GeTe is a prototypical phase change material of high interest for applications in optical and electronic nonvolatile memories. We present an interatomic potential for the bulk phases of GeTe, which is created using a neural network (NN) representation of the potential-energy surface obtained from reference calculations based on density functional theory. It is demonstrated that the NN potential provides a close to ab initio quality description of a number of properties of liquid, crystalline, and amorphous GeTe. The availability of a reliable classical potential allows addressing a number of issues of interest for the technological applications of phase change materials, which are presently beyond the capability of first-principles molecular dynamics simulations.

We provide a consistent description of the electronic levels associated with localized and delocalized excess electrons in liquid water by combining hybrid-functional molecular dynamics simulations with a grand canonical formulation of solutes in aqueous solution. The excess electron localizes in a cavity with an average radius of 1.8 Å and a majority coordination of five water molecules. The vertical binding energy, the optical s-p transitions, and the adiabatic redox level are found to agree closely with their experimental counterparts. The energy level associated with electron delocalization V is inferred to lie at -0.97 eV with respect to the vacuum level.

We investigate redox levels in aqueous solution using a combination of ab initio molecular dynamics (MD) simulations and thermodynamic integration methods. The molecular dynamics are performed with both the semilocal Perdew-Burke-Ernzerhof functional and a nonlocal functional (rVV10) accounting for van der Waals (vdW) interactions. The band edges are determined through three different schemes, namely, from the energy of the highest occupied and of the lowest unoccupied Kohn-Sham states, from total-energy differences, and from a linear extrapolation of the density of states. It is shown that the latter does not depend on the system size while the former two are subject to significant finite-size effects. For the redox levels, we provide a formulation in analogy to the definition of charge transition levels for defects in crystalline materials. We consider the H(+)/H2 level defining the standard hydrogen electrode, the OH(-)/OH(∗) level corresponding to the oxidation of the hydroxyl ion, and the H2O/OH(∗) level for the dehydrogenation of water. In spite of the large structural modifications induced in liquid water, vdW interactions do not lead to any significant structural effect on the calculated band gap and band edges. The effect on the redox levels is also small since the solvation properties of ionic species are little affected by vdW interactions. Since the electronic properties are not significantly affected by the underlying structural properties, it is justified to perform hybrid functional calculations on the configurations of our MD simulations. The redox levels calculated as a function of the fraction α of Fock exchange are found to remain constant, reproducing a general behavior previously observed for charge transition levels of defects. Comparison with experimental values shows very good agreement. At variance, the band edges and the band gap evolve linearly with α. For α ≃ 0.40, we achieve a band gap, band-edge positions, and redox levels in overall good agreement with experiment.

Phase change materials are of great interest as active layers in rewritable optical disks and novel electronic nonvolatile memories. These applications rest on a fast and reversible transformation between the amorphous and crystalline phases upon heating, taking place on the nanosecond time scale. In this work, we investigate the microscopic origin of the fast crystallization process by means of large-scale molecular dynamics simulations of the phase change compound GeTe. To this end, we use an interatomic potential generated from a Neural Network fitting of a large database of ab initio energies. We demonstrate that in the temperature range of the programming protocols of the electronic memories (500-700 K), nucleation of the crystal in the supercooled liquid is not rate-limiting. In this temperature range, the growth of supercritical nuclei is very fast because of a large atomic mobility, which is, in turn, the consequence of the high fragility of the supercooled liquid and the associated breakdown of the Stokes-Einstein relation between viscosity and diffusivity.

We present a general scheme of range-separated hybrid functionals in which the mixing parameters of Fock exchange are fully nonempirical and determined solely from the dielectric function.We showthat the full dielectric dependence leads to an unscreened Fock exchange in the short range, while in the long range the Fock exchange is correctly screened by the macroscopic dielectric constant. The range separation is obtained by fitting to the calculated static dielectric function in the long-wavelength limit. The resulting hybrid functional accurately accounts for electronic and structural properties of various semiconductors and insulators spanning a wide range of band gaps. We present a general scheme of range-separated hybrid functionals in which the mixing parameters of Fock exchange are fully nonempirical and determined solely from the dielectric function. We show that the full dielectric dependence leads to an unscreened Fock exchange in the short range, while in the long range the Fock exchange is correctly screened by the macroscopic dielectric constant. The range separation is obtained by fitting to the calculated static dielectric function in the long-wavelength limit. The resulting hybrid functional accurately accounts for electronic and structural properties of various semiconductors and insulators spanning a wide range of band gaps.