Intermolecular Organization of a Lyotropic Liquid Crystal and Carbon Dot Composite in Microfluidic Channels: Surface and Dynamic Effects

TL;DR

This paper studies how microfluidic channels affect the structure and properties of a nanocomposite material for biomedical applications.

Contribution

The study introduces a novel integration of lyotropic liquid crystal-carbon dot composites with microfluidics and AI for biomedical applications.

Findings

Microchannel wall structure influences the near-wall ordering of the nanocomposite's lamellar structure.

Controlled shear stress and heating-cooling cycles transform vesicles into anisotropic structures with tunable luminescence.

A neural network was trained to recognize nanocomposite organization using texture image datasets.

Abstract

Composites of lyotropic liquid crystals with biocompatible luminescent nanoparticles represent multifunctional materials with high potential for application in molecular diagnostics and biomedicine. Their integration with microfluidics is a new and scarcely studied approach that offers unique opportunities for tuning properties of such nanomaterials and simulating the biological environment of their application. This paper analyzes the impact of the governing microfluidic factors, including wall effects and flow dynamics, on the intermolecular structure and optical properties of the mesogenic luminescent nanocomposite of tetraethylene glycol monododecyl ether and carbon dots. The nanoscale and microscale surface structure of microchannel walls was found to be the dominating factor for additional near-wall ordering of the intrinsic lamellar structure of the composite. A combination of controlled shear stress with heating–cooling cycles allowed for a gradual and reversible transformation of multilamellar vesicles into the axial lamellar structure, and provided the composite with anisotropic luminescence capabilities according to the study of the luminescent behavior of carbon dots. The collected experimental datasets, comprising hundreds of texture images, allowed for training the neural network for subsequent accurate recognition of the composite nanoscale organization and dynamic properties in straight and serpentine microchannels. The results will contribute to developing AI-powered microfluidic chips with integrated biocompatible nanocomposite materials for testing drug delivery systems and simulating biological capillary environment in organ-on-chip prototypes.

Figures