Accurate and efficient calculations of absorption spectra of molecules and materials are essential for the understanding and rational design of broad classes of systems. Solving the Bethe-Salpeter equation (BSE) for electron-hole pairs usually yields
accurate predictions of absorption spectra, but it is computationally expensive, especially if thermal averages of spectra computed for multiple configurations are required. We present a method based on machine learning to evaluate a key quantity entering the definition of absorption spectra: the dielectric screening. We show that our approach yields a model for the screening that is transferable between multiple configurations sampled during first principles molecular dynamics simulations; hence it leads to a substantial improvement in the efficiency of calculations of finite temperature spectra. We obtained computational gains of one to two orders of magnitude for systems with 50 to 500 atoms, including liquids, solids, nanostructures, and solid/liquid interfaces. Importantly, the models of dielectric screening derived here may be used not only in the solution of the BSE but also in developing functionals for time-dependent density functional theory (TDDFT) calculations of homogeneous and heterogeneous systems. Overall, our work provides a strategy to combine machine learning with electronic structure calculations to accelerate first principles simulations of excited-state properties.
Quantum computers hold promise to enable efficient simulations of the properties of molecules and materials; however, at present they only permit ab initio calculations of a few atoms, due to a limited number of qubits. In order to harness the power
of near-term quantum computers for simulations of larger systems, it is desirable to develop hybrid quantum-classical methods where the quantum computation is restricted to a small portion of the system. This is of particular relevance for molecules and solids where an active region requires a higher level of theoretical accuracy than its environment. Here we present a quantum embedding theory for the calculation of strongly-correlated electronic states of active regions, with the rest of the system described within density functional theory. We demonstrate the accuracy and effectiveness of the approach by investigating several defect quantum bits in semiconductors that are of great interest for quantum information technologies. We perform calculations on quantum computers and show that they yield results in agreement with those obtained with exact diagonalization on classical architectures, paving the way to simulations of realistic materials on near-term quantum computers.
We derive a dielectric-dependent hybrid functional which accurately describes the electronic properties of heterogeneous interfaces and surfaces, as well as those of three- and two-dimensional bulk solids. The functional, which does not contain any a
djustable parameter, is a generalization of self-consistent hybrid functionals introduced for homogeneous solids, where the screened Coulomb interaction is defined using a spatially varying, local dielectric function. The latter is determined self-consistently using density functional calculations in finite electric fields. We present results for the band gaps and dielectric constants of 3D and 2D bulk materials, and band offsets for interfaces, showing an accuracy comparable to that of GW calculations.
Carbonates are the main species that bring carbon deep into our planet through subduction. They are an important rock-forming mineral group, fundamentally distinct from silicates in Earths crust in that carbon binds to three oxygen atoms, while silic
on is bonded to four oxygens. Here, we present experimental evidence that under the sufficiently high pressures and high temperatures existing in the lower mantle, ferromagnesian carbonates transform to a phase with tetrahedrally coordinated carbons. Above 80 GPa, in situ synchrotron infrared experiments show the unequivocal spectroscopic signature of the high-pressure phase of (Mg,Fe)CO$_3$. Using ab-initio calculations, we assign the new IR signature to C-O bands associated with tetrahedrally coordinated carbon with asymmetric C-O bonds. Tetrahedrally coordinated carbonates are expected to exhibit substantially different reactivity than low pressure three-fold coordinated carbonates, as well as different chemical properties in the liquid state. Hence this may have significant implications on carbon reservoirs and fluxes and the global geodynamic carbon cycle.
We computed the inter-layer bonding properties of graphite using an ab-initio many body theory. We carried out variational and diffusion quantum Monte Carlo calculations and found an equilibrium inter-layer binding energy in good agreement with most
recent experiments. We also analyzed the behavior of the total energy as a function of interlayer separation at large distances comparing the results with the predictions of the random phase approximation.
Using first principles simulations we have investigated the structural and bonding properties of dense fluid oxygen up to 180 GPa. We have found that band gap closure occurs in the molecular liquid, with a slow transition from a semi-conducting to a
poor metallic state occurring over a wide pressure range. At approximately 80 GPa, molecular dissociation is observed in the metallic fluid. Spin fluctuations play a key role in determining the electronic structure of the low pressure fluid, while they are suppressed at high pressure.
In order to control and tailor the properties of nanodots, it is essential to separate the effects of quantum confinement from those due to the surface, and to gain insight into the influence of preparation conditions on the dot physical properties.
We address these issues for the case of small Ge clusters (1-3 nm), using a combination of empirical and first-principles molecular dynamics techniques. Our results show that over a wide temperature range the diamond structure is more stable than tetragonal, ST12-like structures for clusters containing more than 50 atoms; however, the magnitude of the energy difference between the two geometries is strongly dependent on the surface properties. Based on our structural data, we propose a mechanism which may be responsible for the formation of metastable ST12 clusters in vapor deposition experiments, by cold quenching of amorphous nanoparticles with unsaturated, reconstructed surfaces.
We present a real-space adaptive-coordinate method, which combines the advantages of the finite-difference approach with the accuracy and flexibility of the adaptive coordinate method. The discretized Kohn-Sham equations are written in generalized cu
rvilinear coordinates and solved self-consistently by means of an iterative approach. The Poisson equation is solved in real space using the Multigrid algorithm. We implemented the method on a massively parallel computer, and applied it to the calculation of the equilibrium geometry and harmonic vibrational frequencies of the CO_2, CO, N_2 and F_2 molecules, yielding excellent agreement with the results of accurate quantum chemistry and Local Density Functional calculations.
