Browse
Submit Data
Databases
API
Help

Dataset Information

18 Views

0 Connections

0 Citations

0 Reanalyses

0 Downloads

Omics score: 0

An in-memory computing architecture based on two-dimensional semiconductors for multiply-accumulate operations.

ABSTRACT: In-memory computing may enable multiply-accumulate (MAC) operations, which are the primary calculations used in artificial intelligence (AI). Performing MAC operations with high capacity in a small area with high energy efficiency remains a challenge. In this work, we propose a circuit architecture that integrates monolayer MoS₂ transistors in a two-transistor-one-capacitor (2T-1C) configuration. In this structure, the memory portion is similar to a 1T-1C Dynamic Random Access Memory (DRAM) so that theoretically the cycling endurance and erase/write speed inherit the merits of DRAM. Besides, the ultralow leakage current of the MoS₂ transistor enables the storage of multi-level voltages on the capacitor with a long retention time. The electrical characteristics of a single MoS₂ transistor also allow analog computation by multiplying the drain voltage by the stored voltage on the capacitor. The sum-of-product is then obtained by converging the currents from multiple 2T-1C units. Based on our experiment results, a neural network is ex-situ trained for image recognition with 90.3% accuracy. In the future, such 2T-1C units can potentially be integrated into three-dimensional (3D) circuits with dense logic and memory layers for low power in-situ training of neural networks in hardware.

SUBMITTER: Wang Y

PROVIDER: S-EPMC8184885 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Json Xml

Similar Datasets

Two-dimensional semiconductors pave the way towards dopant-based quantum computing.

Project description:Since the proposal in 1998 to build a quantum computer using dopants in silicon as qubits, much progress has been made in the nanofabrication of semiconductors and the control of charge and spins in single dopants. However, an important problem remains unsolved, namely the control over exchange interactions and tunneling between two donors, which presents a peculiar oscillatory behavior as the dopants relative positions vary at the scale of the lattice parameter. Such behavior is due to the valley degeneracy in the conduction band of silicon, and does not occur when the conduction-band edge is at k = 0. We investigate the possibility of circumventing this problem by using two-dimensional (2D) materials as hosts. Dopants in 2D systems are more tightly bound and potentially easier to position and manipulate. Moreover, many of them present the conduction band minimum at k = 0, thus no exchange or tunnel coupling oscillations. Considering the properties of currently available 2D semiconductor materials, we access the feasibility of such a proposal in terms of quantum manipulability of isolated dopants (for single qubit operations) and dopant pairs (for two-qubit operations). Our results indicate that a wide variety of 2D materials may perform at least as well as, and possibly better, than the currently studied bulk host materials for donor qubits.

| S-EPMC6204835 | biostudies-literature

In-Memory Logic Operations and Neuromorphic Computing in Non-Volatile Random Access Memory.

Project description:Recent progress in the development of artificial intelligence technologies, aided by deep learning algorithms, has led to an unprecedented revolution in neuromorphic circuits, bringing us ever closer to brain-like computers. However, the vast majority of advanced algorithms still have to run on conventional computers. Thus, their capacities are limited by what is known as the von-Neumann bottleneck, where the central processing unit for data computation and the main memory for data storage are separated. Emerging forms of non-volatile random access memory, such as ferroelectric random access memory, phase-change random access memory, magnetic random access memory, and resistive random access memory, are widely considered to offer the best prospect of circumventing the von-Neumann bottleneck. This is due to their ability to merge storage and computational operations, such as Boolean logic. This paper reviews the most common kinds of non-volatile random access memory and their physical principles, together with their relative pros and cons when compared with conventional CMOS-based circuits (Complementary Metal Oxide Semiconductor). Their potential application to Boolean logic computation is then considered in terms of their working mechanism, circuit design and performance metrics. The paper concludes by envisaging the prospects offered by non-volatile devices for future brain-inspired and neuromorphic computation.

| S-EPMC7475900 | biostudies-literature

Two-dimensional ferroelectric channel transistors integrating ultra-fast memory and neural computing.

Project description:With the advent of the big data era, applications are more data-centric and energy efficiency issues caused by frequent data interactions, due to the physical separation of memory and computing, will become increasingly severe. Emerging technologies have been proposed to perform analog computing with memory to address the dilemma. Ferroelectric memory has become a promising technology due to field-driven fast switching and non-destructive readout, but endurance and miniaturization are limited. Here, we demonstrate the α-In2Se3 ferroelectric semiconductor channel device that integrates non-volatile memory and neural computation functions. Remarkable performance includes ultra-fast write speed of 40 ns, improved endurance through the internal electric field, flexible adjustment of neural plasticity, ultra-low energy consumption of 234/40 fJ per event for excitation/inhibition, and thermally modulated 94.74% high-precision iris recognition classification simulation. This prototypical demonstration lays the foundation for an integrated memory computing system with high density and energy efficiency.

| S-EPMC7782550 | biostudies-literature

Phase-matching-free parametric oscillators based on two-dimensional semiconductors.

Project description:Optical parametric oscillators are widely used as pulsed and continuous-wave tunable sources for innumerable applications, such as quantum technologies, imaging, and biophysics. A key drawback is material dispersion, which imposes a phase-matching condition that generally entails a complex design and setup, thus hindering tunability and miniaturization. Here we show that the burden of phase-matching is surprisingly absent in parametric micro-resonators utilizing mono-layer transition-metal dichalcogenides as quadratic nonlinear materials. By the exact solution of nonlinear Maxwell equations and first-principle calculations of the semiconductor nonlinear response, we devise a novel kind of phase-matching-free miniaturized parametric oscillator operating at conventional pump intensities. We find that different two-dimensional semiconductors yield degenerate and non-degenerate emission at various spectral regions due to doubly resonant mode excitation, which can be tuned by varying the incidence angle of the external pump laser. In addition, we show that high-frequency electrical modulation can be achieved by doping via electrical gating, which can be used to efficiently shift the threshold for parametric oscillation. Our results pave the way for the realization of novel ultra-fast tunable micron-sized sources of entangled photons-a key device underpinning any quantum protocol. Highly miniaturized optical parametric oscillators may also be employed in lab-on-chip technologies for biophysics, detection of environmental pollution and security.

| S-EPMC6107017 | biostudies-literature

One-dimensional semimetal contacts to two-dimensional semiconductors

Project description:Two-dimensional (2D) semiconductors are promising in channel length scaling of field-effect transistors (FETs) due to their excellent gate electrostatics. However, scaling of their contact length still remains a significant challenge because of the sharply raised contact resistance and the deteriorated metal conductivity at nanoscale. Here, we construct a 1D semimetal-2D semiconductor contact by employing single-walled carbon nanotube electrodes, which can push the contact length into the sub-2 nm region. Such 1D–2D heterostructures exhibit smaller van der Waals gaps than the 2D–2D ones, while the Schottky barrier height can be effectively tuned via gate potential to achieve Ohmic contact. We propose a longitudinal transmission line model for analyzing the potential and current distribution of devices in short contact limit, and use it to extract the 1D–2D contact resistivity which is as low as 10−6 Ω·cm2 for the ultra-short contacts. We further demonstrate that the semimetal nanotubes with gate-tunable work function could form good contacts to various 2D semiconductors including MoS2, WS2 and WSe2. The study on 1D semimetal contact provides a basis for further miniaturization of nanoelectronics in the future. 2D semiconductors are attracting increasing attention as potentially scalable channels for future transistors, but the scaling of their contact length remains challenging. Here, the authors report the realization of 1D semimetal-2D semiconductor contacts based on individual carbon nanotubes with contact length down to 2 nm.

| S-EPMC9825564 | biostudies-literature

A 619-pixel machine vision enhancement chip based on two-dimensional semiconductors.

Project description:The rapid development of machine vision applications demands hardware that can sense and process visual information in a single monolithic unit to avoid redundant data transfer. Here, we design and demonstrate a monolithic vision enhancement chip with light-sensing, memory, digital-to-analog conversion, and processing functions by implementing a 619-pixel with 8582 transistors and physical dimensions of 10 mm by 10 mm based on a wafer-scale two-dimensional (2D) monolayer molybdenum disulfide (MoS2). The light-sensing function with analog MoS2 transistor circuits offers low noise and high photosensitivity. Furthermore, we adopt a MoS2 analog processing circuit to dynamically adjust the photocurrent of individual imaging sensor, which yields a high dynamic light-sensing range greater than 90 decibels. The vision chip allows the applications for contrast enhancement and noise reduction of image processing. This large-scale monolithic chip based on 2D semiconductors shows multiple functions with light sensing, memory, and processing for artificial machine vision applications, exhibiting the potentials of 2D semiconductors for future electronics.

| S-EPMC9348785 | biostudies-literature

A learnable parallel processing architecture towards unity of memory and computing.

Project description:Developing energy-efficient parallel information processing systems beyond von Neumann architecture is a long-standing goal of modern information technologies. The widely used von Neumann computer architecture separates memory and computing units, which leads to energy-hungry data movement when computers work. In order to meet the need of efficient information processing for the data-driven applications such as big data and Internet of Things, an energy-efficient processing architecture beyond von Neumann is critical for the information society. Here we show a non-von Neumann architecture built of resistive switching (RS) devices named "iMemComp", where memory and logic are unified with single-type devices. Leveraging nonvolatile nature and structural parallelism of crossbar RS arrays, we have equipped "iMemComp" with capabilities of computing in parallel and learning user-defined logic functions for large-scale information processing tasks. Such architecture eliminates the energy-hungry data movement in von Neumann computers. Compared with contemporary silicon technology, adder circuits based on "iMemComp" can improve the speed by 76.8% and the power dissipation by 60.3%, together with a 700 times aggressive reduction in the circuit area.

| S-EPMC4536493 | biostudies-literature

Tunable rectification in a molecular heterojunction with two-dimensional semiconductors.

Project description:Until now, a specifically designed functional molecular species has been recognized as an absolute necessity for realizing the diode's behavior in molecular electronic junctions. Here, we suggest a facile approach for the implementation of a tailored diode in a molecular junction based on non-functionalized alkyl and conjugated molecular monolayers. A two-dimensional semiconductor (MoS2 and WSe2) is used as a rectifying designer at the alkyl or conjugated molecule/Au interface. From the adjustment of band alignment at molecules/two-dimensional semiconductor interface that can activate different transport pathways depending on the voltage polarity, the rectifying characteristics can be implemented and controlled. The rectification ratio could be widely tuned from 1.24 to 1.83?×?104 by changing the molecular species and type and the number of layers of the two-dimensional semiconductors in the heterostructure molecular junction. Our work sets a design rule for implementing tailored-diode function in a molecular heterojunction structure with non-functionalized molecular systems.

| S-EPMC7075907 | biostudies-literature

Fluorescence intermittency originates from reclustering in two-dimensional organic semiconductors.

Project description:Fluorescence intermittency or blinking is observed in nearly all nanoscale fluorophores. It is characterized by universal power-law distributions in on- and off-times as well as 1/f behaviour in corresponding emission power spectral densities. Blinking, previously seen in confined zero- and one-dimensional systems has recently been documented in two-dimensional reduced graphene oxide. Here we show that unexpected blinking during graphene oxide-to-reduced graphene oxide photoreduction is attributed, in large part, to the redistribution of carbon sp2 domains. This reclustering generates fluctuations in the number/size of emissive graphenic nanoclusters wherein multiscale modelling captures essential experimental aspects of reduced graphene oxide's absorption/emission trajectories, while simultaneously connecting them to the underlying photochemistry responsible for graphene oxide's reduction. These simulations thus establish causality between currently unexplained, long timescale emission intermittency in a quantum mechanical fluorophore and identifiable chemical reactions that ultimately lead to switching between on and off states.

| S-EPMC5322502 | biostudies-literature

Thickness Considerations of Two-Dimensional Layered Semiconductors for Transistor Applications.

Project description:Layered two-dimensional semiconductors have attracted tremendous attention owing to their demonstrated excellent transistor switching characteristics with a large ratio of on-state to off-state current, Ion/Ioff. However, the depletion-mode nature of the transistors sets a limit on the thickness of the layered semiconductor films primarily determined by a given Ion/Ioff as an acceptable specification. Identifying the optimum thickness range is of significance for material synthesis and device fabrication. Here, we systematically investigate the thickness-dependent switching behavior of transistors with a wide thickness range of multilayer-MoS2 films. A difference in Ion/Ioff by several orders of magnitude is observed when the film thickness, t, approaches a critical depletion width. The decrease in Ion/Ioff is exponential for t between 20?nm and 100?nm, by a factor of 10 for each additional 10?nm. For t larger than 100?nm, Ion/Ioff approaches unity. Simulation using technical computer-aided tools established for silicon technology faithfully reproduces the experimentally determined scaling behavior of Ion/Ioff with t. This excellent agreement confirms that multilayer-MoS2 films can be approximated as a homogeneous semiconductor with high surface conductivity that tends to deteriorate Ion/Ioff. Our findings are helpful in guiding material synthesis and designing advanced field-effect transistors based on the layered semiconductors.

| S-EPMC4940740 | biostudies-literature

OmicsDI is part of the ELIXIR infrastructure

OmicsDI is an Elixir interoperability service. Learn more ›

Tweets

OmicsDI Databases

PRIDE
PeptideAtlas
MassIVE
JPOST Repository
Physiome Model Repository

EGA
EVA
ENA
LINCS
PAXDB
Cell Collective

MetaboLights
Metabolomics Workbench
MetabolomeExpress
GNPS
BioModels
FAIRDOMHub

ArrayExpress
dbGaP
ExpressionAtlas
GEO
NODE

Information

Databases
Help
API
Contact us
Code on GitHub
Terms of use
Submit Data