ABSTRACT
A structural change between amorphous and crystalline phase provides a basis for reliable and modular photonic and electronic devices, such as nonvolatile memory, beam steerers, solid-state reflective displays, or mid-IR antennas. In this paper, we leverage the benefits of liquid-based synthesis to access phase-change memory tellurides in the form of colloidally stable quantum dots. We report a library of ternary MxGe1-xTe colloids (where M is Sn, Bi, Pb, In, Co, Ag) and then showcase the phase, composition, and size tunability for Sn-Ge-Te quantum dots. Full chemical control of Sn-Ge-Te quantum dots permits a systematic study of structural and optical properties of this phase-change nanomaterial. Specifically, we report composition-dependent crystallization temperature for Sn-Ge-Te quantum dots, which is notably higher compared to bulk thin films. This gives the synergistic benefit of tailoring dopant and material dimension to combine the superior aging properties and ultrafast crystallization kinetics of bulk Sn-Ge-Te, while improving memory data retention due to nanoscale size effects. Furthermore, we discover a large reflectivity contrast between amorphous and crystalline Sn-Ge-Te thin films, exceeding 0.7 in the near-IR spectrum region. We utilize these excellent phase-change optical properties of Sn-Ge-Te quantum dots along with liquid-based processability for nonvolatile multicolor images and electro-optical phase-change devices. Our colloidal approach for phase-change applications offers higher customizability of materials, simpler fabrication, and further miniaturization to the sub-10 nm phase-change devices.
ABSTRACT
The typically nonlinear and asymmetric response of synaptic memristors to positive and negative electrical pulses makes the realization of accurate deep neural networks very challenging. Here, we integrate a two-terminal valence change memory (VCM) into a photonic/plasmonic circuit and show that the switching properties of this memristor become more gradual and symmetric under light irradiation. The added optical input acts on the VCM as a third, independent modulation channel. It locally heats the active area of the device, which enhances the generation of oxygen vacancies and broadens the resulting nanoscale conductive filaments. The measured conductance modulation of the VCM is then inserted into a neural network simulator. Using the MNIST data set of handwritten digits as an application, a light-enhanced recognition accuracy of 93.53% is demonstrated, similar to ideally performing memristors (94.86%) and much higher than those without light (67.37%). Notably, the optical signal does not increase the overall energy consumption by more than 3.2%. Finally, an approach to scale up our electro-optical technology is proposed, which could allow high-density, energy-efficient neuromorphic computing chips.
