Scalable solution-phase epitaxial growth of symmetry-mismatched heterostructures on two-dimensional crystal soft template.

Epitaxial heterostructures with precisely controlled composition and electronic modulation are of central importance for electronics, optoelectronics, thermoelectrics, and catalysis. In general, epitaxial material growth requires identical or nearly identical crystal structures with small misfit in lattice symmetry and parameters and is typically achieved by vapor-phase depositions in vacuum. We report a scalable solution-phase growth of symmetry-mismatched PbSe/BiSe epitaxial heterostructures by using two-dimensional (2D) BiSe nanoplates as soft templates. The dangling bond-free surface of 2D BiSe nanoplates guides the growth of PbSe crystal without requiring a one-to-one match in the atomic structure, which exerts minimal restriction on the epitaxial layer. With a layered structure and weak van der Waals interlayer interaction, the interface layer in the 2D BiSe nanoplates can deform to accommodate incoming layer, thus functioning as a soft template for symmetry-mismatched epitaxial growth of cubic PbSe crystal on rhombohedral BiSe nanoplates. We show that a solution chemistry approach can be readily used for the synthesis of gram-scale PbSe/BiSe epitaxial heterostructures, in which the square PbSe (001) layer forms on the trigonal/hexagonal (0001) plane of BiSe nanoplates. We further show that the resulted PbSe/BiSe heterostructures can be readily processed into bulk pellet with considerably suppressed thermal conductivity (0.30 W/m·K at room temperature) while retaining respectable electrical conductivity, together delivering a thermoelectric figure of merit three times higher than that of the pristine BiSe nanoplates at 575 K. Our study demonstrates a unique epitaxy mode enabled by the 2D nanocrystal soft template via an affordable and scalable solution chemistry approach. It opens up new opportunities for the creation of diverse epitaxial heterostructures with highly disparate structures and functions.