view article

Figure 1
From a time-dependent XRR data set R(q,t) a CNN predicts ten parameters of a time-dependent layer-by-layer growth model. From these parameters, the thin film growth scenario with thickness and roughness evolution is calculated. For different growth scenarios, we create simulated XRR and synthetic R(q,t) data sets that can be used to train the CNN before applying it to real data.

Journal logoJOURNAL OF
APPLIED
CRYSTALLOGRAPHY
ISSN: 1600-5767
Follow J. Appl. Cryst.
Sign up for e-alerts
Follow J. Appl. Cryst. on Twitter
Follow us on facebook
Sign up for RSS feeds