Conspectus Intuitively, chemists see crystals grow atom-by-atom or molecule-by-molecule, very much like a mason builds a wall, brick by brick. It is much more difficult to grasp that small crystals can meet each other in a liquid or at an interface, start to align their crystal lattices and then grow together to form one single crystal. In analogy, that looks more like prefab building. Yet, this is what happens in many occasions and can, with reason, be considered as an alternative mechanism of crystal growth. Oriented attachment is the process in which crystalline colloidal particles align their atomic lattices and grow together into a single crystal. Hence, two aligned crystals become one larger crystal by epitaxy of two specific facets, one of each crystal. If we simply consider the system of two crystals, the unifying attachment reduces the surface energy and results in an overall lower (free) energy of the system. Oriented attachment often occurs with massive numbers of crystals dispersed in a liquid phase, a sol or crystal suspension. In that case, oriented attachment lowers the total free energy of the crystal suspension, predominantly by removal of the nanocrystal/liquid interface area. Accordingly, we should start by considering colloidal suspensions with crystals as the dispersed phase, i.e., “sols”, and discuss the reasons for their thermodynamic (meta)stability and how this stability can be lowered such that oriented attachment can occur as a spontaneous thermodynamic process. Oriented attachment is a process observed both for charge-stabilized crystals in polar solvents and for ligand capped nanocrystal suspensions in nonpolar solvents. In this last system different facets can develop a very different reactivity for oriented attachment. Due to this facet selectivity, crystalline structures with very specific geometries can be grown in one, two, or three dimensions; controlled oriented attachment suddenly becomes a tool for material scientists to grow architectures that cannot be reached by any other means. We will review the work performed with PbSe and CdSe nanocrystals. The entire process, i.e., the assembly of nanocrystals, atomic alignment, and unification by attachment, is a very complex and intriguing process. Researchers have succeeded in monitoring these different steps with in situ wave scattering methods and real-space (S)TEM studies. At the same time coarse-grained molecular dynamics simulations have been used to further study the forces involved in self-assembly and attachment at an interface. We will briefly come back to some of these results in the last sections of this review.

Despite broad interest in colloidal lead halide perovskite nanocrystals (LHP NCs), their intrinsic fast growth has prevented controlled synthesis of small, monodisperse crystals and insights into the reaction mechanism. Recently, a much slower synthesis of LHP NCs with extreme size control has been reported, based on diluted TOPO/PbBr2 precursors and a diisooctylphosphinate capping ligand. We report new insights into the nucleation, growth, and self-assembly in this reaction, obtained by in situ synchrotron-based small-angle X-ray scattering and optical absorption spectroscopy. We show that dispersed 3 nm Cs[PbBr3] agglomerates are the key intermediate species: first, they slowly nucleate into crystals, and then they release Cs[PbBr3] monomers for further growth of the crystals. We show the merits of a low Cs[PbBr3] monomer concentration for the reaction based on oleate ligands. We also examine the spontaneous superlattice formation mechanism occurring when the growing nanocrystals in the solvent reach a critical size of 11.6 nm.

The growth of two-dimensional platelets of the CdX family (X = S, Se, or Te) in an organic solvent requires the presence of both long- and short-chain ligands. This results in nanoplatelets of atomically precise thickness and long-chain ligand-stabilized Cd top and bottom surfaces. The platelets show a bright and spectrally pure luminescence. Despite the enormous interest in CdX platelets for optoelectronics, the growth mechanism is not fully understood. Riedinger et al. studied the reaction without a solvent and showed the favorable role for short-chain carboxylates for growth in two dimensions. Their model, based on the total energy of island nucleation, shows favored side facet growth versus growth on the top and bottom surfaces. However, several aspects of the synthesis under realistic conditions are not yet understood: Why are both short- and long-chain ligands required to obtain platelets? Why does the synthesis result in both isotropic nanocrystals and platelets? At which stage of the reaction is there bifurcation between isotropic and 2D growth? Here, we report an in situ study of the CdSe nanoplatelet reaction under practical synthesis conditions. We show that without short-chain ligands, both isotropic and mini-nanoplatelets form in the early stage of the process. However, most remaining precursors are consumed in isotropic growth. Addition of acetate induces a dramatic shift toward nearly exclusive 2D growth of already existing mini-nanoplatelets. Hence, although myristate stabilizes mini-nanoplatelets, mature nanoplatelets only grow by a subtle interplay between myristate and acetate, the latter catalyzes fast lateral growth of the side facets of the mini-nanoplatelets.

Colloidal 2D semiconductor nanocrystals, the analogue of solidstate quantum wells, have attracted strong interest in material science and physics. Molar quantities of suspended quantum objects with spectrally pure absorption and emission can be synthesized. For the visible region, CdSe nanoplatelets with atomically precise thickness and tailorable emission have been (almost) perfected. For the near-infrared region, PbS nanosheets (NSs) hold strong promise, but the photoluminescence quantum yield is low and many questions on the crystallinity, atomic structure, intriguing rectangular shape, and formation mechanism remain to be answered. Here, we report on a detailed investigation of the PbS NSs prepared with a lead thiocyanate single source precursor. Atomically resolved HAADF-STEM imaging reveals the presence of defects and small cubic domains in the deformed orthorhombic PbS crystal lattice. Moreover, variations in thickness are observed in the NSs, but only in steps of 2 PbS monolayers. To study the reaction mechanism, a synthesis at a lower temperature allowed for the study of reaction intermediates. Specifically, we studied the evolution of pseudo-crystalline templates toward mature, crystalline PbS NSs. We propose a self-induced templating mechanism based on an oleylamine-lead-thiocyanate (OLAM-Pb-SCN) complex with two Pb-SCN units as a building block; the interactions between the long-chain ligands regulate the crystal structure and possibly the lateral dimensions.

Semiconductor superstructures made from assembled and epitaxially connected colloidal nanocrystals (NCs) hold promise for crystalline solids with atomic and nanoscale periodicity, whereby the band structure can be tuned by the geometry. The formation of especially the honeycomb superstructure on a liquid substrate is far from understood and suffers from weak replicability. Here, we introduce 1,4-butanediol as an unreactive substrate component, which is mixed with reactive ethylene glycol to tune for optimal reactivity. It shows us that the honeycomb superlattice has a NC precursor state before oriented attachment occurs, in the form of a self-assembled hexagonal bilayer. We propose that the difference between the formation of the square or honeycomb superstructure occurs during the self-assembly phase. To form a honeycomb superstructure, it is crucial to stabilize the hexagonal bilayer in the presence of solvent-mediated repulsion. In contrast, a square superstructure benefits from the contraction of a hexagonal monolayer due to the absence of a solvent. A second experiment shows the very last stage of the process, where the increasing alignment of NCs is quantified using selected-area electron diffraction (SAED). The combination of transmission electron microscopy (TEM), SAED, and tomography used in these experiments shows that the (100)/(100) facet-to-facet attraction is the main driving force for NC alignment and attachment. These findings are validated by coarse-grained molecular dynamic simulations, where we show that an optimal NC repulsion is crucial to create the honeycomb superstructure.