Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

AKBP: Arbeitskreis Beschleunigerphysik

AKBP 3: Diagnostics

AKBP 3.6: Vortrag

Montag, 9. März 2026, 16:15–16:30, SCH/A117

Automated In-Situ Ion Beam Characterization During Ion Implantation — •Ali Kosari Mehr, René Heller, Oliver Steuer, and Slawomir Prucnal — Institute of Ion Beam Physics and Materials Research, Helmholtz-Zentrum Dresden-Rossendorf, Bautzner Landstr 400, 01328 Dresden, Germany

Ion implantation is a key technique for materials modification, especially in semiconductor processing. However, high-current industrial implanters often struggle to distinguish ion species and energies when projectiles have similar mass-to-charge ratios, such as Al and N*. Because industrial implantation relies on fast throughput and simple mass-analysis stages, ambiguous beam composition can compromise process reliability. This motivates the development of an automated in-situ diagnostic tool. A compact apparatus combining Rutherford backscattering spectrometry and ionoluminescence was therefore developed. It is inserted into the beam path for only a few seconds before implantation and comprises a beam-entry aperture, a dummy target of known composition, an RBS detector, and an ionoluminescence module with mirrors, optical fibres, and a miniature spectrometer. The dual-modality concept enables cross-checked identification of ion species, energies, and molecular fractions. For automated beam-quality decisions, a machine-learning evaluation scheme based on an artificial neural network was implemented and trained on simulated RBS spectra. The optimised model reliably distinguishes target ions under realistic noise and intensity fluctuations, enabling real-time ion beam characterisation.

Keywords: Ion Implantation; In-situ Beam Diagnostics; Rutherford Backscattering Spectrometry; Machine Learning; Artificial Neural Network