Soft Colloidal Robots: Magnetically Guided Liquid Crystal Torons for Targeted Micro-Cargo Delivery

TL;DR

This work shows that topologically protected torons in cholesteric liquid crystals can be induced, propelled, and magnetically steered within confined cells, enabling targeted microcargo transport in microfluidic environments without net LC flow. By combining amplitude-modulated AC electric fields with in-plane magnetic fields, torons act as programmable, cargo-carrying quasiparticles that move as solitonic waves whose speed and size respond to field strength and confinement. The study couples experiments with Frank-Oseen-based simulations to reveal how magnetic alignment reshapes toron structure, velocity, and stability, and demonstrates their use in funnel-like microfluidic geometries that reorganize swarms into orderly jets. Overall, torons offer a robust, uniform, soft-robotic platform for adaptive delivery and exploration of active topological matter, with significant implications for soft microrobotics and lab-on-chip technologies.

Abstract

Quasiparticles in liquid crystals, such as torons and skyrmions, represent a new class of topologically protected solitonic excitations, offering a promising route toward soft microrobotics. Here we demonstrate that torons can be propelled by modulated electric fields and magnetically steered with full directional control, thus achieving programmable trajectories without net liquid flow. Within microfluidic architectures, we guide ensembles of torons through confined channels and realize targeted pick-up, transport, and release of colloidal cargo. By combining experiments and numerical simulations, we uncover how magnetic alignment reshapes toron structure, speed, and stability, while confinement within microchannels gives rise to novel transport regimes. Unlike conventional colloidal inclusions, torons are intrinsically uniform, soft, and reconfigurable, establishing them as both an ideal model system for studying emergent phenomena in active topological matter and a versatile platform for next-generation soft robots, adaptive delivery systems, and smart active matter.

Figures (5)

• Figure 1: (A) Sketch of the experimental setup. The liquid crystal fills the narrow gap between two ITO-coated glass plates, across which we apply an amplitude-modulated AC electric field. Torons of size comparable to the cell gap are formed within the liquid crystal layer. A magnetic field ($\bm{B}$) is applied parallel to the cell. (B) 3D structure of the director field, $\mathbf{n}$, within torons, in the absence of external fields, in the three orthogonal mid-planes. (C) Same as (B), but in the presence of an electric field between the plates. The vector $\vec{c}$ is the projection of $\mathbf{n}$ in the XY mid plane and away from the distortions. The configurations in (B) and (C) are obtained numerically after minimizing the free energy (see text).
• Figure 2: (A) CLC cell ($d/P = 0.79$, $P = 23.5\,\mu$m) with torons being propelled by the action of an AC electric field with AM ($V_p=12$ V, $f_c=1$ kHz, $f_m=20$ Hz). The direction of motion (specified by the velocity vector) is selected through the application of an in-plane magnetic field ($B=400$ mT), as indicated in each panel. Observation between crossed polarizers. See also Movie \ref{['SMov:skyrmion_H_exp']}A. (B) Effect of magnetic field intensity on toron size. Field and velocity directions are shown in the top panel. Observation between crossed polarizers. See also Movie \ref{['SMov:skyrmion_H_exp']}B. Speed (C) and size (D) of torons as a function of the magnetic field intensity, for the same AC driving. The red semi-transparent circles in panel B illustrate the area measurements shown in panel D, which were determined as described in Section 2.4.
• Figure 3: (A-B) Simulations of a toron being driven by an amplitude-modulated DC electric field and oriented with a magnetic field. Color map corresponds to the out-of-plane component of the local director field (see also Movie \ref{['SMov:skyrmion_H_sim']}). The orientation of $\bm{B}$ is changed 90 degrees clockwise between panels A and B. (C) Toron speed as a function of the magnetic field amplitude. In the inset, and for the same data, the toron speed is plotted vs the area (computed with the condition n$_{z}$$>0$) of its intersection with the XY plane, obtained with the condition $n_z > 0$. Periodic boundary conditions are considered in both directions in a system of size $107\,\mu$m $\times$$107\,\mu$m. Cell parameters are $d/P = 0.79$, $P = 23.5\,\mu$m. Electric field parameters are $V_p=3.97$ V, $f_m=20$ Hz. A magnetic field of 0.19 T is applied as indicated in A and B.
• Figure 4: Controlled transport of a passive polystyrene particle with planar LC anchoring on its surface by a toron ($d/P$=0.77, $p$=23.5 $\mu$m) driven by the action of an AM AC electric field ($V_p$=12.7 V, $f_m=20$ Hz $f_c=1$ kHz) and steered with an in-plane permanent magnetic field ($B = 0.4$ mT). (A) POM images of the particle transport. The red circle marks the location of the particle. The green dot marks the location of torons in the field of view. See also Movie \ref{['SMov:transport']}. Elapsed times are indicated. (B) Instantaneous speed of the toron that is loaded at 350 s of elapsed time. Error bars refer to the confidence interval of each measurement.
• Figure 5: (A) Ensemble of torons driven through a microfluidic funnel into a microchannel (see also Movie \ref{['SMov:funnel']}). Cell parameters are $d/P=0.7$, $P=11.6\,\mu$m. (B) Average speed profile along the traveling direction for different $f_m$ ($V_p$ =14 V, $f_c$ = 2 kHz). (C) Average speed profile along the traveling directions for different $V_p$ ($f_m$ = 20 Hz, $f_c$ = 2 kHz). All speeds are normalized with the average value outside the microfluidic device. Error bars refer to the confidence interval of each measurement. (D) Simulations of torons driven along a double-funnel microfluidic device for different angular openings, $\theta$ (see also Movie \ref{['SMov:sim_funnel']}). The length of the straight part decreases with $\theta$. System size is $251\,\mu$m $\times$$168\,\mu$m, with strong boundary conditions on the walls, and periodic boundary conditions in the horizontal direction. Cell parameters are $d/P=0.75$, $P=16.75\,\mu$m. AC modulation conditions are $V_p$ =3.75 V, $f_m$ = 20 Hz. A magnetic field of 0.12 T is applied in the Y direction. (E) Toron speed vs. channel width from simulations within a uniform channel of width $W$. In the inset, speed vs. toron area for the same simulations. Same conditions as in (D).