神经形态工程学
材料科学
场效应晶体管
晶体管
光电子学
领域(数学)
纳米技术
人工神经网络
计算机科学
电气工程
电压
数学
机器学习
工程类
纯数学
作者
S. De Stefano,O. Durante,Anna Sessa,Antonio Politano,Gianluca D’Olimpio,Tsotne Dadiani,Enver Faella,Adrian Dinescu,Cătălin Pârvulescu,Crispin Hetherington,Chia‐Nung Kuo,C. S. Lue,Martino Aldrigo,M. Passacantando,Antonio Di Bartolomeo
标识
DOI:10.1021/acsami.5c11651
摘要
As artificial intelligence continues to evolve, neuromorphic technologies, which emulate biological neural networks, are increasingly seen as a promising direction. Two-dimensional materials are considered promising for neuromorphic applications due to their tunable electrical and optoelectronic properties. In this work, a back-gated tin disulfide (SnS2) field-effect transistor (FET) is electrically and optoelectronically characterized at different temperatures (80, 295, and 380 K), pressures (ambient and 10-4 mbar), and illumination conditions (dark and laser light from 420 to 800 nm). Responsivity peaks of up to ∼100 A/W are recorded. Persistent photoconductivity is observed, with current retention after illumination ranging from 0% to ∼30% of the initial dark current, depending on temperature and gate voltage. The underlying microscopic mechanisms are analyzed, revealing a key role for trap states and ambient adsorbates, and a qualitative model is proposed to explain the observed effects. Trap states within the bandgap, often considered detrimental, are exploited to induce synaptic plasticity, with synaptic weight changes tunable from 0.001 to 3000. Temperature and gate voltage are found to be effective parameters for modulating plasticity, enabling smooth transitions between short-term and long-term behavior. These results clarify the microscopic origin of plasticity in SnS2, demonstrate its robustness under realistic conditions, and lay the foundation for the integration of this two-dimensional material into next-generation neuromorphic architectures.
科研通智能强力驱动
Strongly Powered by AbleSci AI