Our group is engaged in research in the fields of theoretical condensed matter physics and materials science. Our research covers a broad range of materials systems, from bulk materials (metals, semiconductors, and insulators) to those of finite size, such as organic and inorganic nanostructures, to complex multifunctional oxides; phenomena of interest include optical properties, superconductivity, conductance of nanostructures at finite bias, pressure and temperature effects, and dynamics. Particular emphasis is placed on the study of the role of many-particle effects in determining experimentally observed properties.

Our primary goal is to understand and predict materials properties at the most fundamental level using atomistic first principles (or “ab initio”) quantum-mechanical calculations. A variety of different computational approaches are used that require only the atomic number and positions as input. These first principles methods have, in the past, resulted in excellent quantitative agreement with experiment and have predicted with good accuracy materials properties that were later verified experimentally.

We present below some Highlights on recent research projects. For more details, please see our list of publications.

Topological phases in graphene nanoribbons

We predict that semiconducting graphene nanoribbons (GNRs) of different width, edge, and end termination (synthesizable from molecular precursors with atomic precision) belong to different electronic topological classes. The topological phase of GNRs is protected by spatial symmetries and dictated by the terminating unit cell. Our theoretical prediction of junction states and end states arising from the non-trivial topological phases has recently been experimentally observed and confirmed.

Unifying optical selection rules for excitons in two dimensions: band topology and winding numbers

We show that band topology can dramatically change the photophysics of two-dimensional semiconductors. For systems in which states near the band extrema are of multicomponent character, the spinors describing these components (pseudospins) can pick up nonzero winding numbers around the extremal k point. In these systems, we find that the strength and required light polarization of an excitonic optical transition are dictated by the optical matrix element winding number, a unique and heretofore unrecognized topological characteristic. This winding-number physics leads to novel exciton series and optical selection rules, with each valley hosting multiple bright excitons coupled to light of different circular polarization.

Optical properties of transition metal dichalcogenides and monochalcogenides

Two-dimensional (2D) transition metal dichalcogenides (TMDs) are emerging as a new platform for exploring 2D semiconductor physics. Reduced screening in two dimensions results in markedly enhanced electron–electron interactions, which have been predicted to generate giant bandgap renormalization and excitonic effects. We present first-principles calculations of the optical response of different geometries of molybdenum dichalcogenides, employing the GW-Bethe-Salpeter equation (GW-BSE) approach including self-energy, excitonic, and electron-phonon effects. The resulting optical properties are shown to be dominated by excitonic states with very large binding energy. The optical properties are also strongly related to structures, substrates, and defects.

Exchange-driven intravalley mixing of excitons in monolayer transition metal dichalcogenides

Monolayer transition metal dichalcogenides (TMDCs) are promising two-dimensional (2D) semiconductors for application in optoelectronics. Their optical properties are dominated by two series of photo-excited exciton states—A and B—that are derived from direct interband transitions near the band extrema. Here, by using monolayer MoS2 as a prototypical system and solving the first-principles Bethe–Salpeter equations with the newly developed full spinor formalism, we demonstrate a strong intravalley exchange interaction between A and B excitons, indicating that A and B excitons are mixed states instead of pure Ising excitons. In collaboratithe on with experiment using 2D electronic spectroscopy, we observe that an optical excitation of the lower-energy A exciton induces a population of the higher-energy B exciton, manifesting the intravalley exchange interaction.

Two-dimensional ferromagnetism

The realization of long-range ferromagnetic order in two-dimensional van der Waals crystals, combined with their rich electronic and optical properties, could lead to new magnetic, magnetoelectric and magneto-optic applications. In collaboration with the experiment, we report intrinsic long-range ferromagnetic order in pristine Cr2Ge2Te6 atomic layers, as revealed by the experimental scanning magneto-optic Kerr microscopy. In this magnetically soft, two-dimensional van der Waals ferromagnet, we achieve unprecedented control of the transition temperature using a very small field. We explain the observed phenomenon using renormalized spin-wave theory. In another two-dimensional van der Waals material GaSe, we theoretically predict that by a certain amount of hole doping, this material will exhibit robust itinerant ferromagnetism.

Exciton dispersion in monolayer transition metal dichalcogenides and singlet fission in solid pentacene

We use the ab initio GW-Bethe-Salpeter equation method to calculate the dispersion of excitons in monolayer MoS2 and find a nonanalytic lightlike dispersion. This behavior arises from an unusual |Q|-term in both the intra- and intervalley exchange of the electron-hole interaction, which concurrently gives rise to a valley quantum phase of winding number two. A simple effective Hamiltonian to Q2 order with analytic solutions is derived to describe quantitatively these behaviors.  We further developed a new first-principles approach to predict and understand rates of singlet fission with an ab initio Green’s-function formalism based on many-body perturbation theory. We predict a singlet lifetime of 30–70 fs in solid pentacene, in very good agreement with experimental data,

Renormalization of quasiparticle and exciton properties induced by substrate screening

We developed ab initio method to include the RPA dielectric effect from substrates to two-dimensional materials to study the renormalization of GW quasiparticle band structures and exciton excitations. We have studied various materials including transition metal dichalcogenides and few-layer black phosphorus, considering the substrate effects from graphene and wide-gap insulators. Our calculations show nice agreement with experiments.

Electronic properties of bottom-up synthesized graphene nanoribbons

Bottom-up synthesis technique has become an important experimental technique that provides large degrees of freedom to synthesize various types of graphene nanoribbons. One important application is bandgap engineering that is used to create semiconductor heterostructure devices that perform processes such as resonant tunneling and solar energy conversion. In collaboration with experiments, we have studied electronic properties of a wide range of graphene nanoribbons that exhibit many intriguing physics.

Design novel electronic states in few-layer black phosphorus and graphene

Few-layer black phosphorus has recently emerged as a promising 2D semiconductor, notable for its widely tunable bandgap, highly anisotropic properties, and theoretically predicted large exciton binding energies.  We have designed the different forms of black phosphorus to give rise to novel electronic properties such as anisotropic Dirac fermions and gate-switchable transport and optical anisotropy. We have designed several patterned graphene with even disordered potential that gives rise to supercollimation properties.

Probing excitonic dark states in single-layer tungsten disulphide

In this work by applying GW-BSE theory, in combination with experimental two-photon excitation spectroscopy, we study a series of excitonic dark states in single-layer WS2. We prove that the excitons are of Wannier type, meaning that each exciton wavefunction extends over multiple unit cells, but with extraordinarily large binding energy (~0.7 electronvolts). These strongly bound exciton states are observed to be stable even at room temperature. We theoretically reveal an exciton series that deviates substantially from hydrogen models, with a novel energy dependence on the orbital angular momentum.

Ab Initio electron-phonon coupling in understanding hot carriers and temperature-induced topological phase transition

Hot carrier thermalization is a major source of efficiency loss in solar cells. We develop and apply an ab initio approach based on density functional theory and many-body perturbation theory to investigate hot carriers in semiconductors. Our calculations include electron-electron and electron-phonon interactions, and require no experimental input other than the structure of the material. We demonstrate that a hot carrier distribution characteristic of Si under solar illumination thermalizes within 350 fs, in excellent agreement with pump-probe experiments. Our work sheds light on the subpicosecond time scale after sunlight absorption in Si, and constitutes a first step towards ab initio quantification of hot carrier dynamics in materials.


New release of BerkeleyGW 2.0 and recent development in methods and algorithms

BerkeleyGW is a massively parallel computational package for electron excited state properties that are based on many-body perturbation theory employing the ab initio GW and GW plus Bethe-Salpeter equation methodology. We recently released BerkeleyGW 2.0, please visit BerkleyGW website.