Colloidal quantum dots (QDs) based on III-V semiconductors offer solution-processable, size-tunable optical properties that comply with toxic heavy metal regulations in optoelectronic applications. While global research has primarily focused on InP and InAs QDs – spanning a spectral range from the mid-infrared to the visible – colloidal synthesis protocols now extend to InSb QDs for mid-wave infrared absorption, and molten-salt reactions have enabled access to Ga-based III-V compounds. Interestingly, the wavelength range accessible through size tuning often overlaps for In- and Ga-based semiconductors; for example, red-emitting QDs can be synthesized from both InP and GaAs. Such changes in composition and size alter material properties such as lattice parameters and band structure, enabling device performance optimization by tuning QD composition and size while maintaining a fixed operational wavelength. This raises the fundamental question of whether the field should converge on a single material system or continue exploring the compositional diversity of III-V QDs.

In this presentation, we apply density functional theory (DFT) and the Bloch orbital expansion as an unconventional computational approach to link QD geometry with electronic structure. By comparing the fuzzy band structure of QDs with their bulk band structures computed at the same theoretical level, we distinguish strongly confined, delocalized QD orbitals from those derived from bulk surface states, which deviate from the bulk bands. Notably, mid-gap surface states – most detrimental to optoelectronic device performance – can be readily identified as falling within the bulk bands. Additionally, the first delocalized QD orbitals can be identified via this approach, defining a trap band width for each material.

We first apply the method to a 3 nm InP QD model with chloride-passivated (100) and (111) facets, showing that its fuzzy bands exhibit a broad band of occupied surface states related to unpassivated (-111) facets exposing P dangling bonds. Remarkably, changing the composition to GaAs – a material with similar band gap energy – eliminates these mid-gap states. Extending our analysis across various III-V QD sizes and compositions, we observe that (1) indirect band gap materials exhibit larger trap band widths than direct band gap materials, (2) valence trap band width decreases significantly across the P>As>Sb series, with a similar trend for Al>In>Ga, and (3) larger QD models exhibit additional surface states due to extended crystal facets and associated defects. We conclude by relating our findings to previously published dangling bond energies.[1]