Search for: All records

Abstract Non-adiabatic molecular dynamics (NAMD) simulations have become an indispensable tool for investigating excited-state dynamics in solids. In this work, we propose a general framework, N²AMD (Neural-Network Non-Adiabatic Molecular Dynamics), which employs an E(3)-equivariant deep neural Hamiltonian to boost the accuracy and efficiency of NAMD simulations. Distinct from conventional machine learning methods that predict key quantities in NAMD, N²AMD computes these quantities directly with a deep neural Hamiltonian, ensuring excellent accuracy, efficiency, and consistency. N²AMD not only achieves impressive efficiency in performing NAMD simulations at the hybrid functional level within the framework of the classical path approximation (CPA), but also demonstrates great potential in predicting non-adiabatic coupling vectors and suggests a method to go beyond CPA. Furthermore, N²AMD demonstrates excellent generalizability and enables seamless integration with advanced NAMD techniques and infrastructures. Taking several extensively investigated semiconductors as the prototypical system, we successfully simulate carrier recombination in both pristine and defective systems at large scales where conventional NAMD often significantly underestimates or even qualitatively incorrectly predicts lifetimes. This framework offers a reliable and efficient approach for conducting accurate NAMD simulations across various condensed materials. 

Abstract A unified picture of the electronic relaxation dynamics of ionized liquid water has remained elusive despite decades of study. Here, we employ sub-two-cycle visible to short-wave infrared pump-probe spectroscopy and ab initio nonadiabatic molecular dynamics simulations to reveal that the excess electron injected into the conduction band (CB) of ionized liquid water undergoes sequential relaxation to the hydrated electronsground state via an intermediate state, identified as the elusivepexcited state. The measured CB andp-electron lifetimes are 0.26 ± 0.02 ps and 62 ± 10 fs, respectively. Ab initio quantum dynamics yield similar lifetimes and furthermore reveal vibrational modes that participate in the different stages of electronic relaxation, with initial relaxation within the dense CB manifold coupled to hindered translational motions whereas subsequentp-to-srelaxation facilitated by librational and even intramolecular bending modes of water. Finally, energetic considerations suggest that a hitherto unobserved trap state resides ~0.3-eV below the CB edge of liquid water. Our results provide a detailed atomistic picture of the electronic relaxation dynamics of ionized liquid water with unprecedented time resolution. 

The mixed tin (Sn) and lead (Pb) perovskite compositions have shown great potential in perovskite photovoltaic devices due to the significantly enhanced material stability and prolonged carrier lifetime, compared to the pure Sn halide perovskites. In spite of the increasing interest, the behaviors of photo-generated charges and of the intrinsic point defects, such as the metal cation vacancies (V Sn and V Pb ) and the interstitial halogen (i I ), have not been well understood in this class of materials. We report first-principles density functional theory (DFT) calculations combined with ab initio non-adiabatic molecular dynamics (NAMD) simulations on the static and dynamic structures of MA 2 SnPbI 6 with and without these intrinsic defects. We discuss the nature of the defect states and unveil the influence of the intrinsic point defects on the structure, optoelectronic properties, and charge carrier dynamics of MA 2 SnPbI 6 . The i I defect significantly shortens the carrier lifetime by creating mid-gap states that provide new recombination pathways. In comparison, the vacancy defects have much weaker influence on the carrier lifetime. Both V Sn and V Pb produce the defect states just below the valence band maxima (VBMs), and do not alter the band gap. They affect the carrier lifetime through changing the energy dispersions of VBMs and the conduction band minima (CBMs). We suggest that excess cations should be used in the synthesis of perovskites to avoid the appearance of interstitial halogen defects. 

Critical role of hydrogen bond network is unraveled by integrating nuclear quantum effects into nonadiabatic molecular dynamics.