| AVS 63rd International Symposium & Exhibition | |
| Electronic Materials and Photonics | Thursday Sessions |
| Session EM+SS+TF-ThA |
| Session: | Materials and Interfaces for Energy Storage |
| Presenter: | Keren Freedy, University of Virginia |
| Authors: | K.M. Freedy, University of Virginia A. Giri, University of Virginia B.M. Foley, University of Virginia J. Bogan, Dublin City University, Ireland R. O'Conner, Dublin City University, Ireland P.E. Hopkins, University of Virginia S. McDonnell, University of Virginia |
| Correspondent: | Click to Email |
Electron beam evaporation under high vacuum is very widely used for contact deposition in electronic device fabrication. Ti has a low work function and is commonly deposited as a contact or adhesion layer for other metals in silicon-based devices and more recently in 2D semiconductors. Previous work on Ti/MoS2 contacts suggests that electron beam evaporation of Ti under high vacuum results in the formation of TiO2 at the interface while UHV deposition results in unintentional reactions between metallic Ti and the MoS2 substrate (McDonnell et al., 10.1021/acsami.6b00275). Since the majority of reports using Ti as a contact metal utilize HV rather than UHV e-beam processes, understanding the role of process conditions on the properties of this contact/semiconductor interface is of the utmost importance. To avoid the large variability observed in metal/MoS2 contacts, we focus on Si as a case study to investigate the thermal and electronic properties of the Ti/semiconductor interface. The present study examines the effect of deposition rate on the properties of the Ti/Si interface. Electron beam evaporation of Ti onto both hydrogen-terminated and native oxide (001) Si surfaces is performed at varied deposition rates ranging from 0.1 to 5 A/s at pressures of ~10-6 Torr. Prior to ex-situ characterization, the samples are capped with Au in the evaporator to prevent further oxidation in air. Photoelectron spectroscopy measurements reveal that oxide composition is inversely proportional to the deposition rate. X-ray diffraction data shows no evidence of metallic Ti at slower deposition rates. Due to an appreciable partial pressure of O2 at high vacuum, the composition of the deposited material is sensitive to the impingement rate of Ti on the surface. It follows that higher deposition rates should result in a smaller fraction of oxide phases. Electron and phonon transport across the interface will be studied as a function of deposition rate.