Electrochemical charging of films of semiconductor nanocrystals (NCs) allows precise control over their Fermi level and opens up new possibilities for use of semiconductor NCs in optoelectronic devices. Unfortunately, charges added to the semiconductor NCs are often lost due to electrochemical side reactions. In this work, we examine which loss processes can occur in electrochemically charged semiconductor NC films by comparing numerical drift-diffusion simulations with experimental data. Both reactions with impurities in the electrolyte solution, as well as reactions occurring on the surface of the nanomaterials themselves, are considered. We show that the Gerischer kinetic model can be used to accurately model the one-electron transfer between charges in the semiconductor NC and oxidant or reductant species in solution. Simulations employing the Gerischer model are in agreement with experimental results of charging of semiconductor NC films with ideal one-electron acceptors ferrocene and cobaltocene. We show that reactions of charges in the semiconductor NC film with redox species in solution are reversible when the reduction potential is in the conduction band of the semiconductor NC material but are irreversible when the reduction potential is in the band gap. Experimental charging of semiconductor NC films in the presence of oxygen is always irreversible in our system, even when the reduction potential of oxygen is in the conduction band of the semiconductor NC material. We show that the Gerischer model in combination with a coupled reversible-irreversible reaction mechanism can be used to model oxygen reduction. Finally, we model irreversible reduction reactions with the semiconductor NC material itself, such as reduction of ligands or surface ions. Simulations of semiconductor NC cyclic voltammograms in the presence of material reduction reactions strongly resemble experimental cyclic voltammograms of InP and CdSe NC films. This marks material reduction reactions at the semiconductor NC surface as a likely candidate for the irreversible behavior of these materials in electrochemical experiments. These results show that all reduction reactions with redox potentials in the band gap of semiconductor NCs must be suppressed in order to achieve stable charging of these materials.

Indium phosphide (InP) quantum dots (QDs) are considered the most promising alternative for Cd and Pb-based QDs for lighting and display applications. However, while core-only QDs of CdSe and CdTe have been prepared with near-unity photoluminescence quantum yield (PLQY), this is not yet achieved for InP QDs. Treatments with HF have been used to boost the PLQY of InP core-only QDs up to 85%. However, HF etches the QDs, causing loss of material and broadening of the optical features. Here, we present a simple postsynthesis HF-free treatment that is based on passivating the surface of the InP QDs with InF_3. For optimized conditions, this results in a PLQY as high as 93% and nearly monoexponential photoluminescence decay. Etching of the particle surface is entirely avoided if the treatment is performed under stringent acid-free conditions. We show that this treatment is applicable to InP QDs with various sizes and InP QDs obtained via different synthesis routes. The optical properties of the resulting core-only InP QDs are on par with InP/ZnSe/ZnS core-shell QDs, with significantly higher absorption coefficients in the blue, and with potential for faster charge transport. These are important advantages when considering InP QDs for use in micro-LEDs or photodetectors.