Bending Resistant Multibit Memristor for Flexible Precision Inference Engine Application

Pal, Parthasarathi; Lee, Ke-Jing; Thunder, Sunanda; De, Sourav; Huang, Po-Tsang; Kämpfe, Thomas; Wang, Yeong-Her

doi:10.1109/ted.2022.3186965

Cited by 11 publications

(11 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The mechanical characteristics of flexible memristors are also significantly influenced by the electrode materials . Currently, commonly used electrode materials include metal electrodes (Ag (refs , , , , , , , ), Ti (refs , ) Al (refs , , , , , , , , , ) W (ref ) Pt (ref ), and Au (refs , , , , , , )), oxide electrodes (ITO (refs , , , , , , , , , , , , , ) and AZO (ref )), and carbon/nitrogen electrodes (TiN , ). Notably, flexible transparent electrodes such as ITO and AZO, which enable better optoelectronic performance, are attractive for next-generation smart wearables and transparent nanoelectronic devices.…”

Section: Fabrication and Integration Of Flexible Memristorsmentioning

confidence: 99%

“…Therefore, special consideration should be given to the mechanical characteristics, electrical conductivity, heat resistance, chemical stability, and surface uniformity of the flexible substrate. Currently, the commonly used flexible substrates in flexible memristors are PET (refs , , , , , , , , , , , , , , , ), PI (refs , , ), PEN (refs , ), mica (refs , , ), and so on. Among them, PET and PEN are widely used in the preparation of flexible memristors because of their excellent mechanical stability, high transparency, and low cost. , However, lower glass temperatures ( T g ) and coefficients of thermal expansion (CTE) allow them to be used only in low-temperature manufacturing applications.…”

Section: Fabrication and Integration Of Flexible Memristorsmentioning

confidence: 99%

“…56 The device has good thermal stability and recoverability, and the RS performance can be maintained for 10 3 cycles in the bent state with a bending diameter of 15 mm, as shown in Figure 6a,b. A 2-bit HfO x -stacked RRAM prepared on a flexible polyimide substrate was reported by Pal et al 82 This device can be used to simulate neural networks, modeled by erasure characteristics and measured for interdevice and interperiodic variation in standard deviation. Therefore, the device can be used for inference engineering applications with offline training.…”

Section: Flexible Memristors Based On Metalmentioning

confidence: 99%

See 2 more Smart Citations

Flexible Memristor-Based Nanoelectronic Devices for Wearable Applications: A Review

Rao,

Wang,

Mao

et al. 2023

ACS Appl. Nano Mater.

View full text Add to dashboard Cite

In the high-tech and intelligent era of the 21st century, as one of the important electronic devices, wearable electronic products can effectively improve work efficiency and quality of life. Since memristors have broad application prospects in information storage, neural synapses, neural networks, neuromorphic computation, electronic skin (e-skin), and brain-like chips, the development of flexible electronic devices based on memristors has received extensive attention. However, the incompatibility between the flexible substrate and the functional layer, the performance degradation after repeated bending, and the instability of devices in the biological environment greatly limit the commercial application of flexible memristors. In this review, the characteristics and applications of different types of functional layer nanomaterial-based memristors are first analyzed, and the preparation and integration of the devices are discussed. Then the development bottleneck of flexible memristors is introduced, and corresponding solutions are proposed. Finally, we look forward to the development of intelligent flexible memristors and point out the direction for the development of wearable electronic products based on flexible memristors, which promotes research of flexible memristors in artificial intelligence.

show abstract

Section: Fabrication and Integration Of Flexible Memristorsmentioning

confidence: 99%

Section: Fabrication and Integration Of Flexible Memristorsmentioning

confidence: 99%

Section: Flexible Memristors Based On Metalmentioning

confidence: 99%

See 1 more Smart Citation

Flexible Memristor-Based Nanoelectronic Devices for Wearable Applications: A Review

Rao,

Wang,

Mao

et al. 2023

ACS Appl. Nano Mater.

View full text Add to dashboard Cite

show abstract

“…Still, their power-hungry nature makes them inappropriate for end-user applications. The need for low-power, fast and edge-device compatible real-time data processing units motivated the scientists to research non-von-Neumann architectures with higher energy efficiency and lower latency [21][22][23][24][25][26][27][28][29][30].…”

Section: Introductionmentioning

confidence: 99%

Neuromorphic Computing with 28nm High-K-Metal Gate Ferroelectric Field Effect Transistors Based Artificial Synapses

Raffel¹,

Müller²,

Thunder³

et al. 2023

Preprint

View full text Add to dashboard Cite

This paper presents a comprehensive overview of 28 nm high-k-metal gate-based ferroelectric field effect transistor devices for synaptic applications. The device under test was fabricated on 300mm wafers at GlobalFoundries. The fabricated devices demonstrate 103 WRITE-endurance cycles and 104 seconds of data-retention capability at 85°C. We have also assessed the FeFET-based crossbar array’s performance in system-level applications. By simulating the FeFET crossbar array for neuromorphic applications, the system performance was assessed. For datasets from the National Institute of Standards and Technology (MNIST), the crossbar array achieved software-comparable inference accuracy of about 97% using multilayer perceptron (MLP) neural networks.

show abstract

“…Ionic doping and surface treatments in metal oxides are useful techniques that can improve the switching stability of the resistive device performance [25][26][27][28]. Typically, RS can be classified into volatile and non-volatile switching categories, depending on their conducting filament sustainability.…”

Section: Introductionmentioning

confidence: 99%

Controllable coexistence of threshold and non-volatile crosspoint memory for highly linear synaptic device applications

Pal

Singh

Wang

2023

J. Phys. D: Appl. Phys.

Self Cite

View full text Add to dashboard Cite

A highly reliable and versatile resistive memory device that demonstrates threshold and non-volatile (NVM) switching behaviour depending on the compliance current (CC) modulation was utilised by doping a semiconducting (Si) material into a high-k (HfOx) film with highly linear synaptic behaviour. The device shifted towards volatile switching at a CC less than 1 µA and exhibited NVM behaviour at a CC limit above 10 µA. A 3-bit/cell data storage capability on RESET voltage modulation was implemented for high-density memory application. The device exhibited excellent programming linearity of potentiation/depression responses up to 10000 pulses compatible with fast pulse (100 ns) with good ION/IOFF ratio (>10^3), stable data retention capability (10^5 s) at 85°C and high WRITE endurance (~10^7 cycles) with a pulse width of 200 ns. The neuromorphic applications were successfully emulated through neural network simulations using the experimentally calibrated data of the Si-doped HfOx resistive cross-point devices. Simulation results revealed a low nonlinearity of 0.03 with 98.08% pattern recognition accuracy. The estimated results revealed the potential of the device as a low-power selector and high-density NVM storage in large-scale crossbar array in future neuromorphic computing applications.

show abstract

Bending Resistant Multibit Memristor for Flexible Precision Inference Engine Application

Cited by 11 publications

References 42 publications

Flexible Memristor-Based Nanoelectronic Devices for Wearable Applications: A Review

Flexible Memristor-Based Nanoelectronic Devices for Wearable Applications: A Review

Neuromorphic Computing with 28nm High-K-Metal Gate Ferroelectric Field Effect Transistors Based Artificial Synapses

Controllable coexistence of threshold and non-volatile crosspoint memory for highly linear synaptic device applications

Contact Info

Product

Resources

About