Two-dimensional mutually synchronized spin Hall nano-oscillator arrays for neuromorphic computing

• Published in: Nature nanotechnology Vol. 15; no. 1; pp. 47 - 52
• Main Authors: Zahedinejad, Mohammad, Awad, Ahmad A., Muralidhar, Shreyas, Khymyn, Roman, Fulara, Himanshu, Mazraati, Hamid, Dvornik, Mykola, Åkerman, Johan
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 01.01.2020; Nature Publishing Group
• Subjects: 639/166/987; 639/766/119/1001; 639/766/119/2793; 639/766/119/995; 639/925/927/1007; Arrays; Auto-oscillations; Brillouin scattering; Chemistry and Materials Science; Cognitive ability; Cognitive tasks; Computation; Condensed Matter Physics; Constrictions; controlled study; current density; Den kondenserade materiens fysik; Electrical Engineering, Electronic Engineering, Information Engineering; Elektroteknik och elektronik; Light scattering; Magnetic films; Materials Engineering; Materials Science; Materialteknik; Micro-brillouin; microscopy; Microwave frequencies; Microwave oscillators; microwave radiation; Microwave signal generations; Mutual synchronization; Nanotechnology; Nanotechnology and Microengineering; Neuromorphic computing; oscillation; Oscillator networks; Oscillators; priority journal; Q factors; Short time scale; Signal generation; signal processing; Signal quality; Speech recognition; Synchronism; Synchronization; Vowel recognition; White noise
• ISSN: 1748-3387; 1748-3395; 1748-3395
• DOI: 10.1038/s41565-019-0593-9

In spin Hall nano-oscillators (SHNOs), pure spin currents drive local regions of magnetic films and nanostructures into auto-oscillating precession. If such regions are placed in close proximity to each other they can interact and may mutually synchronize. Here, we demonstrate robust mutual synchronization of two-dimensional SHNO arrays ranging from 2 × 2 to 8 × 8 nano-constrictions, observed both electrically and using micro-Brillouin light scattering microscopy. On short time scales, where the auto-oscillation linewidth Δ f is governed by white noise, the signal quality factor, Q = f ∕ Δ f , increases linearly with the number of mutually synchronized nano-constrictions ( N ), reaching 170,000 in the largest arrays. We also show that SHNO arrays exposed to two independently tuned microwave frequencies exhibit the same synchronization maps as can be used for neuromorphic vowel recognition. Our demonstrations may hence enable the use of SHNO arrays in two-dimensional oscillator networks for high-quality microwave signal generation and ultra-fast neuromorphic computing. Synchronization of oscillators can be used to carry out cognitive tasks. Large two-dimensional arrays of synchronized spin Hall nano-oscillators have now been demonstrated, and may in future enable neuromorphic computing on the nanoscale.