Quantum dots (QDs) made of III-V semiconductors have been investigated for many years as a more sustainable alternative for cadmium, lead or mercury based chalcogenides for applications involving visible, short-wave infrared or mid-wave infrared light. Led by progress in QD synthesis, these efforts have mainly focused on In-based materials, including InP, InAs and InSb. These In-based pnictides are all direct band gap semiconductors covering – when accounting for size quantization – a spectral range from the mid IR to the edge of the visible. Recently, synthetic methods for Ga-based pnictides have become available. GaP, however, is an indirect semiconductor, and GaAs and GaSb only cover a relatively narrow spectral range. Hence the question what to expect from Ga-based QDs for opto-electronic applications.

In this presentation, we introduce Bloch orbital expansion as a novel and unconventional computational approach to relate geometry and electronic structure in quantum dots. The method is based on the projection of the QD orbitals on bulk Bloch orbitals, and comparing the resulting QD fuzzy band structure with the bulk band structure computed at the same level of theory. Using this approach, strongly confined, delocalized QD orbitals will overlap with the bulk bands, while QD orbitals derived from bulk surface states will deviate from the bulk bands. Most notably, mid-gap surface states, which are most detrimental for the performance of opto-electronic devices, can be readily identified as falling within the bulk bands. Importantly, when using density functional theory (DFT) to compute QD orbitals for a given QD geometry, this approach provides a direct link between the QD surface termination and the appearance of such surface states.

In a first step, we apply the method to models of PbS, HgTe and HgSe QDs 3-4 nm in size. Interestingly, for all these QDs, we demonstrate that the orbitals are free from coupling to bulk surface states. This finding is rooted in the bulk band structure of these materials, and may explain why films of such QDs truly behave as printed semiconductors. Next, we use the coupling of QD orbitals to bulk surface states as an intrinsic quality-control method to screen the promise of different In- and Ga pnictides as an alternative to restricted Cd, Pb or Hg compounds. Using a fixed, 1116 atom QD model with chloride passivated (100) and (111) facets, we show that InP QDs exhibit a broad band of occupied surface states. These orbitals are related to the P-rich (-111) facets, and extend several 100 meV above the valence-band edge. Such a result can be expected for semiconductors with a p-type valence-band and an s-type conduction-band edge, and reflects charge accumulation at the (-111) facet. In line with this interpretation, we observe a gradual suppression of the coupling of QD orbitals with bulk surface states when reducing the difference in electronegativity between the anion and the cation. In particular, the frontier orbitals of GaAs and GaSb appear as delocalized states, which suggests that these compounds could be used as printed semiconductors with properties superior to In-based equivalents. We end by discussing the impact of these findings for research in III-V QDs.