18 June 2019 in The Journal of Physical Chemistry Letters
26 June 2024 in Chemistry of Materials
17 February 2015 in Physical Review B
The three-dimensional confinement characterizing a nanocrystal (NC) leads to the formation of discrete electronic states. The energy gap between these states in colloidal NCs can be up to an order of magnitude larger than the vibrational energy of the host material. This large energetic mismatch (not given in self-assembled quantum dots) leads to the expectation that an electron occupying an excited state would be unable to release its energy to vibrations and a “phonon bottleneck” should finally be observed. Using large-scale ab initio calculations and a time-dependent formalism, we show that on the contrary, a phonon bottleneck can be observed only in a narrow window of diameters for CdSe and InAs NCs and should not occur at all in Si NCs. Two relaxation pathways enable fast carrier relaxation. For smaller structures (below 20-Å radius), the coupling strength and energy detuning are such that quantum mechanics allows us to effectively bridge electronic gaps much larger than the vibronic energy. For larger structures, the coupling to passivant modes, although very weak, leads to an efficient picosecond carrier relaxation. This work provides insight into the nature of carrier relaxation in colloidal nanostructures and highlights that defects, of any kind, are not necessary to explain the observed fast carrier relaxation.
9 December 2019 in The Journal of Chemical Physics
We calculate the optical properties of InP and CdSe colloidal quantum dots (QDs) within the framework of the atomic effective pseudopotential approach and the screened configuration interaction theory. We obtain an excellent agreement with experiment with our microscopic and space-dependent screening function where the dielectric constant varies in real space with a sharp transition (width of ≈0.18 nm) from the QD material high-frequency bulk value inside the QD to the solvent or passivant high-frequency value outside. We obtain a reasonable agreement (with deviations less than 140 meV) for a computationally less demanding solvent-independent screening using the full high-frequency bulk screening, in contrast to the more commonly used reduced QD radius-dependent screening constant. We show theoretically that for QDs passivated with long-chained organic molecules, the influence of the solvent on the optical gap is in the range of 10 meV, while QDs passivated with short ligands can experience shifts in the order of 100 meV. Experiments on CdSe QDs passivated with octadecylphosphonic acid (ODPA, long-chained ligand) in two different solvents (toluene and chloroform) confirm the bandgap dependence. While the optical gap is weakly affected by the environment, the quasiparticle gap and the exciton binding energy show a strong environmental dependence. Finally, we show that the optical bandgap does not depend significantly on the crystal structure (wurtzite or zincblende) or the morphological details (faceted or “spherical” shape).
27 January 2022 in The Journal of Physical Chemistry C
27 September 2012 in Physical Review B
We derive an analytic connection between the screened self-consistent effective potential from density-functional theory (DFT) and atomic effective pseudopotentials (AEPs). The motivation to derive AEPs is to address structures with thousands to hundreds of thousands of atoms, as given in most nanostructures. The use of AEPs allows us to bypass a self-consistent procedure and to address eigenstates around a certain region of the spectrum (e.g., around the band gap). The bulk AEP construction requires two simple DFT calculations of slightly deformed elongated cells. The ensuing AEPs are given on a fine reciprocal space grid, including the small reciprocal vector components, are free of adjustable parameters, and involve no fitting procedure. We further show how to connect the AEPs of different bulk materials, which is necessary to obtain accurate band offsets. We derive a total of 20 AEPs for group-III-V, II-VI, and IV semiconductors and demonstrate their accuracy and transferability by comparison to DFT calculations of strained bulk structures, quantum wells with varying thickness, and semiconductor alloys.