Memlumor : A Luminescent Memory Device for Energy-Efficient Photonic Neuromorphic Computing

Summary, in English

Neuromorphic computing promises to transform the current paradigm of traditional computing toward non-von Neumann dynamic energy-efficient problem solving. To realize this, a neuromorphic platform must possess intrinsic complexity reflected in the built-in diversity of its physical operation mechanisms. We propose and demonstrate the concept of a memlumor, an all-photonic device combining memory and a luminophore, and being mathematically a full equivalence of the electrically driven memristor. Using CsPbBr₃ perovskites as a material platform, we demonstrate the synergetic coexistence of memory effects within a broad time scale from nanoseconds to minutes and switching energy down to 3.5 fJ. We elucidate the origin of such a complex response to be related to the phenomena of photodoping and photochemistry activated by a tunable light input. When the existence of a history-dependent photoluminescence quantum yield is leveraged in various material platforms, the memlumor device concept will trigger multiple new research directions in both material science and photonics.

• Author: Alexandr Marunchenko; Jitendra Kumar; Alexander Kiligaridis; Dmitry Tatarinov; Anatoly Pushkarev; Yana Vaynzof; Ivan G. Scheblykin
• Department/s: Chemical Physics; LU Profile Area: Light and Materials; NanoLund: Centre for Nanoscience; LTH Profile Area: Nanoscience and Semiconductor Technology; LTH Profile Area: Photon Science and Technology
• Publishing year: 2024
• Language: English
• Pages: 2075-2082
• Publication/Series: ACS Energy Letters
• Volume: 9
• Issue: 5
• Links: http://dx.doi.org/10.1021/acsenergylett.4c00691
• Document type: Journal article
• Publisher: The American Chemical Society (ACS)
• Topic: Condensed Matter Physics (including Material Physics, Nano Physics)
• Status: Published
• ISBN/ISSN/Other: ISSN: 2380-8195