Artificial synapse based on 1,4-diphenylbutadiyne with femtojoule energy consumption

Liu, Jiesong; Li, Zhengjie; Jia, Caihong; Zhang, Weifeng

doi:10.1039/d2cp05417e

Cited by 3 publications

(1 citation statement)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Differently, Li et al recently proposed a circuit strategy for the non-monotonous BCM learning rule with an EDE effect by using load resistance as a regulator for the concentrations of oxygen vacancies and electrons in the active layer of memristors. [31] It is well-known that two competitive working modes with different types of monotonicity are necessary to establish the non-monotonicity of the BCM rule (Figure 1a). [30] Supposing that two different types of monotonicity are reached by introducing two in-built competitive working modes within a device, the non-monotonicity for the BCM rule could be realized using a delicately-designed device, rather than two external stimulus modes or circuit strategy.…”

Section: Introductionmentioning

confidence: 99%

BCM‐Inspired Synapses Constructed with Barrier‐Modulated Coupling Junctions for Enhancing Speech Recognition

Cai,

Liu,

Wang

et al. 2024

Adv Funct Materials

View full text Add to dashboard Cite

Bio‐inspired synaptic devices have garnered considerable interest in neuromorphic computing. The Bienenstock‐Cooper‐Munro (BCM) learning rule stands out as one of the most accurate synaptic models, featuring non‐monotonic behavior and threshold sliding effect, crucial for stable learning processes. The direct device strategy for completely mimicking the BCM rule is a tough issue since the current devices lack two competitive working modes within one device. In this work, a dual‐junction synaptic device with opposite built‐in electric fields using a W/WO2/WO3‐x/Au structure is demonstrated. The devices directly mimic two fundamental features of the BCM rule via a delicately‐designed bandgap engineering strategy. Furthermore, the working mechanisms are investigated and the promising potential of dual‐junction synaptic devices is demonstrated for enhancing speech recognition through Convolutional Neural Network (CNN)‐based digital speech recognition with a remarkable accuracy of 98% through a synaptic array. Even for speech recognition with 13% Gaussian noise, the accuracy remained at 83%. These findings provide a promising strategy for developing BCM‐based synaptic devices for neuromorphic computing applications.

show abstract

Section: Introductionmentioning

confidence: 99%

BCM‐Inspired Synapses Constructed with Barrier‐Modulated Coupling Junctions for Enhancing Speech Recognition

Cai,

Liu,

Wang

et al. 2024

Adv Funct Materials

View full text Add to dashboard Cite

show abstract

Organic synaptic transistor showing ultralow energy consumption with a microscale channel by laser ablation

Wang,

Yin,

Zhang

et al. 2024

Applied Physics Letters

View full text Add to dashboard Cite

Low energy consumption per synaptic event is important for artificial synapses in applications of highly integrated and large-scale neuromorphic computing systems. Reducing the channel length of a synaptic transistor is an effective method to achieve this goal because such devices can work under low operating voltage and current. In this Letter, we use femtosecond laser ablation to fabricate a microscale slit in an Ag film as the channel of an organic synaptic transistor to obtain low energy consumption. The length of the shortest channel is only 1.6 μm. As a result, the device could be driven by a 50 μV drain bias voltage while output 855 pA excitatory postsynaptic current under a gate spike of 50 mV and 30 ms. The calculated energy consumption per synaptic event is 1.28 fJ, which is comparable to that of a biological synapse (1–10 fJ per synaptic event). Femtosecond laser ablation has been demonstrated a rapid and effective process for the fabrication of microscale channel with high resolution for synaptic transistor, showing large potential for the development of neuromorphic electronics.

show abstract

A New Strategy for Optimizing Threshold Voltage of Artificial Neurons

Zhou

Ling

Chen

2023

2023 IEEE International Conference on Control, Electronics and Computer Technology (ICCECT)

View full text Add to dashboard Cite

Artificial synapse based on 1,4-diphenylbutadiyne with femtojoule energy consumption

Abstract: Fast Bienenstock–Cooper–Munro learning rules (BCM), ultra-low energy consumption (∼25 fJ per spike), linear and large conductance changes have been successfully obtained in the 1,4-diphenylbutadiyne (DPDA) memristor device.

Cited by 3 publications

References 50 publications

BCM‐Inspired Synapses Constructed with Barrier‐Modulated Coupling Junctions for Enhancing Speech Recognition

BCM‐Inspired Synapses Constructed with Barrier‐Modulated Coupling Junctions for Enhancing Speech Recognition

Organic synaptic transistor showing ultralow energy consumption with a microscale channel by laser ablation

A New Strategy for Optimizing Threshold Voltage of Artificial Neurons

Contact Info

Product

Resources

About