Sensory Delay as a Feature: Janus Microswimmers That Adapt to Light and Chemistry

Bacteria do not flip their flagella the instant they detect a chemical signal. They integrate information over a finite window of time, comparing concentrations past and present before deciding to keep swimming straight or tumble. This memory, this built-in lag, is essential to the chemotactic strategies that let E. coli climb concentration gradients toward nutrients. Synthetic microswimmers have largely ignored this feature. A new paper from Töpfer and collaborators, published in ACS Nano in 2025, takes a step toward closing that gap, and finds that the delay is not a bug. It is the mechanism.

The team worked with Janus particles, silica spheres coated on one hemisphere with titanium dioxide (TiO₂). In a uniform AC electric field, these particles self-propel through induced-charge electrophoresis: the asymmetric conductivity of the two hemispheres generates an asymmetric electroosmotic flow, and the particle moves. That part is established physics. What is new is the second control channel. TiO₂ is photoconductive: ultraviolet illumination increases its conductivity, and that change alters the propulsion speed, not by adjusting the electric field, but orthogonally to it. The two knobs (electric field amplitude and UV intensity) act independently. Tuning one does not disturb the other.

The Physics of the Delay

When UV light hits the TiO₂ hemisphere, free charge carriers multiply, conductivity climbs, and the propulsion velocity shifts. But this does not happen instantly. The photoconductivity response has a finite rise time governed by how quickly carriers are generated and recombined in TiO₂. The particle's speed at any moment reflects its recent exposure history, not just its current illumination. This is the sensory delay.

The counterintuitive result: particles with a sensory delay localize more sharply in spatiotemporal light patterns than they would with instantaneous response. An instantaneous system reacts only to the local present intensity. A system with memory accumulates a weighted average of recent intensities, and in a light field that varies in time, this lag phase-shifts the particle's speed response relative to its position, building density modulations that a purely instantaneous response would not produce.

The group also showed that methanol, when added to the solution, alters how long charge carriers survive in TiO₂ after the UV source is removed. This shifts the response time. The particle's behavior changes not because of any external actuation, but because a dissolved chemical has modified its internal dynamics, an elementary form of chemical sensing.

Why This Matters

Most active colloid research controls particle behavior from the outside: a field is applied, and particles respond. This paper's particles adjust internally, translating an environmental signal (light, chemistry) into a change in self-propulsion, and doing so with a time constant that can itself be tuned by what is in solution. The two-channel control (field for baseline propulsion, light for modulation, chemistry for response time) opens a design space that earlier platforms did not have.

For the broader active matter field, the sensory delay result carries an immediate implication: the assumption of instantaneous response used in most theoretical models is not just a simplification, it changes the predicted behavior. Models that include a finite response time describe the experimental density patterns more accurately.

Open Questions

The current particles cannot yet steer independently. The speed changes, but the direction is still randomized by rotational diffusion. Combining orientation control, through anisotropic shape, magnetic torque, or chemical feedback on the motor hemisphere, with the adaptive speed mechanism would bring these particles closer to the navigation capability of real microorganisms. Whether the density modulation observed here can be channeled into directional accumulation, a synthetic phototaxis, is the next logical question.

In the Broader Landscape

This work fits a shift underway in the active colloid field: from particles that simply move to particles that respond. The Péclet number alone no longer characterizes what an active particle does. The relevant parameter space now includes the ratio of the particle's response time to the timescale of environmental variation, an analog of the bacterial adaptation time that controls chemotaxis efficiency. Designing that ratio deliberately, rather than accepting whatever the material provides, is where the field is heading.Töpfer, U., Bailey, M. R., Schreiber, S., Paratore, F., & Isa, L. (2025). Density Modulations in Active Colloidal Systems through Orthogonal Propulsion Control and Sensory Delays. ACS Nano, 19(45), 39210–39219. DOI: 10.1021/acsnano.5c12596