Imaging Orientation of a Single Molecular Hierarchical Self-Assembled Sheet: The Combined Power of a Vibrational Sum Frequency Generation Microscopy and Neural Network

Abstract

In this work, we determined the tilt angles of molecular units in hierarchical self-assembled materials on a single-sheet level, which were not available previously. This was achieved by developing a fast linescanning vibrational sum frequency generation (VSFG) hyperspectral imaging technique in combination with neural network analysis. Rapid VSFG imaging enables polarization resolved images on a single sheet level to be measured within a short time period, circumventing technical challenges due to long term optical setup instability. The polarization resolved hyperspectral images were then used to extract the supramolecular tilt angle of a self-assembly through a set of spectra-tilt angle relationships which were solved through neural network techniques. This unique combination of both novel techniques offers a new pathway to resolve molecular level structural knowledge of self-assembled materials. Understanding these properties can further drive self-assembly design from a bottom-up approach for applications in biomimetic and drug delivery researches.

Revealing the Molecular Physics of Lattice Self-Assembly by Vibrational Hyperspectral Imaging

“…The image quality is also improved as the shape of the SDS@2β-CD sheets could be resolved. This is highlighted by comparing hyperspectral images of SDS@2β-CD sheets with optical images of the same area, which show high correspondence (Figure a–c) …”

“…The quantification of the tilt angle of the building blocks sheds light on how they pack within the self-assembled materials, providing a protocol for studying the structure–property correlations of other hierarchical structures developed from the SDS@2β-CD nanosheets. This method could be applied to other systems to quantify similar structural properties that influence the overall structure and function of self-aseemblies …”