Increased performance of nanophotonic devices by utilizing structural fluctuation information with deep learning

2022 Fiscal Year Final Research Report

• Project/Area Number: 21K18912
• Research Category: Grant-in-Aid for Challenging Research (Exploratory)
• Allocation Type: Multi-year Fund
• Review Section: Medium-sized Section 30:Applied physics and engineering and related fields
• Research Institution: Keio University
• Principal Investigator: Tanabe Takasumi 慶應義塾大学, 理工学部(矢上), 教授 (40393805)
• Project Period (FY): 2021-07-09 – 2023-03-31
• Keywords: 光エレクトロニクス / フォトニック結晶 / ナノフォトニクス

Outline of Final Research Achievements

We have demonstrated a spectrometer using a chirped photonic crystal waveguide. Nanophotonic devices usually have limits due to fabrication precision. In particular, the effects of fabrication errors are significant. On the other hand, in this study, we have achieved a wavelength resolution exceeding the fabrication precision by using the light localization patterns resulting from fabrication fluctuations. We measure the localization pattern at each wavelength in advance and map it into a two-dimensional matrix. By multiplying the inverse matrix of this two-dimensional matrix by the pattern of unknown wavelength input, we reconstruct the spectrum. We have successfully achieved a spectral resolution of less than 0.8 nm.

Free Research Field

光エレクトロニクス

Academic Significance and Societal Importance of the Research Achievements

フォトニック結晶(PhC)やプラズモニクス，メタマテリアル，集積光回路を始めとするナノフォトニクス研究は，光科学の一大分野を形成しているが，その性能は作成誤差によって制限をうける．従来は作成誤差との戦いであったが，本研究はその戦略を180度転換するものであり，作成誤差をソフトウェアの力も借りながら積極的に活用しようとするものである．本手法は全く新しいものである，ナノフォトニクス分野の発展に大きな変革をもたらすことが期待される．