Nanoscale Events on Cyanobiphenyl-Based Self-Assembled Droplets Triggered by Gas Analytes

Possibilities and limitations of convolutional neural network machine learning architectures in the characterisation of achiral orthogonal smectic liquid crystals

Machine learning is becoming a valuable tool in the characterisation and property prediction of liquid crystals. It is thus worthwhile to be aware of the possibilities but also the limitations of current machine learning algorithms. In this study we investigated a phase sequence of isotropic - fluid smecticA - hexatic smectic B - soft crystal CrE - crystalline. This is a sequence of transitions between orthogonal phases, which are expected to be difficult to distinguish, because of only minute changes in order. As expected, strong first order transitions such as the liquid to liquid crystal transition and the crystallisation can be distinguished with high accuracy. It is shown that also the hexatic SmB to soft crystal CrE transition is clearly characterised, which represents the transition from short- to long-range order. Limitations of convolutional neural networks can be observed for the fluid to hexatic SmA to SmB transition, where both phases exhibit short-range ordering.

Biosensing with Oleosin-Stabilized Liquid Crystal Droplets

Liquid crystals (LCs) are emerging as novel platforms for chemical, physical, and biological sensing. They can be used to detect biological amphiphiles such as lipids, fatty acids, digestive surfactants, and bacterial endotoxins. However, designing LC-based sensors in a manner that preserves their sensitivity and responsiveness to these stimuli, and possibly improves biocompatibility, remains challenging. In this work, the stabilization of LC droplets by oleosins, plant-sourced and highly surface active proteins due to their extended amphipathic helix, is investigated. Purified oleosins, at sub-micromolar concentrations, are shown to readily stabilize nematic LC droplets without switching their alignment, allowing them to detect surfactants at micromolar concentrations. Direct evidence of localization of oleosins at the LC-water interface is provided with fluorescent labeling, and the stabilized droplets remain stable over months. Interestingly, chiral LC droplets readily switch in the presence of nanomolar oleosin concentrations, an unexpected behavior that is explained by accounting for the energy barriers required for switching the alignment between the two cases. This leads thus to a twofold conclusion: oleosin-stabilized nematic LC droplets present a biocompatible alternative for bioanalyte detection, while chiral LCs can be further investigated for use as highly sensitive sensors for detecting amphipathic helices in biological systems.

Machine learning classification of polar sub-phases in liquid crystal MHPOBC

Experimental polarising microscopy texture images of the fluid smectic phases and sub-phases of the classic liquid crystal MHPOBC were classified as paraelectric (SmA*), ferroelectric (SmC*), ferrielectric (SmC1/3*), and antiferroelectric (SmCA*) using convolutional neural networks, CNNs. Two neural network architectures were tested, a sequential convolutional neural network with varying numbers of layers and a simplified inception model with varying number of inception blocks. Both models are successful in binary classifications between different phases as well as classification between all four phases. Optimised architectures for the multi-phase classification achieved accuracies of (84 ± 2)% and (93 ± 1)% for sequential convolutional and inception networks, respectively. The results of this study contribute to the understanding of how CNNs may be used in classifying liquid crystal phases. Especially the inception model is of sufficient accuracy to allow automated characterization of liquid crystal phase sequences and thus opens a path towards an additional method to determine the phases of novel liquid crystals for applications in electro-optics, photonics or sensors. The outlined procedure of supervised machine learning can be applied to practically all liquid crystal phases and materials, provided the infrastructure of training data and computational power is provided.

Heads or tails: investigating the effects of amphiphile features on the distortion of chiral nematic liquid crystal droplets

Liquid crystal-based sensing has fast become a growing field, harnessing the sensitivity of liquid crystals to their surroundings to provide information about the analytes present, including surface-active amphiphiles such as biological lipids. Amphiphiles can impart ordering to a liquid crystal and, in the case of chiral nematic liquid crystals (CLCs), distort the helical texture. The cause and degree to which this distortion occurs is not fully clear. In this work, the effects of different amphiphiles on the final colour textures as well as the pitch of chiral nematic liquid crystals are investigated. We find that the tails of amphiphiles and their orientation play a more important role in determining the final distortions of the liquid crystal by the direct interactions they have with the host, whereas the headgroups do not play a significant role in affecting these distortions. Our findings may find implications in designing CLC-based biosensors, where the tails will likely have more impact on the CLC response, while the headgroups will remain available for further functionalization without having significant effects on the signal readout.

Effect of Polymer Hydrophobicity in the Performance of Hybrid Gel Gas Sensors for E-Noses

Relative humidity (RH) is a common interferent in chemical gas sensors, influencing their baselines and sensitivity, which can limit the performance of e-nose systems. Tuning the composition of the sensing materials is a possible strategy to control the impact of RH in gas sensors. Hybrid gel materials used as gas sensors contain self-assembled droplets of ionic liquid and liquid crystal molecules encapsulated in a polymeric matrix. In this work, we assessed the effect of the matrix hydrophobic properties in the performance of hybrid gel materials for VOC sensing in humid conditions (50% RH). We used two different polymers, the hydrophobic PDMS and the hydrophilic bovine gelatin, as polymeric matrices in hybrid gel materials containing imidazolium-based ionic liquids, [BMIM][Cl] and [BMIM][DCA], and the thermotropic liquid crystal 5CB. Better accuracy of VOC prediction is obtained for the hybrid gels composed of a PDMS matrix combined with the [BMIM][Cl] ionic liquid, and the use of this hydrophobic matrix reduces the effect of humidity on the sensing performance when compared to the gelatin counterpart. VOCs interact with all the moieties of the hybrid gel multicomponent system; thus, VOC correct classification depends not only on the polymeric matrix used, but also on the IL selected, which seems to be key to achieve VOCs discrimination at 50% RH. Thus, hybrid gels' tunable formulation offers the potential for designing complementary sensors for e-nose systems operable under different RH conditions.

Where Nanosensors Meet Machine Learning: Prospects and Challenges in Detecting Disease X

Disease X is a hypothetical unknown disease that has the potential to cause an epidemic or pandemic outbreak in the future. Nanosensors are attractive portable devices that can swiftly screen disease biomarkers on site, reducing the reliance on laboratory-based analyses. However, conventional data analytics limit the progress of nanosensor research. In this Perspective, we highlight the integral role of machine learning (ML) algorithms in advancing nanosensing strategies toward Disease X detection. We first summarize recent progress in utilizing ML algorithms for the smart design and fabrication of custom nanosensor platforms as well as realizing rapid on-site prediction of infection statuses. Subsequently, we discuss promising prospects in further harnessing the potential of ML algorithms in other aspects of nanosensor development and biomarker detection.

Designing Biological Microsensors with Chiral Nematic Liquid Crystal Droplets

Biosensing using liquid crystals has a tremendous potential by coupling the high degree of sensitivity of their alignment to their surroundings with clear optical feedback. Many existing set-ups use birefringence of nematic liquid crystals, which severely limits straightforward and frugal implementation into a sensing platform due to the sophisticated optical set-ups required. In this work, we instead utilize chiral nematic liquid crystal microdroplets, which show strongly reflected structural color, as sensing platforms for surface active agents. We systematically quantify the optical response of closely related biological amphiphiles and find unique optical signatures for each species. We detect signatures across a wide range of concentrations (from micromolar to millimolar), with fast response times (from seconds to minutes). The striking optical response is a function of the adsorption of surfactants in a nonhomogeneous manner and the topology of the chiral nematic liquid crystal orientation at the interface requiring a scattering, multidomain structure. We show that the surface interactions, in particular, the surface packing density, to be a function of both headgroup and tail and thus unique to each surfactant species. We show lab-on-a-chip capability of our method by drying droplets in high-density two-dimensional arrays and simply hydrating the chip to detect dissolved analytes. Finally, we show proof-of-principle in vivo biosensing in the healthy as well as inflamed intestinal tracts of live zebrafish larvae, demonstrating CLC droplets show a clear optical response specifically when exposed to the gut environment rich in amphiphiles. Our unique approach shows clear potential in developing on-site detection platforms and detecting biological amphiphiles in living organisms.