Influence of the Dispersion of the Size of the Si Nanocrystals on their Emission Spectra

  • PDF / 252,781 Bytes
  • 5 Pages / 414.72 x 648 pts Page_size
  • 6 Downloads / 194 Views

DOWNLOAD

REPORT


J. B. KHURGIN*, E. W.FORSYTHE*', S. I. KIM", B. S. SYWE.", B. A. KHAN****, and G. S. TOMPA'* * The Johns Hopkins University, Department of ECE, Baltimore MD. *" Structured Materials Incorporated, Piscataway, NJ. Rutgers University, Department of ECE, Piscataway, NJ Philips Laboratories, Briarcliff Manor, NY

ABSTRACT

A systematic study of the PL spectra of Si quantum nanocrystals in the

Si0 2 matrix has been performed. The results have been fitted to a quantum-

confinement model that includes the nanocrystal size dispersion rather than a specific size of the nanocrystal. This serves as a strong confirmation of the confinement-induced nature of the PL. It has been shown that if the dispersion is taken into account, the position of the emission peak as well as the PL width can always be correlated with the average size of the nanocrystal. I. INTRODUCTION Since the discovery of efficient visible light photoluminescence (PL) in porous Si [1, 2] in 1990 there has been renewed interest in the emission properties of all kinds of quantum-sized nanostructures. Strong PL was observed in a wide range of wavelengths - from infrared (IR) to ultraviolet (UV) in a variety of structures such as free-standing porous Si [2] and small oxidized nanocrystallites (NC's) of Si and Ge [3] Electroluminescence studies has also been done on Si and Ge NC's. [4, 5, 6]. While an impressive amount of experimental results have been accumulated in the last few years, the basic question about the origin of the PL in materials that have an indirect bandgap in the bulk form and show little or no PL, remains unresolved. In the first works on the subject the visible PL was attributed to the quantum confinement of the photoexcited carriers [2, 7, 8, 9] and there is a large amount of work providing supporting evidence for this point of view. But, as mentioned in Refs.[10, 11] the dependence of the peak PL wavelength on the mean size of the NC does not fit neatly into the quantum2 confinement model that predicts the blue shift to be proportional to d- , where d is the size of the NC. Therefore, alternative models such as oxidation [12], surface states [13, 14] and others have been proposed. None of these models, however, have been able to explain all features of the light emission from the Si NC and porous Si. The difficulties in finding the proper answer can be traced to two facts: 1. The size of the crystallites, a few nanometers, is of the same order of magnitude, if not smaller, than the size of the exciton in the silicon oxide or siloxene, as well as the spatial extent of the surface state. In other words, the probability of electron-hole recombination is determined by the volume in which they are confined. Since this volume is roughly the same for the exciton, surface state, defect, or for quantum confinement, the oscillator strength should be also roughly the same. 193 Mat. Res. Soc. Symp. Proc. Vol. 358 01995 Materials Research Society

In this respect, making the distinction between the quantum confinement, surface state, and defect makes ver