In recent years, research into cadmium- and lead-free nanocrystals (NCs) has intensified due to toxicity concerns associated with the use of heavy-metals in consumer products. Copper indium sulfide (CIS) has emerged as a promising alternative material due to its size dependent optical properties that are tunable from the visible to the near-infrared. However, bare CIS NCs possess low photoluminescence quantum yields (PLQY) and low stability under ambient conditions. A commonly used strategy to improve these properties is to overcoat the CIS NCs with a ZnS shell, which yields NCs with PLQYs as high as 70% and high photostability. The overgrowth of ZnS over CIS cores is typically accompanied by a blue shift of both the absorption and the PL spectra. The origin of this blue shift is however still unclear, and several possible explanations have been proposed: etching and size reduction of the CIS core prior to ZnS shell overgrowth, alloying, cation exchange, or interfacial strain due to the lattice mismatch between the ZnS shell and the CIS core. These different possibilities are nevertheless not mutually exclusive, and it is thus possible that the synthesis outcome depends on the interplay between several competing processes. Understanding the factors that determine the balance between these processes is crucially relevant, since it would lead to a better control over the properties of CIS/ZnS core/shell NCs.

In this study, we investigated the role of the reaction conditions (precursor reactivity and temperature) in the process of overgrowing a ZnS shell around 2.5 nm diameter CIS NCs. Our results show that the compositional profile of the product CIS/ZnS NCs is dictated by a delicate competition between cation exchange, interdiffusion, and heteroepitaxial overgrowth, which can lead to gradient (CIS,ZnS) alloy NCs with variable compositional profiles or to CIS/ZnS core/shell hetero-NCs, depending on the reaction conditions. The conditions required to achieve controlled ZnS overgrowth are however very stringent, and therefore typically some degree of alloying occurs, leading to the spectral blue shifts commonly reported for CIS/ZnS NCs. Nevertheless, we show that, by choosing the right reaction parameters, CIS/ZnS NCs exhibiting red-shifted optical spectra can be obtained, consistent with a core/shell heteroarchitecture in which shell overgrowth occurred in the absence of significant cation exchange and interdiffusion. The ability to shift the optical spectra to both higher and lower energies provides an extra design tool for CIS NCs.