Research Nonlinear Optics Topological Materials 2D Materials and Moiré Superlattices Photovoltaics & Exciton Funnel Battery, Memory, and Neuromorphics Multiscale Materials Modeling

First-Principles Theory and Method Development of Nonlinear Optics, Optoelectronics, and Electronics
Support by NSF DMR-1753054: CAREER: First-Principles Predictive Theory and Microscopic Understanding of Nonlinear Light-Matter Interactions towards Designer Nonlinear Optical Materials

Nonlinear light-matter interaction plays a key role in the understanding, probing, and ultimately controlling light and matter. Novel materials and nanostructures with strong nonlinear optical (NLO) responses are highly desirable for many scientific disciplines and technologically important applications, e.g. ultrafast nonlinear optics, nonlinear biosensing and imaging, efficient generation of entangled photon pairs for quantum computing and quantum sensing, and all-optical transistor and computer. Recent discoveries of giant second and third harmonic generation (SHG, THG) in two distinct classes of materials – 2D crystals and topological materials – raise fundamental questions on our knowledge of NLO materials: What are the intrinsic mechanisms in the extraordinary nonlinear light-matter interactions observed in the 2D materials and topological materials? Are they different from conventional materials? Does a fundamental upper limit exist in second and higher order NLO responses?

Supported by NSF CAREER Award, our group is developing first-principles electronic structure theory of NLO responses and carrying out theoretical studies of 2D materials and topological materials to address the above questions. Specifically, the proposed research will 1) formulate first-principles approaches and develop computational modules for efficient and accurate calculation of general second and third order NLO properties; 2) apply the above approaches combined with group theory to understand intrinsic and extrinsic factors of the extraordinary NLO responses in 2D materials and generalize NLO materials design principles; 3) elucidate the role of symmetry, spin-orbit coupling, and topological phase as well as surface in the giant NLO responses of topological materials; and 4) integrate materials theory and simulation into undergraduate and graduate curriculum, outreach activities to K-12 students, underrepresented groups and secondary school teachers, and undergraduate and graduate research.

Since 2017, we have investigated several types of NLO responses in a number of distinct classes of materials as well as the associated microscopic mechanisms. For example, we have studied SHG, shift photocurrent, and circular photocurrent in the 2D semiconducting/ferroelectric materials. Distinct from conventional linear responses, they exhibit ferroelectricity-driven nonlinear photocurrent switching (for example in group IV monochalcogenides GeSe, GeS, SnS, SnSe, SnTe), that is, nonreciprocal behavior whose current flow direction can be controlled by manipulating ferroelectric polarization. Recently, we proposed a theory of ferroelectric nonlinear anomalous Hall effect (FNAHE) in semimetals and topological materials, and predicted an even-odd layer oscillation of FNAHE in few-layer topological semimetals (bilayer, trilayer, and four layer WTe2). Through a close collaboration with Dr. Aaron Lindenberg's group at Stanford University and Dr. Xiang Zhang's group at UC Berkeley, our theoretically-predicted FNAHE and corresponding intriguing low-energy ferroelectric transition pathway in few-layer WTe2 were experimentally demonstrated.

In our theory of FNAHE, Berry curvature dipole and shift dipole are not only treated on an equal footing to account for intraband and interband contributions to nonlinear anomalous Hall effect, but also established as new order parameters for noncentrosymmetric materials. This suggests that ferroelectric metals and Weyl semimetals may be suitable for the development of nonlinear quantum electronics. Moreover, FNAHE provides a facile approach for direct readout of ferroelectric states, which, combined with vertical ferroelectric writing, may realize nonlinear multiferroic memory such as Berry curvature memory. In addition, the distinct ferroelectric transformation pathway may provide potential routes to achieving non-abelian reciprocal braiding of Weyl nodes. These new findings therefore reveal an underexplored realm beyond classical linear Hall effect and conventional ferroelectrics with exciting new opportunities for FNAHE-based nonlinear quantum electronics using ferroelectric metals and Weyl semimetals.

We believe these microscopic theory and insights of nonlinear optical phenomena from the proposed research (obtained so far with more forthcoming) together with their symmetry principles will offer stupendous opportunities for the discovery and design of nonlinear optical materials and enable novel devices such as nonlinear quantum electronics, spintronics, magnetoelectronics, and dynamic quantum materials which may foster the second quantum evolution with unprecedented impact.

Selected Publications:

Topological Materials and Topological Phase Transition

The seminal discovery of quantum spin Hall (QSH) effect engendered a new chapter of topological materials research in condensed matter physics and materials science, followed by the discoveries of three-dimensional topological insulator, quantum anomalous Hall insulator, topological crystalline materials, Weyl semimetals etc. These exotic materials share a general aspect, that is, the presence of special surface/edge states that are topologically protected against weak perturbations, hence inelastic scattering induced heat dissipation is minimized. In contrast, conventional electronics suffers from severe local heating as any structural defect or chemical impurity could cause additional scattering and reduce carrier transmission. These topological phases, if materialized and integrated at the device level, could be advantageous for many novel low-power and low-dissipation electronic applications. Novel materials with nontrivial electronic and photonic band topology are, therefore, highly desired for utilizing their topological nature and realizing novel devices with low power consumption and heat dissipation and quantum computing free of decoherence. Furthermore, the ability of controlling topological invariants is also highly desirable for developing configurable topological electronics/photonics.

Our group dedicates special effort to the discovery and design of topological materials using first-principles theoretical approaches. We develop first-principles effective Hamiltonian from density funcional theory or many-body perturbation theory calculations and compute the corresponding topological invariants and surface/edge states in the presence/absence of external electric/stress field using Green's function method. Along with materials discoveries with individual topological phases, our group is also interested in understanding topological phase transition, for examples, from nontrivial to trivial topology and from one nontrivial phase to another through chemical doping, elastic strain engineering, electric and magnetic field etc.

Since 2014, we have theoretically predicted a number of new topological materials, including quantum spin Hall materials in 1T'-transition metal chalcogenides (TMDC, 1T'-MX2 with M=W, Mo & X=Te, Se, and S), topological crystalline insulators in monolayer IV–VI semiconductors, quantum spin Hall phase and Weyl semimetallic phase in ternary transition metal chalcogenides. We also discovered topological phase transition, including (a) electric field induced Z2 nontrivial-to-trivial topological phase transition in 1T'-MX2 TMDC, (b) electrically controlled band gap and topological phase transition in two-dimensional multilayer germanane, and (c) vdW interlayer spacing induced topological phase transition from quantum spin Hall insulator to Weyl Semimetal owing to the symmetry-breaking upon stacking and hence the creation and annihilation of Weyl fermions.

Our theoretical predictions of 1T'-WTe2, 1T'-MoTe2, as well as ternary TMDC topological materials have been experimentally demonstrated using angle-resolved photoemission spectroscopy (ARPES), four-probe conductance measurement, scanning tunneling microscope (STM), etc. The quantum spin Hall effect from the conductance measurement was observed up to 100 kelvin in monolayer 1T'-WTe2, and more strikingly it was demonstrated to be the first 2D materials that performs as both topological insulator and superconductor, showing great promise for realizing Majorana modes towards topological quantum computing.

Selected Publications:

2D Ferroelectric, Ferromagnetic, and Multiferroic Materials as well as Moiré Superlattices
Support by NSF DMR-2103842: Collaborative Research: Probing quasiparticle excitations in TMDC Moiré superlattices for revealing and understanding novel two-dimensional correlated phases

Ultrafast sensing and control of the state of matter at nanoscale are highly attractive for advanced applications such as energy harvesting and conversion, high-performance high-density information storage, and quantum computing and quantum simulation. Nanoscale multiferroics is an ideal materials platform which allows for direct manipulation and cross-control of charge, spin, and lattice with potentially superior performance (e.g. ultrafast switching and sensing, low-power consumption) owing to reduced dimensionality. However, due to stringent symmetry and coupling constraints, room-temperature multiferroics with strong ferroic coupling has not been demonstrated. In addition, depolarization-induced instability (i.e. ferroic order vanishing below a few nanometers) poses another challenge to nanoscale multiferroics.

Recent breakthroughs in 2D ferroics and multiferroics (including several works from our group) open up a very exciting yet largely-underexploited realm within ultimate thickness of ~1nm. 2D multiferroics are fundamentally different from their bulk counterpart with (a) significantly reduced dielectric screening, (b) increased joint density of states, (c) distinct symmetry, (d) large Rashba spin splitting, and (e) strong many-body interaction (e.g. excitonic photoabsorption and photoluminescence).

Since 2017, we have showed that it is possible to achieve 2D ferroicity/multiferroicity. For example, monolayer group IV monochalcogenides MX (M=Ge, Sn; X=S, Se) possess room-temperature 2D multiferroicity with strongly-coupled large spontaneous in-plane polarization and lattice strain. Encouragingly, 2D in-plane ferroelectricity was recently observed in their cousin atomic thick tin telluride (SnTe) and few odd-layer tin sulfide (SnS). Moreover, we discovered intrinsic coupling between multiferroicity and nonlinear optical response, predicted a new class of 2D ferromagnetic semiconductors in CrSBr and CrSeBr with Curie temperature around ~150K, and predicted the first class of 2D ferroelectric-ferromagnetic multiferroics in monolayer transition metal phosphorus chalcogenides (TMPCs)-CuMP2X6 (M = Cr, V; X = S, Se), suggesting great opportunity in 2D multiferroics.

Motivated by the recent experiment discovery of ferroelectric 2D metal in bilayer and trilayer WTe2, we theoretically studied ferroelectricity in another interesting material - few-layer WTe2. While individual monolayer 1T'-WTe2 is centrosymmetric without spontaneous polarization, in-plane stacking breaks inversion symmetry and induces out-of-plane polarization while maintaining in-plane metallicity. Subsequently, in-plane interlayer gliding leads to low-energy ferroelectric polarization switching in both even-layer and odd-layer WTe2. Ferroelectric transformation, however, is fundamentally different in even-layer and odd-layer WTe2, i.e. inversion in odd-layer WTe2 and mirror plus glide in even-layer WTe2, which leads to Berry curvature and shift vector switching in the case of odd-layer WTe2 only. Hence, Berry curvature dipole and shift dipole exhibit a striking even-odd layer oscillation upon electric polarization switching, giving rise to ferroelectric nonlinear anomalous Hall effect (FNAHE) fundamentally rooted in nonlinear optics. As a result, Berry curvature dipole and shift dipole are formally established as new order parameters for noncentrosymmetric materials. Our theory and results provide a generalized picture of electric polarization and current based on the zero-th order, linear-order, and higher order dipole. It also provides an important practical guidance for developing new electric and optical tools to characterize materials in greater detail and developing novel devices based on the generalized higher-order polarization/current. Through a close collaboration with Dr. Aaron Lindenberg's group at Stanford University and Dr. Xiang Zhang's group at UC Berkeley, our theoretically-predicted intriguing low-energy ferroelectric transition pathway and ferroelectric nonlinear anomalous Hall effect in few-layer WTe2 were experimentally demonstrated.

Selected Publications:

Thin-Film Photovoltaics and Exciton Funnel

The last two decades have witnessed the increasing demand for renewable and green energy for sustainable society as conventional non-renewable energy resources such as fossil fuels will be depleted by 2060. Photovoltaics, converting solar energy to electric power, provides one of the most efficient approaches to harvest renewable, affordable and environment friendly energy. A variety of solar cells have been developed in the past using different absorbers, including silicon, CdTe, CZTS, CIGS, organic/polymer, and perovskite solar cells. Among them, silicon-, CdTe- and CIGS-based solar cells are dominating the current commercial photovoltaics productions with certified power conversion efficiency (PCE) over 22%. However, the current photovoltaic technologies still face some issues. For example, despite large predominance in photovoltaic market, crystalline silicon (c-Si) solar cells suffer from high cost. The toxicity of Cd and the scarcity of Te are two notable issues of CdTe solar cell. Besides, Indium and gallium in CIGS are not earth abundant. The complexity of defect control hinders the further PCE improvement in CZTS solar cell. Perovskites have attracted tremendous attention in the last decade, and their PCE is approaching 25%, making them very promising for commercialization. Nevertheless, several challenges need to be addressed, such as stability and toxicity of Pb-based perovskites.

Since 2017, we have been collaborating with experimentalists to develop and improve low-dimensional antimony chalcogenides as solar absorbers for thin-film photovoltaic applications. We have investigated the electronic and optical properties of antimony chalcogenides with unique quasi-one-dimensional structure for carrier transport. We have collaborated with Dr. Feng Yan's group at University of Alabama and successfully fabricated Sb2Se3 and Sb2S3 based thin-film solar cells using close space sublimation which is fully compatible with the current thin-film manufacturing process. The power conversion efficiency (PCE) of Sb2Se3 with graphite as electrodes has been improved from 4% to 7% via interfacial engineering. Furthermore, we found the low open-circuit voltage (Voc) limits the further improvement of PCE. We will further study defect physics in antimony chalcogenides using first-principles density functional theory and understand its role in the mid-trap states and the low Voc. We will seek for suitable extrinsic dopants guided by the defect diagram calculations and experimental measurements to provide an effective way to suppress potential defect-induced trap states. In addition to improving the device performance, our group will investigate the role of grain boundary in antimony chalcogenides as well as extrinsic doping.

Besides the development of thin-film photovoltaics, we have collaborated with Dr. Xiaolin Zheng's group at Stanford University and demonstrated the exciton funneling effect in inhomogeneously strained MoS2 which my colleague and I theoretically proposed in 2012. Recently, the concept of inhomogeneously strain engineered "artificial atom" was also explored for enhancing catalytic activities and as single-photon emission center to provide single-photon source for quantum information science applications.

Selected Publications:

Electronic and Ionic Transport at Nanoscale: Batteries, Nonvolatile Memories, Neuromorphic Computing

Over the past few years, our group has been studying electron and ion dynamics in battery materials, nonvolatile memories, electro/photocatalysts, and recently neuromorphic materials.

In lithium-ion batteries, the movement of charge carriers induces the local atomic lattice distortion, forming new quasiparticles called electron-polaron or hole-polaron. Energy barrier of polaron migration largely determines electronic conductivity, which is one of the rate-limiting factors during charge and discharge. Microscopic mechanism of electron, ion, and/or coupled electron-ion dynamics at nanoscale is therefore crucial for further improvement and the discovery of new battery materials. Coupled electronic-ionic motion also exists in nonvolatile random access memories. In nonvolatile memories, the applied voltage drives charge carriers to overcome the potential energy barrier, which strongly couples the electronic and ionic degrees of freedom, leading to either metallic filament formation or local structural distortion associated with negative U-centers. Microscopic mechanisms of resistance switching highly depends on materials systems, and some of them are still not very clear.

Recently, neuromorphic computing emerges as one of the two potential disruptive technologies (together with quantum computing) which may transform the paradigm of computing and information processing. The current data storage, processing, and transmission in cloud computing, edge computing, or Internet of Things (IoT) using inefficient and energy-hungry legacy hardware face the von Neumann bottleneck of the classical computing architecture. Materials with complex nonlinear dynamical behavior such as nonlinear conductance switch and self-oscillation could directly emulate neural elements. Circuits built with those materials could achieve much higher energy efficiency and outperform systems built from thousands of transistors.

The development of neuromorphic computing is in its early stage at present. While initial neural circuits have been realized based on conventional complementary metal-oxide-semiconductor (CMOS) processors, such architecture is still not energy efficient for neuromorphic computing. Hence, there is a surging demand for an integrated investigation from the discovery and design of novel neuromorphic materials, to synthesis and characterization, to the realization of neural elements, and to novel computing and circuit design. We are currently collaborating with several experiment and theory groups to explore and develop new materials systems for energy efficient neuromorphic computing, including strongly-correlated oxides, 2D materials, and molecular materials.

Selected Publications:

Development of Multiscale Materials Modeling
Supported by NSF OAC-1835690: Elements: Software: Autonomous, Robust, and Optimal In-Silico Experimental Design Platform for Accelerating Innovations in Materials Discovery

Advances in high-performance computing have led to the rapidly growing interests in computation-aided materials discovery and device design. However, the complexity in both experiment and theory poses great challenges to the development of more reliable and accurate simulation methods across different lengths and time scales. These challenges are particularly important for many energy applications including understanding fundamental mechanisms of energy, charge, spin, and mass transport. In the past, Qian has developed/co-developed several related computational approaches, including first-principles tight-binding method, time-dependent density functional theory with PAW method and Ultrasoft pseudopotentials, automatic basin filling method for exploring potential energy surface, accelerated many-body perturbation theory within the GW approximation, and nonequilibrium quantum transport with first-principles tight-binding Hamiltonian.

Recently our group has been working on the following new directions to further advance computational materials science:

  • Active Machine Learning Approach for Accelerated and Convergent Model Generation (Supported by NSF CSSI)
  • Bridging Quantum Mechanics to Classical Molecular Dynamics via Machine Learning Force Field
  • Exploring First-Principles Potential Energy Surface under External Perturbations
  • Developing Efficient and Accurate First-Principles Methods for Large Systems

Selected Publications: