Effect of dielectric confinement on optical properties of colloidal nanostructures

  • PDF / 785,651 Bytes
  • 13 Pages / 612 x 792 pts (letter) Page_size
  • 58 Downloads / 251 Views

DOWNLOAD

REPORT


ontribution for the JETP special issue in honor of L.V. Keldysh’s 85th birthday

Effect of Dielectric Confinement on Optical Properties of Colloidal Nanostructures1 A. V. Rodinaa,* and Al. L. Efrosb,** a Ioffe

Physicotechnical Institute, Russian Academy of Sciences, St. Petersburg, 194021 Russia b Naval Research Laboratory DC 20375, Washington, USA *e-mail: [email protected] **e-mail: [email protected] Received November 17, 2015

Abstract—We review the effects caused by a large difference in the dielectric constants of a semiconductor and its surrounding in colloidal semiconductor nanostructures (NSs) with various shapes, e.g., nanocrystals, nanorods, and nanoplatelets. The difference increases the electron–hole interaction and consequently the exciton binding energy and its oscillator transition strength. On the other hand, this difference reduces the electric field of a photon penetrating the NS (the phenomenon is called the local field effect) and reduces the photon coupling to an exciton. We show that the polarization properties of the individual colloidal NSs as well as of their randomly oriented ensemble are determined both by the anisotropy of the local field effect and by the symmetry of the exciton states participating in optical transitions. The calculations explain the temperature and time dependences of the degree of linear polarization measured in an ensemble of CdSe nanocrystals. DOI: 10.1134/S1063776116030183

1. INTRODUCTION In low-dimensional semiconducting structures, the free motion of electrons and holes is limited by their spatial confinement. In quantum wells, which are referred to as 2D structures in what follows, their motion is limited in one dimension. In quantum wires, to be referred to as 1D structures, their motion is limited in two dimensions, and, finally, in quantum dots and nanocrystals, which are sometimes called 0D structures, their motion is limited in all three dimensions. These confinements generally increase the energy of electron–hole Coulomb interactions and modify (most commonly, increase) the oscillator transition strength of the band-edge optical transitions. In addition to the spatial confinement of free-carrier motion, the low-dimensional semiconductor structures are characterized by a difference in the dielectric constants inside the structures, e in , and in the surrounding media, e out . Such a difference leads to an additional dielectric confinement and affects the electron–hole Coulomb interaction. The dielectric confinement effect is usually very small in semiconductor/semiconductor heterostructures due to very close 1The article is published in the original.

values of the dielectric constants, i.e., when the ratio or dielectric contrast is k = e in /e out ≈ 1. In 1979, Keldysh [1] demonstrated that the dielectric confinement strongly affects excitons in the semiconductor/dielectric structures. Several dielectric confinement effects were discussed for semiconductor/dielectric and semiconductor/vacuum structures [1–6], where the contrast k can be as large