Elucidating the Evolving Atomic Structure in Atomic Layer Deposition Reactions with in Situ XANES and Machine Learning

Orlando Trejo, Anup L. Dadlani, Francisco De La Paz, Shinjita Acharya, Rob Kravec, Dennis Nordlund, Ritimukta Sarangi, Fritz B. Prinz, Jan Torgersen, Neil P. Dasgupta

Publikation: Beitrag in FachzeitschriftArtikelBegutachtung

23 Zitate (Scopus)


Precision synthesis of thin films requires an improved mechanistic understanding of the structural evolution of materials at the atomic scale. Atomic layer deposition (ALD) is a critical nanofabrication technique that enables fine-tuning of atomic structure and thickness as a result of its layer-by-layer growth behavior. In this study, in situ X-ray absorption spectra of the S K-edge during ALD growth of ZnS thin films on TiO2 nanoparticles were collected and analyzed. The experimental results show that both sulfide and sulfate species form during the nucleation phase of ZnS on TiO2. As film growth proceeds, the S K-edge of the in situ ZnS film converges to that of a representative ex situ ALD ZnS film. By building representative atomistic models, a high-throughput screening method was developed to determine the most probable atomic configurations as the film structure evolves. The screening method consisted of a supervised machine learning analysis of thousands of simulated X-ray absorption near edge structure (XANES) spectra. Atomic-level insight was gained into changes in the coordination environment of surface species as they transitioned from the nucleation phase toward the crystalline ZnS phase. The experimental and computational strategies presented herein provide an example of how in situ synchrotron-based characterization can be leveraged using robust modeling approaches to elucidate the ordering of atoms during thin-film growth.

FachzeitschriftChemistry of Materials
PublikationsstatusAngenommen/Im Druck - 2019
Extern publiziertJa


Untersuchen Sie die Forschungsthemen von „Elucidating the Evolving Atomic Structure in Atomic Layer Deposition Reactions with in Situ XANES and Machine Learning“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren