Quantify point defects in monolayer tungsten diselenide

Abstract

Abstract Point defects may significantly influence the electrical and optoelectronic properties of two-dimensional (2D) tungsten diselenide (WSe2), while precise information about point defects distribution (e.g. species and concentration) in monolayer (ML-) WSe2 are hard to obtain. In this letter, we tried to partly fill this knowledge gap via performing quantitative and statistical analysis of intrinsic point defects in WSe2 monolayers prepared by three so-called main-stream approaches i.e. mechanical exfoliation (ME), chemical vapor deposition (CVD), and molecular beam epitaxy (MBE), which are promising for providing high-quality samples. Via a conjunction of statistic atomic-resolution annular dark-field scanning transmission electron microscopy imaging, software-based automated defect identification and counting, together with image simulations, defect species and concentrations were quantitatively determined. Seven types of intrinsic point defects were identified in ML-WSe2 and the most dominant one is selenium mono-vacancy (VSe) (corresponding to one Se atom missing), irrespective of the synthetic route and growth conditions. Exact contents and diversity of point defects depend on the specific preparation method: CVD grown ML-WSe2 is the most defective (for example, the density of VSe reaches 1.48% in atomic ratio), followed by ME (∼0.85 at% for VSe) and MBE grown samples (∼0.49 at% for VSe). Our results, though still with limited sampling, provide preliminary quantitative information of point defects in ML-WSe2, which can serve as a reference to achieve the precisely controlled large-scale sample growth and establish the structure-property relationship of 2D transition-metal dichalcogenides materials.