Regensburg 2022 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

HL: Fachverband Halbleiterphysik

HL 30: Poster 2

HL 30.3: Poster

Thursday, September 8, 2022, 11:00–13:00, P3

Machine Learning enhanced in-situ electron beam lithography of photonic nano-structures — •Jan Donges, Marvin Schlischka, Ching-Wen Shih, Monica Pengerla, Imad Limame, Johannes Schall, Lucas Rickert, Sven Rodt, and Stephan Reitzenstein — Technische Universitaet Berlin, Hardenbergstr. 36, 10623 Berlin, Germany

The unique in-situ electron beam lithography (iEBL) nanotechnology concept enables us to embed single quantum emitters into photonic nano-structures with a 34nm precision. To obtain the fitted position of the quantum emitter with high accuracy, the signal-to-noise ratio of their catholuminescence (CL) needs to be well above unity even for ms integration times. Here we show that for samples with dark quantum emitters, such as telecom quantum dots or low planar photon-extraction efficiency, machine learning (ML) is very well suited to drastically improve the performance of iEBL at low CL intensities. The machine learning software utilizes computer algorithms, which were trained through data samples to denoise data with a barely visible quantum dot emission. We present experimental results for InGaAs quantum dots, which could be successfully embedded into circular Bragg grating structures with the assistance of machine learning. Our experimental results yield that by using ML, the CL sensitivity and the alignment accuracy could be increased by more than an order of magnitude compared to standard iEBL.