Donor-acceptor-type conjugated polymers are widely used in memristors due to their unique push-pull electron structures and charge transfer mechanisms. However, the inherently inhomogeneous microstructure of polymer films and their low crystallinity produce randomness that destabilizes formed conductive channels, giving polymer-based memristors unstable switching behavior. In this contribution, we prepared a synaptic device based on PM6-MoS 2 QD (molybdenum disulfide quantum dot) nanocomposites. In the composites, MoS 2 QDs provided the active centers for forming conductive channels via electron trapping and detrapping. They also controlled the directional formation of conductive channels between PM6 and MoS 2 QDs, reducing randomness and giving devices a narrow switching voltage range and cycling longevity. The device exhibited continuous multistage conductance states under a direct current voltage sweep and simulated a variety of synaptic functions, including long-term potentiation, long-term depression, short-term potentiation, short-term depression, paired-pulse facilitation, spiking-rate-dependent plasticity, and "learning experience" behavior. The memristor could also perform arithmetic, including "counting" and "subtraction" operations. This work provides a new approach to improving the performance of memristors for neuromorphic computing.
Keyphrases
- reduced graphene oxide
- gold nanoparticles
- particulate matter
- air pollution
- depressive symptoms
- room temperature
- quantum dots
- photodynamic therapy
- prefrontal cortex
- solar cells
- visible light
- polycyclic aromatic hydrocarbons
- heavy metals
- computed tomography
- magnetic resonance imaging
- highly efficient
- mass spectrometry
- high intensity