Semiconductor nanocrystals (SNCs) have become the most important material for colloidal nanophotonics – a rapidly advancing research field that evolves into a powerful technological platform for lighting, biomedicine, lasing, photovoltaics, etc. Over the years major strides in this field were made for II-VI and IV-VI SNCs, due to the availability of precursors and relative simplicity of synthetic protocols for obtaining high-quality SNCs with varying shape, tunable composition and structure, and surface properties. However recently their attractiveness was severely undercut by tightening restrictions on the use of elements comprising such crystals (e.g. Cd, Pb and Hg) due to their toxicity. This highlights the importance of alternative materials among which III-V semiconductors stand out: The nature of their composing elements is relatively benign and they exhibit improved chemical stability due to a higher covalent bond share in the materials.

Although the first syntheses of III-V colloidal NCs can be traced back to the early 90s, their development was much slower than that of their II-VI and IV-VI counterparts, and many unresolved challenges still exist. One such attractive research avenue is the realization of shape control III-V NCs, in particular the synthesis of two-dimensional (2D) nanoplatelets (NPls). The interest in such nanocrystals is mainly driven by the studies of cadmium chalcogenide NPls which showed that they exhibit much superior properties to their counterparts with other shapes, i.e., among others, narrow emission and absorbance bands, high absorption coefficients directed emission, etc. Regardless of the recent advancements in understanding the driving forces of the directed growth of cadmium chalcogenide NPls [1,2], the proposed models have limited predictive power and currently, it is impossible to straightforwardly use or adapt existing protocols to the direct synthesis of III-V NPls.

In this work, we investigate a more general indirect approach consisting in the Cu-for-In cation exchange in pre-synthesised Cu_3-xP NPls. This approach benefits from the inherent structural anisotropy of Cu_3-xP driving their 2D growth into NPls, which can subsequently serve as templates for InP NPls. In addition, this approach can be potentially further expanded to other indium pnictides as well as a broad range of heterostructured NPls, which are hard or impossible to obtain using direct synthesis. A few reports have already demonstrated the possibility of cation exchange in Cu_3-xP, however, they mostly focused on the process itself and have not yielded InP NPls with pronounced absorbance features and noticeable emission, which was related to the excessive amounts of the residual copper atoms. [3,4] In this work, we further optimize the cation exchange conditions, such as the effect of ligands, and complexing additives to tune the balance between incoming and outgoing cations, which is important to avoid a considerable amount of vacancies and domains that act as centers for nonradiative recombination of charge carriers.