Critical differences in 3D atomic structure of individual ligand-protected nanocrystals in solution

Abstract

Precise three-dimensional (3D) atomic structure determination of individual nanocrystals is a prerequisite for understanding and predicting their physical properties. Nanocrystals from the same synthesis batch display what are often presumed to be small but possibly important differences in size, lattice distortions, and defects, which can only be understood by structural characterization with high spatial 3D resolution. We solved the structures of individual colloidal platinum nanocrystals by developing atomic-resolution 3D liquid-cell electron microscopy to reveal critical intrinsic heterogeneity of ligand-protected platinum nanocrystals in solution, including structural degeneracies, lattice parameter deviations, internal defects, and strain. These differences in structure lead to substantial contributions to free energies, consequential enough that they must be considered in any discussion of fundamental nanocrystal properties or applications.

Chemical and bonding analysis of liquids using liquid cell electron microscopy

“…[12] One of the most effective methods to side-step this problem is to encapsulate the liquid between sheets of 2D materials, [18][19][20] simultaneously reducing the window and liquid thicknesses while also allowing the material of interest to comprise a much higher volumetric fraction of the liquid cell. The advent of graphene and boron-nitride liquid cells (GLCs an BNLCs respectively) has resulted in an outpouring of atomistic structural analysis of dynamic processes in colloidal nanoparticles, [16,[21][22][23][24][25] the compositional analysis of hydrated biological specimens and soft matter outside of cryogenic conditions, [26][27][28][29][30][31][32] and the direct analysis of liquids and their various phases. [15,30,[33][34][35] Perhaps the most critical advantage of the nanoscopic volume of supporting material in the 2D layer liquid cells is the opportunity to use low-dose techniques to mitigate and control beam damage and radiolysis in liquid cells.…”

“…[21] Recently, utilizing modern DEDs and improvements in reconstruction algorithms allowed the team to achieve true atomic resolution imaging in 3D to determine the coordinates of 9 Pt nanoparticles with approximately 2 nm diameters. [22] A nearly perfect NP and a NP with a defect are shown in Figure 4C…”

AutoDetect-mNP: An Unsupervised Machine Learning Algorithm for Automated Analysis of Transmission Electron Microscope Images of Metal Nanoparticles

“…[11][12][13][14][15][16][17][18] To date, transmission electron microscopy (TEM) remains the most reliable and widely used method for characterizing the morphology of NPs, for which rapid advancement in automated highthroughput electron microscopy has drastically increased both the acquisition rate and the quality of TEM data. [19][20][21][22][23] Increased efficiency in TEM data acquisition now enables the scale of NP shape characterization to increase from tens or hundreds of particles to orders of magnitude more, extracting information at the level of more informative statistical distributions. The information from TEM data at such a scale far exceeds the capability of a human analyst, and hence the development of automated methods for TEM image analysis is imperative.…”

Determination of the 3D Atomic Structures of Nanoparticles

“…[ 21–25 ] These structural deviations are also sensitive to the local environment on the surface, such as the specific binding chemistry of ligands that passivate surfaces and interacting molecules in the solution or gas phase. [ 25–27 ] Such change in ligands often leads to an increase or decrease in lattice parameters. [ 27 ] In addition, the high surface energy of nanoparticles is strongly correlated with the introduction of defects, including vacancies, stacking faults, and dislocations.…”