Emerging 2D Ferroelectric Devices for In-Sensor and In-Memory Computing

An Electrode Design Strategy to Minimize Ferroelectric Imprint Effect

The phenomenon of ferroelectric imprint, characterized by an asymmetric polarization switching behavior, poses significant challenges in the reliability and performance of ultra-low-voltage ferroelectric devices, including MagnetoElectric Spin-Orbit devices, Ferroelectric Random-Access Memory, Ferroelectric Field-Effect Transistors, and Ferroelectric Tunnel Junctions. In this study, the influence of electrode configuration in different device architectures are systematically investigated on their imprint effect. By tuning the work function of La0.7Sr0.3MnO3 (LSMO) electrodes through oxygen pressure during deposition, precise control over the built-in voltage offset (Voffset) in ferroelectric capacitors are demonstrated. This results reveal that higher oxygen pressures increase the work function of LSMO, effectively compensating for Voffset and enhancing device stability. Finally, a ferroelectric device with a hybrid bottom electrode of LSMO and SrRuO3 is optimized to mitigate the imprint effect. The optimal device showcases small coercive voltage of 0.3 V, a minimal Voffset of 0.06 V, excellent endurance (electrical cycle up to 109), and robust zero bias applied polarization retention. These findings provide a practical guideline for electrode design in ferroelectric devices, addressing the imprint effect and improving operational reliability. This approach, combining material tuning and in situ diagnostics, offers a pathway to optimize ferroelectric device performance, with implications for advancing ultra-low-power electronics.

Plasmonic Hot-Electron Effect Enhanced WSe2 Based Transistor Based on Asymmetric Schottky Contacts for Self-Powered Photodetection and Visual Synapse

Schottky interfaces in metal-semiconductor contacts are crucial in optoelectronics, with a focus on enhancing the detection performance. The plasmonic hot-electron effect offers efficient photon-to-electricity conversion, boosting the sensitivity in self-powered photodetectors as well as expanding detection wavelength ranges and improving the functionality of metal-semiconductor-metal (M1-S-M2) structured photodetectors. Utilizing two-dimensional WSe2 nanoflakes, we fabricate a transistor with a M1-S-M2 structure featuring Au and Ag electrodes. Under 405 nm light, we achieve a maximum specific detectivity (D*) of 9.23 × 1011 Jones with a photocurrent density of 4.6 mA cm-2 and a peak on/off ratio of 6.88 × 105. Compared with the Au/WSe2/Au transistor also on the poly(ethylene terephthalate) (PET) substrate, the maximum current density measured for the Au/WSe2/Ag transistor under the light of 405 nm is 1.3 times that of the former, and the D* measured under the light of 1064 nm increases to 20 times the original value; it is now capable of detecting the 1550 nm light, which was undetectable previously. These data clearly demonstrate that the transistor exhibits excellent photodetection performance in the visible and near-infrared spectra. In addition, under biased voltage conditions, the transistor can effectively simulate the visual synaptic behavior under the stimulation of visible light and near-infrared light. Due to its simple structure, wide detection range, excellent light detection performance, and remarkable synaptic plasticity characteristics, this transistor has great potential in various applications of light detection technology and artificial vision systems.

Ion-Electron Interactions in 2D Nanomaterials-Based Artificial Synapses for Neuromorphic Applications

With the increasing limitations of conventional computing techniques, particularly the von Neumann bottleneck, the brain's seamless integration of memory and processing through synapses offers a valuable model for technological innovation. Inspired by biological synapse facilitating adaptive, low-power computation by modulating signal transmission via ionic conduction, iontronic synaptic devices have emerged as one of the most promising candidates for neuromorphic computing. Meanwhile, the atomic-scale thickness and tunable electronic properties of van der Waals two-dimensional (2D) materials enable the possibility of designing highly integrated, energy-efficient devices that closely replicate synaptic plasticity. This review comprehensively analyzes advancements in iontronic synaptic devices based on 2D materials, focusing on electron-ion interactions in both iontronic transistors and memristors. The challenges of material stability, scalability, and device integration are evaluated, along with potential solutions and future research directions. By highlighting these developments, this review offers insights into the potential of 2D materials in advancing neuromorphic systems.

Multistate Polarization and Enhanced Nonreciprocal Transport in Two-Dimensional van der Waals Ferroelectric Heterostructures

Achieving multiple switchable polarization states at the nanoscale is crucial to high-density nonvolatile multistate memory beyond bistable ferroelectric architectures. Here, we propose a novel strategy to realize multistate polarization and enhance nonreciprocal transport in two-dimensional (2D) van der Waals ferroelectric heterostructures. By integrating two distinct 2D ferroelectric materials with substantial spontaneous polarizations, we demonstrate that the Bi/SnTe heterostructure can support up to eight distinct polarization states. Our first-principles analysis of transforming paths and corresponding energy barriers reveals that these states can be mutually switched by applying external electric fields, facilitated by a combination of intralayer polar distortion and interlayer sliding. Moreover, the Bi/SnTe heterostructure exhibits significantly enhanced nonlinear Hall and kinetic magnetoelectric effects, closely correlated to the multistate in-plane and persistent out-of-plane polarization. These findings open new possibilities for designing advanced ferroelectric devices with multiple polarization states and enhanced nonreciprocal transport, offering a pathway toward next-generation memory and nanoelectronics.

Photonic-Electronic Modulated a-IGZO Synaptic Transistor with High Linearity Conductance Modulation and Energy-Efficient Multimodal Learning

Brain-inspired neuromorphic computing is expected to overcome the von Neumann bottleneck by eliminating the memory wall between processing and memory units. Nevertheless, critical challenges persist in synaptic device implementation, particularly regarding nonlinear/asymmetric conductance modulation and multilevel conductance states, which substantially impede the realization of high-performance neuromorphic hardware. This study demonstrates a novel advancement in photonic-electronic modulated synaptic devices through the development of an amorphous indium-gallium-zinc oxide (a-IGZO) synaptic transistor. The device demonstrates biological synaptic functionalities, including excitatory/inhibitory post-synaptic currents (EPSCs/IPSCs) and spike-timing-dependent plasticity, while achieving excellent conductance modulation characteristics (nonlinearity of 0.0095/-0.0115 and asymmetric ratio of 0.247) and successfully implementing Pavlovian associative learning paradigms. Notably, systematic neural network simulations employing the experimental parameters reveal a 93.8% recognition accuracy on the MNIST handwritten digit dataset. The a-IGZO synaptic transistor with photonic-electronic co-modulation serves as a potential critical building block for constructing neuromorphic architectures with human-brain efficiency.

Challenges and Opportunities of Upconversion Nanoparticles for Emerging NIR Optoelectronic Devices

Upconversion nanoparticles (UCNPs), incorporating lanthanide (Ln) dopants, can convert low-energy near-infrared photons into higher-energy visible or ultraviolet light through nonlinear energy transfer processes. This distinctive feature has attracted considerable attention in both fundamental research and advanced optoelectronics. Challenges such as low energy-conversion efficiency and nonradiative losses limit the performance of UCNP-based optoelectronic devices. Recent advancements including optimized core-shell structures, tailed Ln-doping concentration, and surface modifications show significant promise for improving the efficiency and stability. In addition, combining UCNPs with functional materials can broaden their applications and improve device performance, paving the way for the innovation of next-generation optoelectronics. This paper first categorizes and elaborates on various upconversion mechanisms in UCNPs, focusing on strategies to boost energy transfer efficiency and prolong luminescence. Subsequently, an in-depth discussion of the various materials that can enhance the efficiency of UCNPs and expand their functionality is provided. Furthermore, a wide range of UCNP-based optoelectronic devices is explored, and multiple emerging applications in UCNP-based neuromorphic computing are highlighted. Finally, the existing challenges and potential solutions involved in developing practical UCNPs optoelectronic devices are considered, as well as an outlook on the future of UCNPs in advanced technologies is provided.

Phase-Engineered In2Se3 Ferroelectric P-N Junctions in Phototransistors for Ultra-Low Power and Multiscale Reservoir Computing

Two-dimensional (2D) ferroelectric field-effect transistors (Fe-FETs) based on p-n junctions are the basic units of future neuromorphic hardware. The In2Se3 semiconductor with ferroelectric, photoelectric, and phase transition properties possesses great application potential for in-sensor computing, but its ferroelectric p-n junction (FePNJ) is not well investigated. Here, we present an optoelectronic synapse made of uniformly full-coverage α-In2Se3/WSe2 FePNJ, achieving ultralow-power classification recognition and multiscale signal processing. Using chemical vapor deposition (CVD), we can obtain β'-In2Se3/WSe2 subferroelectric p-n junctions by direct growth on SiO2/Si substrate and α-In2Se3/WSe2 FePNJ by phase transition. Modulated by the synergistic effect of the polarization electric field and the built-in electric field, the FePNJ exhibits significantly enhanced and highly tunable synaptic effects (memory retention >2500 s and >8 multilevel current states under single optical/electrical pulses), along with power consumption down to atto-joule levels. Utilizing these photoelectric properties, we constructed an all-ferroelectric in-sensor reservoir computing system, comprising both reservoir and readout networks, achieving ultralow-power handwritten digit recognition. We also created a multiscale reservoir computing system through the gate-voltage-modulated relaxation time scale of the FePNJ, which can efficiently detect motions in the 1 to 100 km h-1 speed range.

Synergistic Control of Ferroelectric and Optical Properties in Molecular Ferroelectric for Multiplexing Nonvolatile Memory

Utilizing the correlation among diverse physical properties to facilitate multiplexing and multistate memory is anticipated to emerge as an efficient strategy to enhance memory capacity, achieve device miniaturization, and ensure information security. As an important functional material, ferroelectrics have long been considered as a potential candidate in multistate memory devices. Furthermore, the integration of optical response offers an alternative path to realizing multiplexing features, further enhancing the versatility and efficiency of these devices. However, combining ferroelectricity and optical activity is always challenging because ferroelectricity is very sensitive to the crystal structure. In this study, on the correlation between ferroelectric polarization (FP) and optical properties in molecular ferroelectric material, trimethylchloromethyl ammonium trichloromanganese (TMCM-MnCl3) is reported. This research demonstrated that the FP can modulate the photoluminescence (PL) emission, while optical illumination can trigger FP reversal. Based on these, both electric-writing optical-reading (EWOR) and optical-writing electrical-reading (OWER) modes have been conclusively established, and the seamless transition between these two modes can be achieved by adjusting the excitation light intensity. These findings reveal an intriguing physical interconnection and imply the viability of implementing multiplexing and multistate memory functionalities in systems based on ferroelectrics.

Emerging 2D materials hardware for in-sensor computing

The advent of the novel in-sensor/near-sensor computing paradigm significantly eliminates the need for frequent data transfer between sensory terminals and processing units by integrating sensing and computing functions into a single device. This approach surpasses the traditional configuration of separate sensing and processing units, thereby greatly simplifying system complexity. Two-dimensional materials (2DMs) show immense promise for implementing in-sensor computing systems owing to their exceptional material properties and the flexibility they offer in designing innovative device architectures with heterostructures. This review highlights recent progress and advancements in 2DM-based in-sensor computing research, summarizing the unique physical mechanisms that can be leveraged in 2DM-based devices to achieve sensory responses and the essential biomimetic synaptic characteristics for computing functions. Additionally, the potential applications of 2DM-based in-sensor computing systems are discussed and categorized. This review concludes with a perspective on future development directions for 2DM-based in-sensor computing.

Self-Powered Infrared-Detectable BP/Ta2NiS5 Heterojunction and Its Application in Energy-Efficient Optoelectronic Synapses

The development of energy-efficient and high-performance optoelectronic devices is crucial for the advancement of modern optoelectronic and microelectronic systems. Although the self-powered devices and optoelectronic synapses based on 2D heterojunction show great application prospects, the high energy consumption and infrared band detection of self-powered optoelectronic synapses are still an urgent problem to be solved. In this report, a BP/Ta2NiS5 heterojunction is constructed to achieve infrared detection by leveraging differences in Fermi energy levels. This heterojunction exhibits a high specific detectivity of 6.57 × 1010, 2.6 × 1010, and 1.12 × 1010 Jones and responsivity of 20, 10.6, and 5.9 mA W-1 for 1064, 1550, and 2200 nm infrared light at 0 bias voltage, respectively. In addition, under the 2200 nm light, by applying an ultra-low bias voltage of 800 µV, the heterojunction exhibits ultra-low power and energy consumption of 28.8 pW and 0.64 pJ, successfully simulates a variety of synaptic behaviors under infrared light, and demonstrates its image perception and image memory capabilities. These findings position the BP/Ta2NiS5 heterojunction as an ideal candidate for a multifunctional optoelectronic device crucial for advanced photodetection, neuromorphic computing, and artificial intelligence.

Ferroelectric Perovskite/MoS2 Channel Heterojunctions for Wide-Window Nonvolatile Memory and Neuromorphic Computing

Ferroelectric materials commonly serve as gate insulators in typical field-effect transistors, where their polarization reversal enables effective modulation of the conductivity state of the channel material, thereby realizing non-volatile memory. Currently, novel 2D ferroelectrics unlock new prospects in next-generation electronics and neuromorphic computation. However, the advancement of these materials is impeded by limited selectivity and narrow memory windows. Here, new concepts of 2D ferroelectric perovskite/MoS2 channel heterostructures field-effect transistors are presented, in which 2D ferroelectric perovskite features customizable band structure, few-layered ferroelectricity, and submillimeter-size monolayer wafers. Further studies reveal that these devices exhibit unique charge polarity modulation (from n- to p-type channel) and remarkable nonvolatile memory behavior, especially record-wide hysteresis windows up to 177 V, which enables efficient imitation of biological synapses and achieves high recognition accuracy for electrocardiogram patterns. This result provides a device paradigm for future nonvolatile memory and artificial synaptic applications.

Advanced Materials Research at CUHK: From Biomedicine to Electronics and Beyond

This special issue spans a diverse array of topics, including nanomedicine, tissue engineering, regenerative medicine, organs-on-chips, biosensing, soft robotics, smart devices, nanofabrication, energy saving and storage, catalysis, spintronics, soft electronics, and neuromorphic computing. It showcases the breadth and depth of advanced materials research at the Chinese University of Hong Kong (CUHK), highlighting the innovation, collaboration, and excellence of CUHK's materials scientists.

Object Motion Detection Enabled by Reconfigurable Neuromorphic Vision Sensor under Ferroelectric Modulation

Increasing the demand for object motion detection (OMD) requires shifts of reducing redundancy, heightened power efficiency, and precise programming capabilities to ensure consistency and accuracy. Drawing inspiration from object motion-sensitive ganglion cells, we propose an OMD vision sensor with a simple device structure of a WSe2 homojunction modulated by a ferroelectric copolymer. Under optical mode and intermediate ferroelectric modulation, the vision sensor can generate progressive and bidirectional photocurrents with discrete multistates under zero power consumption. This design enables reconfigurable devices to emulate long-term potentiation and depression for synaptic weights updating, which exhibit 82 states (more than 6 bits) with a uniform step of 6 pA. Such OMD devices also demonstrate nonvolatility, reversibility, symmetry, and ultrahigh linearity, achieving a fitted R2 of 0.999 and nonlinearity values of 0.01/-0.01. Thus, a vision sensor could implement motion detection by sensing only dynamic information based on the brightness difference between frames, while eliminating redundant data from static scenes. Additionally, the neural network utilizing a linear result can recognize the essential moving information with a high recognition accuracy of 96.8%. We also present the scalable potential via a uniform 3 × 3 neuromorphic vision sensor array. Our work offers a platform to achieve motion detection based on controllable and energy-efficient ferroelectric programmability.

Simulating and Implementing Broadband van der Waals Artificial Visual Synapses Based on Photoconductivity and Pyroconductivity Mechanisms

With advancements in artificial neural networks and information processing technology, a variety of neuromorphic synaptic devices have been proposed to emulate human sensory systems, with vision being a crucial information source. Moreover, as practical applications become increasingly complex, the need for multifunctional visual synapses to expand the application range becomes urgent. This study introduces a MoS2/WSe2 van der Waals (vdW) heterojunction and utilizes it to replicate artificial visual synapses by harnessing the cooperative effect of photoconductivity and pyroconductivity mechanisms. By adjusting the optical power, pulse width, and pulse number of the optical stimulus, the heterojunction effectively simulates synaptic properties. Under the combined action of an external electric field and the built-in electric field (Ebi), the heterojunction exhibits broadband synaptic properties in the visible to near-infrared spectrum (405-1550 nm) while consuming low power of 0.3-1.1 pJ per spike. The heterojunction can detect ultraweak optical signals at 660 nm with an optical power intensity of 14 μW/cm2, displaying a high specific detectivity (D*) of 3.98 × 1011 Jones. Furthermore, at 405, 808, 1064, and 1550 nm, the D* of the heterojunction is 4.16 × 1011, 3.61 × 109, 4.96 × 107, and 1.64 × 107 Jones, respectively. Visual synaptic devices based on the MoS2/WSe2 vdW heterojunction hold significant promise for the future development of integrated sensing and memory processing devices.

Near-Infrared Response Organic Synaptic Transistor for Dynamic Trace Extraction

The development of neuromorphic hardware capable of detecting and recognizing moving targets through an in-sensor computing strategy is considered to be an important component of the construction of edge computing systems with distributed computation. In addition to responsiveness to visible light, the implementation of neuromorphic hardware should also demonstrate the ability to sense and process nonvisible light, which is essential for tracking target object trajectories in specialized environments. In this work, we fabricated an organic synaptic transistor with a near-infrared (NIR) response by incorporating doped LaF3: Yb/Ho upconversion quantum dots (UCQDs) into the channel of a Poly3-hexylthiophene (P3HT)-based organic field effect transistor (FET), serving as charge trapping and infrared sensing sites. The obtained synaptic transistor not only replicates common synaptic behaviors when exposed to NIR illumination but also demonstrates potential applications for the dynamic trajectory recognition of animals in the dark. Compared to other monitoring technologies, P3HT transistors doped with LaF3: Yb/Ho UCQDs exhibit distinct advantages, including a NIR response, high-efficiency computing, and sensitivity, which provide an experimental foundation and a design reference for the development of next-generation intelligent dynamic image recognition systems.