top of page

RESEARCH INTERESTS

 

Nano-Electronics:

 

Next Generation Nanoscale Transistors and Memories Technologies

Low Dimensional Semiconductor Materials (e.g. Elemental & 2D-Transition Metal Dichalcogenides)

Advanced CMOS Process Technologies

 

Nano-Photonics:

 

Active & Passive Photonics Devices (Photodetector, Modulator, Laser, Waveguide)

New Optoelectronics Materials (e.g. 2D Semiconductor and Heterostructures, Group-IV Materials)

Bio-Photonics Sensors

Selected Publications:

​​

  1. "A MoS2 Hafnium Oxide based Ferroelectric Encoder for Temporal-Efficient Spiking Neural Network," Advanced Materials 34, 2204949 (2023).

  2. "In-memory computing using memristor arrays with ultrathin 2D PdSeOx/PdSe2 heterostructure," Advanced Materials 34, 2201488 (2022).

  3. "Interface modulated resistive switching in Mo-irradiated ReS2 for neuromorphic computing," Advanced Materials 34, 2202722 (2022).

  4. "Wafer-scale two-dimensional hafnium diselenide based memristor crossbar array for energy-efficient neural network hardware," Advanced Materials 34, 2103376 (2022).

  5. "Anomalous resistive switching in memristors based on two dimensional palladium diselenide using heterophase grain boundaries," Nature Electronics 4, 348-356 (2021).

  6. "Self-selective multi-terminal memtransistor crossbar array for in-memory computing," ACS Nano 15, 1764 (2021).

  7. "Artificial synapses based on multi-terminal memtransistors for neuromorphic application," Advanced Functional Materials 29, 1901106, 2019.

  8. "Waveguide-integrated black phosphorus photodetector for mid-infrared applications," ACS Nano 13, 913-921, 2019.

  9. "High mobility anisotropic black phosphorus nanoribbons field-effect transistor," Advanced Functional Materials 28, 1801524, 2018.

  10. "A black phosphorus carbide infrared phototransistor," Advanced Materials 30, 1705039, 2018.

  11. "Few-layer black phosphorus carbide field-effect transistor via carbon doping," Advanced Materials 29, 1700503, 2017.

Nano Electronics Project
2D Transition Metal Dichalcogenides
​

This project aims to exploit the exceptional material properties of 2D layered semiconductors to realize novel devices for ubiquitous electronics and optoelectronics applications.  By taking advantage of its superior carrier transport, electrostatic control and optical transparency properties, immense opportunities are available to make great scientific and technological impact on the application of 2D-TMD in next generation field-effect transistors, bio-sensors and optical detectors.  A holistic integration approach will be developed in which all module challenges need to be addressed in totality rather than standalone to allow the realization of devices with optimized performance.  New process technologies that are CMOS-compatible and that facilitate process simplicity will be developed to increase its likelihood of future industry adoption for large-scale integration and low cost manufacturing.

 

Research Highlights:

Adv Mat 2021_HfSe2 memristor.png
Advanced Materials 33, 2103376, 2021.

DOI: 10.1002/adma.202103376 [Link] [PDF]

​​

Memristor crossbar with programmable conductance could overcome the energy consumption and speed limitations of neural networks when executing core computing tasks in image processing. Here, a memristor crossbar array is demonstrated using wafer-scale polycrystalline HfSe2 grown by molecular beam epitaxy, and a metal-assisted van der Waals transfer technique.  The memristor exhibits small switching voltage (0.6 V), low switching energy (0.82 pJ), and simultaneously achieves emulation of synaptic weight plasticity.  Furthermore, the CBA enables artificial neural network with a high recognition accuracy of 93.34%.  Hardware multiply-and-accumulate (MAC) operation with a narrow error distribution of 0.29% is also demonstrated, and a high power efficiency of greater than 8-trillion operations per second per Watt is achieved.  Based on the MAC results, hardware convolution image processing can be performed using programmable kernels, which constitutes a vital function for neural network hardware.

Nat Electron 2021.png
Nature Electronics 4, 348-356, 2021.

DOI: 10.1038/s41928-021-00573-1 [Link] [PDF]

​​

The implementation of memristive synapses in neuromorphic computing is hindered by the limited reproducibility and high energy consumption of the switching behaviour. Here, we report memristors based on two-dimensional pentagonal palladium diselenide (PdSe2) that can exhibit anomalous resistive switching behaviour with two interchangeable reset modes: total reset and quasi-reset. Heterophase grain boundaries are formed in the PdSe2 via local phase transitions induced by electron-beam irradiation, which leads to residual filaments along the grain boundaries that can guide the formation of conductive filaments. When operated in the quasi-reset mode, the memristors show a sixfold improvement in switching variation compared with devices operating in the total-reset mode, as well as a low set voltage (0.6 V), long retention times and programmable multilevel resistance states. We also show that the devices can emulate synaptic plasticity and that multipattern memorization can be implemented using a crossbar array architecture.

ACS nano_memtransistor.png
ACS Nano 15(1), 1764–1774, 2021.

DOI: 10.1021/acsnano.0c09441 [Link] [PDF]

​​

Two-terminal resistive switching devices are commonly plagued with longstanding scientific issues including interdevice variability and sneak current that lead to computational errors and high-power consumption. This necessitates the integration of a separate selector in a one-transistor-one-RRAM (1T-1R) configuration to mitigate crosstalk issue, which compromises circuit footprint. Here, we demonstrate a multiterminal memtransistor crossbar array with increased parallelism in programming via independent gate control, which allows in situ computation at a dense cell size of 3−4.5 F2 and a minimal sneak current of 0.1 nA. Moreover, a low switching energy of 20 fJ/bit is achieved at a voltage of merely 0.42 V. The architecture is capable of performing multiply-and-accumulate operation, a core computing task for pattern classification. A high MNIST recognition accuracy of 96.87% is simulated owing to the linear synaptic plasticity. Such computing paradigm is deemed revolutionary toward enabling data-centric applications in artificial intelligence and Internet-of-things.

Adv Func Mat 2021_Universal Model.png
Advanced Functional Materials 31, 2105003, 2021.

DOI: 10.1002/adfm.202105003 [Link] [PDF]

​​

2D semiconductor field-effect transistors (2D FETs) have been limited by the metal-2D semiconductor contact, and the non-negligible contact resistance (RSD) not only deteriorates the on-state current but also hinders the direct characterization of the intrinsic properties of 2D semiconductors (e.g., intrinsic charge carrier mobility, μint). Therefore, a proper extraction technique that can independently characterize the metal-2D semiconductor contact behavior and the intrinsic properties of a 2D semiconducting layer is highly desired. In this study, a universal yet simple method is developed to accurately extract the critical parameters in 2D FETs, including characteristic temperature (To), threshold voltage, RSD, and μint. The practicability of this method is extensively explored by characterizing the temperature-dependent carrier transport behavior and the strain-induced band structure modification in 2D semiconductors. TCAD simulation is employed to verify the precision of RSD extraction. Furthermore, the universality of the proposed method is validated by successfully implementing the extraction to various 2D semiconductors, including BP, InSe, MoS2, ReS2, and WS2 with top- and bottom-gated configurations.

npj 2D_ReS2.png
npj 2D Materials and Applications 5, 1, 2021.

DOI: 10.1038/s41699-020-00190-0 [Link] [PDF]

​​

Owing to the stochastic formation of filament, the set/reset voltage of vertical MIM memristors is difficult to control, which results in poor temporal and spatial switching uniformity. Here, a two-terminal lateral memristor based on electron-beam-irradiated rhenium disulfide (ReS2) is realized, which unveils a resistive switching mechanism based on Schottky barrier height (SBH) modulation. The devices exhibit a forming-free, stable gradual RS characteristic, and simultaneously achieve a small transition voltage variation during positive and negative sweeps (6.3%/5.3%). The RS is attributed to the motion of sulfur vacancies induced by voltage bias in the device, which modulates the ReS2/metal SBH. The gradual SBH modulation stabilizes the temporal variation in contrast to the abrupt RS in MIM-based memristors. Moreover, the emulation of long-term synaptic plasticity of biological synapses is demonstrated using the device, manifesting its potential as artificial synapse for energy-efficient neuromorphic computing applications.

Appl Phy Rev 2020_BP photonics.png
Applied Physics Reviews 7, 031302, 2021.

DOI: 10.1063/5.0005641 [Link] [PDF]

​​

The direct bandgap of black phosphorus is tunable by layer number, vertical electric field, and chemical doping, covering a broad spectrum for efficient light manipulation. The optical anisotropy further enables the identification and control of light polarization. Along with high carrier mobility, nonlinear optical properties, and integration capability due to its layered lattice structure, black phosphorus manifests itself as a promising multipurpose material for chip-scale optoelectronics. In this manuscript, we review the research on black phosphorus photonics, with a focus on the most fundamental active functions in photonic circuits: photodetection, electro-optic modulation, light emission, and laser pulse generation, aiming at evaluating the feasibility of integrating these black phosphorus-based components as a compact system for on-chip applications.

AEM_Printed Synapse.png
Advanced Electronic Materials 5, 1900740, 2019.

DOI: 10.1002/aelm.201900740 [Link] [PDF]

​​

Realization of memristors capable of storing and processing data on flexible substrates is a key enabling technology towards “system-on-plastics”. Here, we realize an aerosol-jet-printed Ag/MoS2/Ag memristor in a cross-bar structure by developing a scalable and low temperature printing technique utilizing functional molybdenum disulfide (MoS2) ink platform. The fully printed devices exhibit an ultra-low switching voltage (0.18 V), a high switching ratio (1E7), a wide tuneable resistance states (10-1E10 Ω) for multi-bit data storage, as well as a low standby power consumption of 1 fW and a switching energy of 4.5 fJ per transition set. Moreover, the MoS2 memristor exhibits both volatile and non-volatile resistive switching behaviours by controlling the current compliance levels, which efficiently mimic the short-term and long-term plasticity of biological synapses, demonstrating its potential to enable energy-efficient artificial neuromorphic computing.

MoS2 circuits_AEM.png
Advanced Electronic Materials 5, 1900393, 2019.

DOI: 10.1002/aelm.201900393 [Link] [PDF]

​​

To date most reported works have been limited to exfoliated MoS2 nanosheets primarily due to the difficulty in synthesizing large-area and high-quality MoS2 thin film. A demonstration of wafer-scale monolayer MoS2 synthesis is reported by chemical vapor deposition (CVD), enabling transistors, memristive memories, and integrated circuits to be realized simultaneously. Specifically, building on top-gated FETs with a high-κ gate dielectric (HfO2), Boolean logic circuits including inverters and NAND gates are successfully demonstrated using direct-coupled FET logic technology, with typical inverters exhibiting a high voltage gain of 16, a large total noise margin of 0.72 VDD at VDD = 3 V, and perfect logic-level matching. Additionally, resistive switching is demonstrated in a MoS2-based memristor, indicating that they have great potential for the development of resistive random-access memory. By virtue of scalable CVD growth capability, the way toward practical and large-scale electronic applications of MoS2 is indicated. 

Adv Func Mat 2019_Synapse.png
Advanced Functional Materials 29, 1901106, 2019.

DOI: 10.1002/adfm.201901106 [Link] [PDF]

​​

Here, we realize a memtransistor-based artificial synapse by integrating memristor and selector transistor into a multi-terminal device using monolayer polycrystalline-MoS2 grown by a scalable chemical vapor deposition (CVD) process.  Notably, the memtransistor offers both drain- and gate-tunable non-volatile memory functions, which efficiently emulates the long-term potentiation/depression, spike-amplitude and spike-timing-dependent plasticity of biological synapses.  Moreover, the gate tunability function that is not achievable in two-terminal memristors, enables significant bipolar resistive states switching up to four orders-of-magnitude and high cycling endurance. First-principles calculations reveal a new resistive switching mechanism driven by the diffusion of double sulfur vacancy perpendicular to the MoS2 grain boundary, leading to a conducting switching path without the need for a filament forming process.  The seamless integration of multi-terminal memtransistors may offer another degree-of-freedom to tune the synaptic plasticity by a third gate terminal for enabling complex neuromorphic learning. 

ACS Nano_WG BP.png
ACS Nano 13, 913-921, 2019.

DOI: 10.1021/acsnano.8b08758 [Link] [PDF]

​​

Here, we realize an integration of silicon-on-insulator (SOI) waveguides with black phosphorus (BP) photodetectors.  When operating near BP’s cut-off wavelength where absorption is weak, the light-BP interaction is enhanced by exploiting the optical confinement in Si waveguide and grating structure to overcome the limitation of absorption length constrained by the BP thickness. Devices with different BP crystal orientation and thickness are compared in terms of their responsivity and noise equivalent power (NEP). Spectral photoresponse from 3.68 µm to 4.03 µm was investigated. Additionally, power-dependent responsivity and gate-tunable photocurrent were also studied.  At a bias of 1 V, the BP photodetector achieved a responsivity of 23 A/W at 3.68 µm and 2 A/W at 4 µm, and a NEP less than 1 nW/Hz1/2 at room-temperature.  The integration of passive Si photonics and active BP photodetector is envisaged to offer a potential pathway towards MIR sensing applications. 

Advanced Functional Materials 28, 1801524, 2018.

DOI: 10.1002/adfm.201801524 [Link] [PDF]

​​

Achieving excellent electrostatic control and immunity to short channel effects have been a formidable challenge in ultra-scaled devices. Here, by leveraging on the merits of 3D structure, we have demonstrated high-mobility black phosphorus nanoribbons field-effect transistors (BPNR-FET) and systematically investigate its anisotropic transport properties.  A simple top-down reactive ion etching (RIE) method is used to realize both armchair- and zigzag-oriented nanoribbons with various widths down to 60 nm.  The mobility of BPNR-FET is found to be width- and thickness-dependent, with the highest hole mobility of ~862 cm2/Vs demonstrated in armchair-oriented device at room temperature by combining high-k gate dielectric and hydrogen treatment to reduce sidewall scattering.  Furthermore, hydrogenation effectively passivates the nanoribbon dangling bonds, leading to hysteresis and contact resistance improvement.  

Advanced Materials 30, 1705039, 2018.

DOI: 10.1002/adma.201705039 [Link] [PDF]

​​

Here, a novel 2DM, black phosphorous carbide (b-PC) with a wide absorption spectrum up to 8000 nm is synthesized and a b-PC phototransistor with a tunable responsivity and response time at an excitation wavelength of 2004 nm is demonstrated. The b-PC phototransistor achieves a peak responsivity of 2163 A W−1 and a shot noise equivalent power of 1.3 fW Hz^−1/2 at 2004 nm. In addition, it is shown that a response time of 0.7 ns is tunable by the gating effect, which renders it versatile for high-speed applications. Under the same signal strength (i.e., excitation power), its performance in responsivity and detectivity in room temperature condition is currently ahead of recent top-performing photodetectors based on 2DMs that operate with a small bias voltage (Vd) of 0.2 V.  

ACS Nano 11, 7416-7423, 2017.

DOI: 10.1021/acsnano.7b03703 [Link] [PDF]

​​

Here, we demonstrate a monolithically integrated complementary inverter made using a homogeneous black phosphorus (BP) nanosheet on flexible substrates. Controllable electron concentration is achieved by accurately modulating the aluminum (Al) donor doping, which realizes BP n-FET with a room-temperature on/off ratio >10^3. Simultaneously, work function engineering is employed to obtain a low Schottky barrier contact electrode that facilities hole injection, thus enhancing the current density of the BP p-FET by 9.4 times. The flexible inverter circuit shows a clear digital logic voltage inversion operation along with a larger-than-unity direct current voltage gain, while exhibits alternating current dynamic signal switching at a record high frequency up to 100 kHz and remarkable electrical stability upon mechanical bending with a radii as small as 4 mm. 

Advanced Materials 29, 1700503, 2017.

DOI: 10.1002/adma.201700503 [Link] [PDF]

​​

Black phosphorus carbide (b-PC) is a new family of layered semiconducting material that has recently been predicted to have the lightest electrons and holes among all known two dimensional semiconductors, yielding a p-type mobility (~10^5 cm2V−1s−1) at room temperature that is approximately five times larger than the maximum value in black phosphorus. Here, we report on a high-performance composite few-layer b-PC field-effect transistor fabricated via a novel carbon doping technique which achieved a high hole mobility of 1,995 cm2V-1s-1 at room temperature.  The absorption spectrum of this material covers an electromagnetic spectrum in the infrared regime not served by black phosphorus and is useful for range finding applications as the earth atmosphere has good transparency in this spectral range.  

Advanced Functional Materials 27, 1604638, 2017.

DOI: 10.1002/adfm.201604638 [Link] [PDF]

​​

2D p-n junction is a fundamental building block for nanoelectronics device applications, which has been predominantly realized using van der Waals heterostructures or electrostatic-gated junctions due to the lack of controllable doping technique.  Here, near-ideal black phosphorus p-n homojunction diodes are made possible by a novel and facile Al-atoms doping.  The diode achieves a near-unity ideality factor of 1.001 along with an on/off ratio of ≈5.6 × 10^3 at a low bias of 2 V, allowing for low-power dynamic current rectification without signal decay or overshoot. When operated under a photovoltaic regime, the diode's dark current can be significantly suppressed. The presence of a built-in electric field additionally gives rise to temporal short-circuit current and open-circuit voltage under zero bias. 

Sponsors:

bottom of page