Simulating warm dense matter that undergoes a wide range of temperatures and densities is challenging. Predictive theoretical models, such as quantum-mechanics-based first-principles molecular dynamics (FPMD), require a huge amount of computational resources. Herein, we propose a deep learning based scheme called electron temperature dependent deep potential molecular dynamics (TDDPMD), which can be readily applied to study larger systems with longer trajectories, yielding more accurate properties. We take warm dense beryllium (Be) as an example with the training data from FPMD simulations spanning a wide range of temperatures (0.4–2500 eV) and densities (3.50–8.25 g/cm3). The TDDPMD method well reproduces the principal Hugoniot curve and radial distribution functions from the FPMD method. Furthermore, it depicts the reflection point of the Hugoniot curve more smoothly and provides more converged diffusion coefficients. We also show the new model can yield static structure factors and dynamic structure factors of warm dense Be.
19 February 2021 in Matter and Radiation at Extremes
We propose an efficient scheme that combines density functional theory (DFT) with deep potentials (DPs), to systematically study convergence issues in the computation of the electronic thermal conductivity of warm dense aluminum (2.7 g/cm3 and temperatures ranging from 0.5 eV to 5.0 eV) with respect to the number of k-points, the number of atoms, the broadening parameter, the exchange-correlation functionals, and the pseudopotentials. Furthermore, we obtain the ionic thermal conductivity using the Green–Kubo method in conjunction with DP molecular dynamics simulations, and we study size effects on the ionic thermal conductivity. This work demonstrates that the proposed method is efficient in evaluating both electronic and ionic thermal conductivities of materials.
20 November 2025 in The Journal of Chemical Physics
ABACUS (Atomic-orbital Based Ab initio Computation at USTC) is an open-source software for first-principles electronic structure calculations and molecular dynamics simulations. It mainly features density functional theory (DFT) and molecular dynamics functions and is compatible with both plane wave basis sets and numerical atomic orbital basis sets. ABACUS serves as a platform that facilitates the integration of various electronic structure methods, such as Kohn–Sham DFT, stochastic DFT, orbital-free DFT, real-time time-dependent DFT, etc. In addition, with the aid of high-performance computing, ABACUS is designed to perform efficiently and provide massive amounts of first-principles data for generating general-purpose machine learning potentials, such as deep potential with attention models. Furthermore, ABACUS serves as an electronic structure platform that interfaces with several artificial intelligence-assisted algorithms and packages, such as DeePKS-kit, DeePMD, DP-GEN, DeepH, DeePTB, HamGNN, etc.
1 January 2024 in Matter and Radiation at Extremes
In traditional finite-temperature Kohn–Sham density functional theory (KSDFT), the partial occupation of a large number of high-energy KS eigenstates restricts the use of first-principles molecular dynamics methods at extremely high temperatures. However, stochastic density functional theory (SDFT) can overcome this limitation. Recently, SDFT and the related mixed stochastic–deterministic density functional theory, based on a plane-wave basis set, have been implemented in the first-principles electronic structure software ABACUS [Q. Liu and M. Chen, Phys. Rev. B 106, 125132 (2022)]. In this study, we combine SDFT with the Born–Oppenheimer molecular dynamics method to investigate systems with temperatures ranging from a few tens of eV to 1000 eV. Importantly, we train machine-learning-based interatomic models using the SDFT data and employ these deep potential models to simulate large-scale systems with long trajectories. Subsequently, we compute and analyze the structural properties, dynamic properties, and transport coefficients of warm dense matter.
Traditional finite-temperature Kohn-Sham density functional theory (KSDFT) based on the diagonalization (DG) method has an unfavorable scaling with respect to the electron number or at high temperatures. The evaluation of the ground-state density in KSDFT can be replaced by the Chebyshev trace (CT) method. In addition, the use of stochastic orbitals within the CT method leads to the stochastic density functional theory [ Phys. Rev. Lett. 111, 106402 (2013)] (SDFT) and its improved theory, mixed stochastic-deterministic density functional theory [ Phys. Rev. Lett. 125, 055002 (2020)] (MDFT). We have implemented the above four methods based on the plane-wave basis set within the first-principles package ABACUS. In addition, the four methods are adapted to the norm-conserving pseudopotentials and periodic boundary conditions with the use of -point sampling in the Brillouin zone. By testing the Si and C systems from the DG method as benchmarks, we systematically evaluate the accuracy and efficiency of the CT, SDFT, and MDFT methods by examining a series of physical properties, which include the electron density, free energy, atomic forces, stress, and density of states. We conclude that the CT, SDFT, and MDFT methods not only reproduce the DG results with a sufficient accuracy but also exhibit several advantages over the DG method. We expect these methods can be of great help in studying high-temperature and large-size extended systems such as warm dense matter and dense plasmas.
By adopting stochastic density functional theory (SDFT) and mixed stochastic–deterministic density functional theory (MDFT) methods, we perform first-principles calculations to predict the shock Hugoniot curves of boron (pressure P = 7.9 × 103–1.6 × 106 GPa and temperature T = 25–2800 eV), silicon (P = 2.6 × 103–7.9 × 105 GPa and T = 21.5–1393 eV), and aluminum (P = 5.2 × 103–9.0 × 105 GPa and T = 25–1393 eV) over wide ranges of pressure and temperature. In particular, we systematically investigate the impact of different cutoff radii in norm-conserving pseudopotentials on the calculated properties at elevated temperatures, such as pressure, ionization energy, and equation of state. By comparing the SDFT and MDFT results with those of other first-principles methods, such as extended first-principles molecular dynamics and path integral Monte Carlo methods, we find that the SDFT and MDFT methods show satisfactory precision, which advances our understanding of first-principles methods when applied to studies of matter at extremely high pressures and temperatures.