Multiscale modeling of semiconductor nanocrystal synthesis in templating media
Several liquid-phase and vapor-phase techniques are reported in the literature for growing nanostructured materials. The use of templates in the synthesis of nanostructured materials is attractive as it combines precise control of size and shape with easy scale up for industrial production. This work aims to elucidate the underlying mechanisms controlling the growth rate and morphology of II-VI compound semiconductor nanocrystals (also called quantum dots) in templating media by employing theory, modeling, and simulation. The focus of the work is on zinc selenide (ZnSe), which can form nanocrystals emitting in the blue and violet part of the visible spectrum when excited by ultra violet radiation.
We developed a lattice Monte Carlo simulation technique to describe the formation of ZnSe quantum dots by reacting diethylzinc with hydrogen selenide in the spherical nanodroplets of a microemulsion formed by self-assembly of a ternary system consisting of an amphiphilic block copolymer, a polar continuous phase (formamide) and a non-polar dispersed phase (heptane) . The stochastic model describes diffusion of diethylzinc molecules in heptane, nucleation of ZnSe through a fast reaction between diethylzinc and hydrogen selenide at the interface between the droplet and the continuous phase, as well as diffusion and coalescence of ZnSe clusters inside the nanodroplet, eventually leading to the formation of a single nanocrystal per nanodroplet. The motion of molecules and clusters in the lattice is programmed according to their diffusivity, which is estimated by the Stokes-Einstein equation. A deterministic diffusion-reaction model describing diethylzinc depletion in a spherical droplet due to a fast interfacial reaction was used to investigate different growth regimes and compare its predictions with those of the stochastic model. In the early stages of the nanocrystal formation process, slow diffusion of hydrogen selenide through the surfactant layer is the rate determining step. The zinc precursor is progressively depleted inside the nanodroplet and its diffusion to the interface becomes the rate controlling step. This transition can be tracked precisely by our stochastic model without any assumptions, but not by the aforementioned deterministic model. The formation of stable clusters (also called "magic clusters") of ZnSe with a fullerene-like close-caged structure has been included in the stochastic simulations. The predicted size variation of the final nanocrystals due to the formation of such clusters has been studied using this stochastic model. (Abstract shortened by UMI.)