Chemo-Hydrodynamic Transceivers

• Chemo-hydrodynamic transceivers are systems that convert chemical reactions into localized fluid disturbances for propulsion, sensing, and communication.
• They employ designs such as catalytic MXene sheets, Janus particles, and microfluidic gating to modulate and decode hydrodynamic signals.
• These systems enable secure underwater communication and biomimetic sensing by harnessing solutal buoyancy, viscous shear, and controlled reaction flows.

A chemo-hydrodynamic transceiver is a system in which chemical processes generate hydrodynamic fields and those hydrodynamic fields are subsequently decoded into motion, sensing, or communication. In the cited literature, the expression is used for catalytic active matter, receptor-bearing biological appendages, and microfluidic molecular-communication architectures, and it is closely related to hydrodynamic transceivers that operate on localized pressure fields rather than on long-range acoustics (Wang et al., 19 Jun 2026, Zhang et al., 11 Feb 2026, Hood et al., 2018, Bolhassan et al., 2023, Chen, 26 Dec 2025). Across these works, the defining feature is not merely the coexistence of chemistry and flow, but a transduction chain in which chemical or mechanically induced nonequilibrium structure in the fluid acts as the transmitted signal and a downstream hydrodynamic response acts as the received quantity.

1. Definitional core and scope

In the most literal usage, the active MXene sheet converts a chemical input—the catalytic decomposition of hydrogen peroxide by asymmetrically placed catalase—into a hydrodynamic signal in the surrounding fluid, and then converts that fluid signal into directed mechanical motion of the sheet itself (Wang et al., 19 Jun 2026). In the catalytic Janus-particle model for the Internet of Bio-Nano Things, a single external command signal simultaneously determines both molecular emission and active self-propulsion, so the particle is not treated as a separate “transmitter” and “mobile node,” but as a single chemo-hydrodynamic transceiver (Zhang et al., 11 Feb 2026). In marine crustaceans, hairy appendages are described as chemo-hydrodynamic transceivers because the hairs both shape local flow and sample the fluid’s chemical content (Hood et al., 2018). In microfluidic molecular communication, hydrodynamic gating is presented as a transceiver architecture candidate because controlled flow switching generates information-bearing molecular pulses in a platform that can eventually support matched emission and sensing (Bolhassan et al., 2023). A related but chemically simpler limit is the underwater “Hydrodynamic Whispering” system, where a vibrating source and an Artificial Lateral Line array realize a hydrodynamic transceiver through localized pressure fluctuations (Chen, 26 Dec 2025).

Context	Hydrodynamic carrier	Output function
Active MXene sheet	Buoyancy-driven chemo-hydrodynamic wave	Directed mechanical motion
Catalytic Janus particle	Phoretic slip and stochastic active displacement	Molecular emission and propulsion
Microfluidic gating device	Pressure-driven laminar gating flow	Time-shaped molecular pulses
Artificial lateral line link	Localized dipole pressure field	Demodulated bits
Crustacean hair bed	Boundary-layer-mediated transport	Sensing or feeding mode

This comparison suggests that the term denotes a class of coupled physicochemical systems rather than a single device architecture. The commonality is the conversion of a chemically or mechanically generated fluid disturbance into an actionable downstream state variable.

2. Governing transport, flow, and transduction laws

For reaction-driven active matter, the core description is a coupled transport–flow problem. In the MXene-sheet simulations, species transport is governed by

$$\frac{\partial c_i}{\partial t} + \nabla \cdot \mathbf{J}_i + \mathbf{u} \cdot \nabla c_i = R_i,$$

with diffusive flux

$\mathbf{J}_i = -D_i \nabla c_i,$

and fluid motion is governed by the Navier–Stokes equation with incompressibility and buoyancy forcing (Wang et al., 19 Jun 2026). The reaction term follows Michaelis–Menten kinetics for catalase. The reported dimensionless estimates, $Re \approx 0.15$, $Pe \approx 150$, and $Ra \approx 9 \times 10^4$, place that system in a low-Reynolds-number but strongly buoyancy-driven regime where convection is important (Wang et al., 19 Jun 2026).

In the IoBNT Janus-particle model, the same control input enters both the propulsion law and the emission law. The paper derives

$U(t)=K_{\text{control}}\, I_{\text{ext}}(t),$

and

$q(t)=\kappa_{\text{em}}\, I_{\text{ext}}(t),$

so control simultaneously sets swimming speed and molecular release (Zhang et al., 11 Feb 2026). The mobility model is planar active Brownian dynamics, and the sampled receiver output is linearized in the far field as a Gaussian observation whose variance contains both passive and active control-dependent terms. This is the mathematical basis of the propulsion–transmission trade-off.

In the hydrodynamic communication setting, the transmitter is modeled by potential flow theory as an oscillating dipole source. With a rigid sphere oscillating as $x(t)=A\sin(\omega t)$, the pressure field follows

$$p(\mathbf{r}, t) = \frac{\rho a^3}{2r^2}\cos\theta \cdot \dot{U}(t),$$

so the amplitude scales as $p \propto 1/r^2$ (Chen, 26 Dec 2025). The steep near-field decay is central because it defines a localized communication bubble and underlies the Low Probability of Interception characterization.

In the microfluidic molecular-communication architecture, ligand transport is modeled by a 2D convection-diffusion equation,

$\mathbf{J}_i = -D_i \nabla c_i,$0

with $\mathbf{J}_i = -D_i \nabla c_i,$1 because chemical reactions are neglected (Bolhassan et al., 2023). The analytical approximation then reduces the waveform to three descriptors—pulse amplitude, pulse width, and pulse delay—after decomposing the problem into pulse generation and pulse propagation compartments.

In crustacean hair arrays, the control parameter is the Reynolds number. For the experimental channel system,

$\mathbf{J}_i = -D_i \nabla c_i,$2

and for shear-flow comparisons,

$\mathbf{J}_i = -D_i \nabla c_i,$3

with the single-hair boundary-layer depth used as the predictor of array-scale flow phase (Hood et al., 2018).

3. Propulsion-centered realizations

The MXene-sheet realization is a freestanding, thin Ti$\mathbf{J}_i = -D_i \nabla c_i,$4C$\mathbf{J}_i = -D_i \nabla c_i,$5T$\mathbf{J}_i = -D_i \nabla c_i,$6 MXene sheet fabricated by vacuum filtration, with one half selectively modified with catalase by a mask-guided deposition process while the other half remains bare (Wang et al., 19 Jun 2026). The sheet is placed at the air–liquid interface of an aqueous H$\mathbf{J}_i = -D_i \nabla c_i,$7O$\mathbf{J}_i = -D_i \nabla c_i,$8 solution in a square chamber, with buoyancy keeping it afloat. Because the enzyme is distributed asymmetrically, the catalase-coated side decomposes peroxide faster, creating a local concentration deficit and density difference. This asymmetry is crucial, and a fully coated sheet was shown to be almost motionless (Wang et al., 19 Jun 2026).

The experiments use synchronized dual-view particle image velocimetry. One camera records the top view and tracks the centroid of the moving sheet, while the second camera records the side view and resolves the local flow field near the interface. The side-view PIV reveals a pair of counter-rotating vortices, whereas the top view shows nearly rectilinear translation. The motion is described as quasi-one-dimensional because the experiment occurs in a 3D fluid but the asymmetric enzyme pattern constrains the sheet to move predominantly along one horizontal direction, with only slight curvature due to finite-width hydrodynamics and fabrication imperfections (Wang et al., 19 Jun 2026).

The paper distinguishes this locomotion from other propulsion modes in two ways. First, unlike bubble propulsion, there is no observed bubble ejection event to account for thrust, and no significant bubble detachment or rupture was observed in the side-view recordings. Second, unlike Marangoni propulsion, the surface-tension contribution is negligible because the measured change in surface tension with H$\mathbf{J}_i = -D_i \nabla c_i,$9O$Re \approx 0.15$0 concentration is negligible. The mechanism is therefore identified as solutal buoyancy, with reaction-generated density gradients producing convective flow rolls beneath and beside the sheet (Wang et al., 19 Jun 2026).

A central result is that the flow behaves like a propagating chemo-hydrodynamic wave. The wave is most clearly seen in the magnitude of the horizontal velocity component, $Re \approx 0.15$1, measured along a horizontal plane near the lower boundary of the sheet. In the direct-wave picture, the wave and the sheet move in the same direction: the peak of the fluid signal and the centroid of the sheet both propagate leftward, from the enzyme-coated side toward the bare side. The motion is periodically modulated rather than steady, and Fourier analysis shows a dominant low-frequency component (Wang et al., 19 Jun 2026).

The force analysis makes the transceiver formulation explicit. Because the sheet floats symmetrically at the interface and is very thin, horizontal pressure contributions cancel to leading order. The net driving force is attributed to viscous shear stress from the liquid side:

$Re \approx 0.15$2

with the boundary-layer approximation retaining the vertical gradient of the horizontal flow as the key contribution. The total driving force is obtained by integrating the shear stress over the wetted sheet area. The resulting force is time-dependent and tracks the chemo-hydrodynamic wave, which supports the interpretation that the wave is the transmitted signal carrying the mechanical drive (Wang et al., 19 Jun 2026).

At the nanoscale, the Janus-particle formulation generalizes the same logic. An external optical control drives a catalytic cap reaction $Re \approx 0.15$3, where species $Re \approx 0.15$4 is the information molecule, and the same control determines the phoretic slip flow and self-propulsion (Zhang et al., 11 Feb 2026). The mean received signal grows with actuation, but the observation variance contains both a passive $Re \approx 0.15$5 term and an active $Re \approx 0.15$6 term:

$Re \approx 0.15$7

The paper therefore reports a unimodal reliability profile with a critical actuation level beyond which the signal-to-noise ratio collapses, an analytically characterizable optimal control level, and an optimal control that scales approximately linearly with link distance (Zhang et al., 11 Feb 2026). Compared with Brownian-mobility baselines, neglecting active motility noise can underestimate the bit error probability by orders of magnitude.

4. Communication-oriented architectures

“Hydrodynamic Whispering” defines a near-field silent communication paradigm in which the transmitter is a vibrating spherical source and the receiver is a bio-inspired Artificial Lateral Line array mounted on the receiving AUV hull (Chen, 26 Dec 2025). The transmitter converts digital information into controlled mechanical oscillations of the fluid, and the receiver senses the resulting pressure field, suppresses turbulence noise through spatial processing, and recovers the bits via coherent demodulation. The communication channel is deliberately local: because the pressure amplitude scales as $Re \approx 0.15$8, the useful range is only about 10 body lengths, or less than 1 m, which is the basis of its secure “communication bubble” and Low Probability of Interception properties (Chen, 26 Dec 2025).

The modulation scheme is Binary Phase Shift Keying adapted for mechanical actuator inertia. The paper defines the Cycle-per-Symbol Ratio

$Re \approx 0.15$9

For the successful case $Pe \approx 150$0 and $Pe \approx 150$1, $Pe \approx 150$2, meaning each symbol spans two carrier cycles; at $Pe \approx 150$3, $Pe \approx 150$4, and demodulation fails because the actuator cannot complete even one oscillation per bit period (Chen, 26 Dec 2025). The receiver uses a 24-sensor conformal dual-line array and a two-stage processing chain: Spatial Matched-Field Beamforming followed by Temporal Coherent Demodulation. Because the beamforming weights are matched to the dipole field, the array produces an array gain of approximately 13.8 dB, and the paper states that the system achieves BER $Pe \approx 150$5 at SNR $Pe \approx 150$6 after spatial processing within the effective range (Chen, 26 Dec 2025).

The microfluidic molecular-communication architecture realizes the chemical side of the same transceiver logic. It uses a cross-shaped PDMS microfluidic channel fabricated with soft lithography and operated with pressure-driven flows, with four main branches: the supply inlet/channel $Pe \approx 150$7, the gating inlet/channel $Pe \approx 150$8, the gating outlet/channel $Pe \approx 150$9, and the propagation channel $Ra \approx 9 \times 10^4$0 (Bolhassan et al., 2023). When the gating flow is ON, the vertical flow diverts the supply stream to the gating outlet and prevents ligands from entering the propagation channel. When the gating flow is OFF for a short duration $Ra \approx 9 \times 10^4$1, molecules from the supply inlet briefly enter the propagation channel, after which a short molecular pulse convects and diffuses downstream.

The analytical approximation reduces the waveform to Gaussian pulse descriptors. The generated pulse amplitude is assumed to satisfy $Ra \approx 9 \times 10^4$2, the generated width is

$Ra \approx 9 \times 10^4$3

and at the sampling point the propagated pulse is parameterized by amplitude $Ra \approx 9 \times 10^4$4, width $Ra \approx 9 \times 10^4$5, and delay $Ra \approx 9 \times 10^4$6, with

$Ra \approx 9 \times 10^4$7

The fitting constants were optimized using a genetic algorithm in MATLAB over 30 scenarios, yielding $Ra \approx 9 \times 10^4$8, $Ra \approx 9 \times 10^4$9, $U(t)=K_{\text{control}}\, I_{\text{ext}}(t),$0, and $U(t)=K_{\text{control}}\, I_{\text{ext}}(t),$1 (Bolhassan et al., 2023). COMSOL comparisons show that the analytical curves closely follow the simulation trends for pulse amplitude, width, and delay, while successive-pulse simulations demonstrate increasing overlap and intersymbol interference with propagation distance.

These two communication architectures emphasize different limits of transceiver design. One relies on localized hydrodynamic pressure fluctuations for reliable and secure short-range underwater networking; the other uses hydrodynamic gating to generate time-shaped molecular concentration pulses for molecular communication. Taken together, they show that hydrodynamic processing can be used either to carry the message directly or to temporally sculpt the chemical message before downstream detection.

5. Biological antecedents and biomimetic interpretation

Marine crustaceans provide a biological realization of chemo-hydrodynamic transduction. Their appendages are covered with rigid, receptor-bearing hairs, and the appendage is described as a chemo-hydrodynamic transceiver because the hairs both shape the local flow and sample the fluid’s chemical content (Hood et al., 2018). As the appendage moves through water, each hair creates a no-slip boundary layer; the collective behavior of these boundary layers determines whether fluid is largely excluded from the hair bed, passes through it, or enters and then escapes sideways.

The paper identifies three flow phases. In rake, boundary layers overlap strongly, the gap between hairs is effectively blocked, fluid inside the bed is nearly stagnant, and streamlines bend around the array. In deflection, fluid penetrates the hair bed but is deflected laterally and exits before reaching the downstream end; a recirculation or wake region appears downstream. In sieve, boundary layers do not overlap, fluid passes through the gaps between hairs, and flow inside the bed is comparable in speed to the external flow (Hood et al., 2018). The deflection regime is explicitly identified as a third transitional phase rather than a mere interpolation between blocked and fully permeable behavior.

The reduced-order model predicts phase from the boundary-layer depth on a single hair. Using the level set

$U(t)=K_{\text{control}}\, I_{\text{ext}}(t),$2

with $U(t)=K_{\text{control}}\, I_{\text{ext}}(t),$3, the critical boundary-layer radius $U(t)=K_{\text{control}}\, I_{\text{ext}}(t),$4 is measured from the single-hair contour and compared with the center-to-center spacing $U(t)=K_{\text{control}}\, I_{\text{ext}}(t),$5. The phase rule is:

• $U(t)=K_{\text{control}}\, I_{\text{ext}}(t),$6 $U(t)=K_{\text{control}}\, I_{\text{ext}}(t),$7 rake
• $U(t)=K_{\text{control}}\, I_{\text{ext}}(t),$8 $U(t)=K_{\text{control}}\, I_{\text{ext}}(t),$9 sieve
• $q(t)=\kappa_{\text{em}}\, I_{\text{ext}}(t),$0 $q(t)=\kappa_{\text{em}}\, I_{\text{ext}}(t),$1 deflection (Hood et al., 2018)

The experiments used a rectangular channel with rigid steel rods mounted as a $q(t)=\kappa_{\text{em}}\, I_{\text{ext}}(t),$2 hair bed, with hair diameter $q(t)=\kappa_{\text{em}}\, I_{\text{ext}}(t),$3 mm, length $q(t)=\kappa_{\text{em}}\, I_{\text{ext}}(t),$4 mm, spacing $q(t)=\kappa_{\text{em}}\, I_{\text{ext}}(t),$5 varied from 2 to 10 mm, and Reynolds number varied over roughly $q(t)=\kappa_{\text{em}}\, I_{\text{ext}}(t),$6 to $q(t)=\kappa_{\text{em}}\, I_{\text{ext}}(t),$7 (Hood et al., 2018). Flow was visualized with tracer particles and particle image velocimetry, and the reduced-order criterion matched the observed phase diagram. The same framework also agrees with measurements of chemo-sensing and suspension-feeding crustaceans: aesthetasc arrays of stomatopods and lobsters lie in the deflection/sieve region, whereas suspension-feeding appendages tend to lie in the rake region (Hood et al., 2018).

This biological case is significant because the transceiver function does not require explicit digital modulation or engineered catalysts. Geometry, motion speed, and boundary-layer overlap already implement a form of flow-mediated encoding and decoding that determines whether odor-bearing water is retained, exchanged, or redirected.

6. Design principles, misconceptions, and broader significance

Several recurrent misconceptions are explicitly addressed in the literature. In the MXene-sheet system, the motion is not simply a classical Marangoni swimmer and not a bubble-driven motor; the reported evidence instead supports a direct chemo-hydrodynamic wave-driven mode based on solutal buoyancy and viscous shear traction (Wang et al., 19 Jun 2026). In hydrodynamic whispering, the link does not rely on long-range propagating sound waves, but on localized hydrodynamic pressure patterns that vanish rapidly outside a limited bubble around the transmitter (Chen, 26 Dec 2025). In IoBNT modeling, the common reservoir abstraction and the mobility decoupling assumption miss the fact that stronger actuation can increase both emission and propulsion noise, so conventional decoupled models can be dangerously optimistic (Zhang et al., 11 Feb 2026). In crustacean arrays, function is not determined by hair geometry alone; it is determined by the relation between geometry and boundary-layer thickness at the relevant Reynolds number (Hood et al., 2018).

The cited works also provide concrete design rules. For hydrodynamically driven active motion, the MXene study proposes engineering the spatial layout of catalytic regions so that the reaction generates a directed, propagating fluid signal whose shear traction is aligned with the desired motion; the authors suggest tuning enzyme type, catalytic area, fuel concentration, and fluid depth, and they note that multiple actuators could potentially be combined for collective behavior (Wang et al., 19 Jun 2026). For hydrodynamic communication, matched-field weighting is essential because the dipole field contains positive and negative lobes that would cancel under naive summation; the 24-sensor conformal array therefore acts as a spatial front end for coherent recovery (Chen, 26 Dec 2025). For microfluidic molecular communication, the model indicates that shorter gating durations generate narrower pulses, higher flow velocity reduces delay, $q(t)=\kappa_{\text{em}}\, I_{\text{ext}}(t),$8 should be tuned for sharp gating transitions, and pulse spacing must exceed the effective broadened width to reduce intersymbol interference (Bolhassan et al., 2023). For active Janus transceivers, the recommended policy is to keep actuation within a safe envelope, use distance-aware control, prefer shorter symbol durations when possible, and account for medium viscosity because active-noise effects can worsen in viscous environments (Zhang et al., 11 Feb 2026). For biological or biomimetic hair beds, tuning hair length, spacing, and motion speed selects among rake, deflection, and sieve transport modes (Hood et al., 2018).

Taken together, these results define chemo-hydrodynamic transceivers as coupled propulsion–communication–sensing systems in which chemistry encodes a signal, hydrodynamics transmits it, and mechanics or receptor geometry decodes it into action. The concept spans active sheets, catalytic colloids, microfluidic emitters, and biological appendages, but the central principle remains the same: the relevant signal is not only chemical concentration or only flow, but the structured coupling between them.