| Optical and electronic properties of small Si nanocrystals and selected nanoclusters are examined and critically reviewed within the framework of high level and high accuracy ab initio calculations based on "static" and time-dependent density functional theory, using a hybrid nonlocal exchange and correlation functional. These calculations are supported by sophisticated quadratic configuration interaction and other multi-reference methods, such as multi-reference second order perturbation theory. Similar calculations for Ge and SiGe nanocrystals examine the origin of the gap, the role of surface oxygen or hydrogen and the critical dimensions for visible photoluminescence. The agreement of our theoretical predictions with accurate experimental results is excellent, despite other conflicting experimental and theoretical results. The main sources of conflict in the experimental results are oxygen contamination, preparation conditions and difficulties in size determination. The theoretical controversies are due to either poor treatment of correlation and exchange, or incorrect fitting of empirical parameters. In most of the conflicting studies agreement with experiment or theory is claimed. We illustrate, using as an additional example the Si6 nanocluster, how easily a seemingly good agreement between theoretical and experimental results could be fortuitous and misleading. The results presented here are very promising for future extensions to the accurate study of nanowires and nanoropes as well as for band-gap engineering. |
full paper (pdf, 432 Kb)