Abstract:The von-Neumann bottleneck has constrained computing systems from efficiently operating on the increasingly large demand in data from networks and devices. Silicon (Si) photonics offers a powerful solution for this issue by providing a platform for high-bandwidth, energy-efficient interconnects. Furthermore, memristors have emerged as a fundamental building block for non-volatile data storage and novel computing architectures with powerful in-memory processing capabilities. In this paper, we integrate an Al2O3 memristor into a heterogeneous Si quantum dot microring laser to demonstrate the first laser with non-volatile optical memory. The memristor alters the effective optical modal index of the microring laser cavity by the plasma dispersion effect in the high resistance state (HRS) or Joule heating in the low resistance state (LRS), subsequently controlling the output wavelength of the laser in a non-volatile manner. This device enables a novel pathway for future optoelectronic neuromorphic computers and optical memory chips.

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is to achieve non - volatile optical wavelength tuning by integrating aluminum oxide (Al₂O₃) memristors into heterogeneous silicon quantum - dot microring lasers. This innovation aims to overcome the impact of the von Neumann architecture bottleneck on the performance of computing systems, especially in response to the increasing data processing demands of big - data networks and devices. Specifically, the paper shows how to use memristors to change the effective optical mode index of the microring laser cavity, thereby controlling the output wavelength of the laser in the high - resistance state (HRS) or low - resistance state (LRS). This non - volatile wavelength - tuning ability paves new paths for future optoelectronic neuromorphic computers and optical memory chips. ### Main contributions of the paper 1. **Non - volatile optical memory**: For the first time, a laser diode with non - volatile optical memory function has been realized, and the output wavelength of the laser is controlled by the state change of the memristor. 2. **Technological breakthrough**: Combining the advantages of silicon photonics and memristors, an efficient, low - power optical interconnection solution is provided. 3. **Application prospects**: It provides a basis for future optoelectronic neuromorphic computing and optical memory chips, and is expected to significantly improve the efficiency of data processing and storage. ### Technical details - **Device structure**: The device includes a quantum - dot laser based on indium arsenide (InAs)/gallium arsenide (GaAs), located on a 50 - micron - diameter silicon microring resonator. The memristor consists of an approximately 10 - nanometer - thick Al₂O₃ layer, sandwiched between a silicon layer and an n - GaAs layer, forming a semiconductor - insulator - semiconductor (SIS) structure. - **Working principle**: - **High - resistance state (HRS)**: The effective mode refractive index is reduced through the plasma dispersion effect, resulting in a blue shift of the laser wavelength. - **Low - resistance state (LRS)**: The effective mode refractive index is increased through the Joule heating effect, resulting in a red shift of the laser wavelength. - **Experimental results**: By applying different voltage biases, the switching of the laser wavelength between HRS and LRS can be achieved. The experimental results show that the change in the laser wavelength is consistent with the state change of the memristor, verifying the function of non - volatile optical memory. ### Future prospects - **Multi - level storage**: By adjusting the input voltage bias, multiple low - resistance states can be achieved, thereby supporting multi - bit storage. - **Integrated applications**: Future designs will explore the integration of multiple memristor lasers into circuits for various applications, such as in - memory computing. - **Performance improvement**: Further optimize the device structure and material selection to reduce the switching voltage, increase the switching speed and reliability. In conclusion, this paper proposes an innovative non - volatile optical tuning method, providing new possibilities for future high - performance computing and data - storage technologies.

Heterogeneously Integrated Laser on Silicon with Non-Volatile Wavelength Tuning

High-Speed and Energy-Efficient Non-Volatile Silicon Photonic Memory Based on Heterogeneously Integrated Memresonator

Pulsed Laser Deposition of a Ga2O3 Thin Film for an Optoelectronic Synaptic Device.

Energy-Efficient Photonic Memory Based on Electrically Programmable Embedded III-V/Si Memristors: Switches and Filters

Silicon Photonic Integrated Circuits with Electrically Programmable Non-Volatile Memory Functions.

Silicon Optical Memory: Non-Volatile Optoelectronic Devices via Si-SiO$_2$ Hysteresis Effect

Integrated Photonics with Programmable Non-Volatile Memory

Novel Semiconductor Lasers and Integrated Photonic Devices

Tunable Quantum Dot Lasers Grown Directly on Silicon

Non-volatile heterogeneous III-V/Si photonics via optical charge-trap memory

Photon-Memristive System for Logic Calculation and Nonvolatile Photonic Storage

Optofluidic memory and self-induced nonlinear optical phase change for reservoir computing in silicon photonics

Electrical Programmable Low-loss high cyclable Nonvolatile Photonic Random-Access Memory

Memlumor: a luminescent memory device for photonic neuromorphic computing

Heterogeneous integration of 2D materials on Si charge-coupled devices as optical memory

Fast and energy-efficient non-volatile III-V-on-silicon photonic phase shifter based on memristors

Silicon Photonics Circuits With Programmable Memory Functionality

Laser-amplified Nonvolatile Charge Trapping Effect in Semiconductor Quantum Dot Structures

Integrated optical memristors

Experimental photonic quantum memristor