Abstract

Electrically detected magnetic resonance (EDMR) is a powerful technique for the observation and categorization of paramagnetic defects within semiconductors. The interpretation of the recorded EDMR spectra has long proved to be challenging. Here, defect spectra are identified by comparing EDMR measurements with extensive ab initio calculations. The defect identification is based upon the defect symmetry and the form of the hyperfine (HF) structure. A full description is given of how an accurate spectrum can be generated from the theoretical data by considering some thousand individual HF contributions out of some billion possibilities. This approach is illustrated with a defect observed in nitrogen implanted silicon carbide (SiC). Nitrogen implantation is a high energy process that gives rise to a high defect concentration. The majority of these defects are removed during the dopant activation anneal, shifting the interstitial nitrogen to the desired substitutional lattice sites, where they act as shallow donors. EDMR shows that a deep-level defect persists after the dopant activation anneal. This defect is characterized as having a gc∥B=2.0054(4) and gc⊥B=2.0006(4), with pronounced hyperfine shoulder peaks with a 13 G peak to peak separation. The nitrogen at a carbon site next to a silicon vacancy (NCVSi) center is identified as the persistent deep-level defect responsible for the observed EDMR signal and the associated dopant deactivation.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call