Data fusion by artificial neural network for hybrid metrology development - Université Grenoble Alpes Accéder directement au contenu
Article Dans Une Revue Journal of Vacuum Science & Technology A Année : 2021

Data fusion by artificial neural network for hybrid metrology development

Résumé

Deposition of titanium nitride thin films by plasma enhanced atomic layer deposition has been realized on thermal silicon oxide substrates in an inductively coupled plasma reactor. The plasma step involves a H2 (40 sccm)/N2 (5 sccm)/Ar (10 sccm) gas mixture, and growth has been followed by in situ ellipsometric measurements. A tunable substrate bias voltage has been applied in the vicinity of the substrate to modulate plasma-ion energy and investigate its impact on the growth mechanism. We have observed that an increase in the applied bias power leads to a gradual TiN nucleation delay of up to 30 cycles at 80 W radio frequency bias power. An increase in the H2 content of the plasma gas mixture shows that hydrogen species from the plasma can significantly deactivate the SiO2 substrate, thanks to reduction reactions induced by H3+, Ar+, and ArH+ ions leading to the formation of Si–H surface bonds. A nitrogen-rich plasma gas mixture results in N atom incorporation on the substrate surface, which in turn favors subsequent TiN growth. The combination of hydrogen-rich plasma chemistry with a high applied substrate bias power leads to a TiN growth delay larger than 50 cycles. These results provide a valuable implementation for the development of area-selective deposition processes.
Fichier non déposé

Dates et versions

hal-04627269 , version 1 (27-06-2024)

Identifiants

Citer

Lucien Penlap Woguia, Jerome Rêche, Maxime Besacier, Patrice Gergaud, Guido Rademaker. Data fusion by artificial neural network for hybrid metrology development. Journal of Vacuum Science & Technology A, 2021, 39 (1), pp.84. ⟨10.1117/12.2583590⟩. ⟨hal-04627269⟩
0 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More