betaFeedback

WorksPhysics and Astronomy

Dynamical-state switching controls active microrod navigation in porous media

Abstract

Adaptive search and travel strategies enable biological swimmers to explore complex environments, but remain difficult to encode in minimal synthetic microswimmers. Here, we unveil multiple dynamical and orientational states of rod-like Janus microswimmers actuated via electric fields: planar active, oriented chiral and persistent rods. We discover a hidden dynamical state after pre-orienting them, which allows us to switch on demand from strong chiral to persistent motion. We harness this phenomenological orientation and chirality switch for exploration in microstructured porous arrays. Here, strong chirality promotes local scanning but drives circular trapping and poor net displacement, whereas the pre-aligned dynamical state enables efficient obstacle sorting, and trap escape. Comparing experiments with simulations of active Brownian particles in the very same experimentally-extracted obstacle geometries, we quantify navigation efficiency via first-passage time statistics, revealing how chirality tuning optimizes the trade-off between local exploration and long-range transport.

Navigating through disordered and complex environments is a fundamental requirement for microorganisms. From sperm cells reaching an egg to immune cells patrolling tissues, and bacteria foraging for nutrients in heterogeneous soils or mucus [1, 2, 3, 4, 5], the ability to move effectively through disordered media is essential for survival. To succeed, these organisms employ adaptive strategies that alternate between persistent directed motion, reorientation events to escape entrapment, and chirality or tumbling to probe their local surroundings [6, 7, 8, 9, 10].

Designing motility at the microscale to mimic biological microswimmers is a current open challenge critical for realizing next-generation microscopic vehicles for cargo transport, in particular targeted drug delivery [11, 12, 13, 14, 15, 16, 17], or noninvasive microsurgeries [11, 15, 17, 18, 19, 20]. In this context, synthetic active colloids are widely used as a platform for engineering bioinspired microswimmers aiming at these applications [8, 11, 21, 22, 23]. Active colloids self-propel under diverse mechanisms [3]; however, unlike their biological counterparts, they typically lack an intrinsic ability to adapt their swimming strategy to environmental constraints. Moreover, their navigation in porous or complex media is often hindered by interactions with solid interfaces, leading to wall accumulation [24, 25, 26] or trapping in obstacles and blind pores due to steric and hydrodynamic effects [27, 28, 29]. At the same time, particle–interface interactions can be leveraged to guide motion and bias transport by designing obstacle arrangements or tuning particle–surface interactions [30, 31, 32].

Efficient navigation through porous and complex media, therefore, requires implementing refined motility strategies that include controlled reorientation mechanisms, which are indispensable for efficient dispersion in porous networks [2]. Inspired by biological navigation, current approaches to control dynamical patterns of active colloids use external fields to modulate propulsion and reorientation [33, 34, 35, 36], or exploit internal material properties and geometry/composition reconfiguration to tune persistence length [37] or mimic run-and-tumble behavior [13, 34, 38].

On-demand control of chiral motion offers a minimal dynamical knob to reshape turning events, enabling a direct trade-off between local exploration and long-range persistence in synthetic microswimmers [38]. In simulations of chiral active systems, obstacles can either enhance or hinder transport, with direct implications for controlling active motion in disordered media [39]. Moreover, the interplay between angular velocity and alignment interactions can determine whether convex obstacles suppress or enhance surface accumulation [40]. At larger (granular) scales, experiments with chiral active particles have shown that effective diffusivity becomes sensitive to the obstacle-lattice structure, a feature absent in achiral swimmers [41]. At the microscale, however, chirality control is often not deliberate: fabrication imperfections can induce unintended symmetry breaking, generating persistent torques and non-adaptive or tunable chiral trajectories [42, 43]. Although controlled chirality modulation has been achieved in bulk using feedback strategies and responsive materials [38, 44, 45], its use as an in situ control variable for navigation in porous media remains largely unexplored.

Here, we present a strategy that overcomes these limitations, enabling on-demand switching between chiral and persistent motion of colloidal rods using a frequency-switchable AC electric field. By identifying distinct frequency regimes, we successfully decouple electro-orientation from propulsion. This control suppresses the inherent chirality of the rods, converting circular trapping trajectories into persistent transport. We validate this approach in micro-structured porous media, demonstrating that the switching protocol allows rods to escape geometric traps and achieve optimal spreading, whereas constant-frequency ones lead to chiral swimmers remaining confined. This capability endows the colloids with a rudimentary form of physical intelligence [46, 47], enabling a biomimetic search-and-travel mode fundamentally different from standard protocols that simply turn motility on or off.

Results

Revealing hidden dynamical states via frequency pre-alignment

In this work, we present the dynamical behavior and navigation efficiency of rod-like Janus active colloids under electric fields. We first fabricate the Janus rod particles by drying on a flat substrate a monolayer from a colloidal suspension of rod-like SU-8 particles [48] (diameter of D=1 μmD=1\,\mu\mathrm{m}, aspect ratio ⟨L/D⟩=4.6±2.0\langle L/D\rangle=4.6\pm2.0). Then, we metalized one side with a platinum layer of dPt=6 nmd_{\mathrm{Pt}}=6\,\mathrm{nm} using a sputter coater (see Methods for more details). After the sputter coating step, particles were immersed in MilliQ water and redispersed via ultrasonication. The asymmetry in particle coverage is confirmed by energy-dispersive X-ray spectroscopy imaging as shown in Figure 1a. From the stock solution of the Janus rod-like particles, we take a volume and put it in a chamber of a height of z=120 μmz=120\,\mu\mathrm{m}, in between two parallel electrodes as shown in Figure 1a (see Methods for more details). Due to their size and density, the colloids sediment close to the bottom electrode, where they exhibit Brownian diffusion (see Supplementary Note 1, Fig. S1 and Supplementary Movie S1). Once all rod particles have sedimented, a perpendicular AC electric field is applied to investigate their dynamics at f=1f = 1 kHz, 10 kHz, and 1 MHz, and Erms=2.95E_{\mathrm{rms}} = 2.95–29.47 V mm−129.47\ \mathrm{V\ mm}^{-1} (ΔV=1\Delta V = 1–10 Vpp_{\mathrm{pp}}); representative trajectories are shown in Figure 1bFigure 1-bFigure 3.

Figure 1. Frequency-dependent behaviour of Janus rods under AC electric fields. a Plano-parallel electrodes set-up scheme. Inset: Scanning Transmission electron microscopy image of a Janus rod, and elemental analysis showing the spatial distribution of Nitrogen (green) and Platinum (magenta). Scale bar: 2 μm2\,\mu\mathrm{m}. b Diagrams and fluorescence microscopy images showing the behaviour under indicated AC electric field protocols (b1) f=1 kHzf=1\,\mathrm{kHz} and ΔV=10 Vpp\Delta V=10\,\mathrm{V}_{\mathrm{pp}}; (b2) f=10 kHzf=10\,\mathrm{kHz} and ΔV=10 Vpp\Delta V=10\,\mathrm{V}_{\mathrm{pp}}; (b3) f=1 MHzf=1\,\mathrm{MHz} and ΔV=8 Vpp\Delta V=8\,\mathrm{V}_{\mathrm{pp}}; and b4 dropping ff from f=1 MHzf=1\,\mathrm{MHz} to 1 kHz1\,\mathrm{kHz} and ΔV=10 Vpp\Delta V=10\,\mathrm{V}_{\mathrm{pp}} (orientation locking protocol, OLP) (scale bars: 10 μm10\,\mu\mathrm{m} in b1, b2, b3 and 20 μm20\,\mu\mathrm{m} in b4). c Time evolution of oriented and active population of rod-like Janus colloids under OLP. d Birefringence signal (Δn\Delta n) as a function of time. The orange curve represents the birrefringence for a f=1 kHzf=1\,\mathrm{kHz} pulse, while the purple curve shows it under the OLP application, where an initial high-frequency pulse (f=1 MHzf=1\,\mathrm{MHz}) is brought to f=1 kHzf=1\,\mathrm{kHz} at t≈7.5 st\approx7.5\,\mathrm{s}. e Rod-like Janus colloids velocity vv as a function of the intensity of the applied field ErmsE_{\mathrm{rms}}: orange squares for f=1 kHzf=1\,\mathrm{kHz}; green triangles for f=10 kHzf=10\,\mathrm{kHz}; blue circles for f=1 MHzf=1\,\mathrm{MHz}; and for OLP (purple stars). Error bars are calculated as the standard error of the mean.

Figure 3. Dynamics of rod-like Janus colloids in a porous media under AC electric fields. a Fluoresce microscopy image of polystyrene microparticles built-up porous medium, scale bar 20 μ\mum. Inset: Schematic diagram of the 2D porous media via microparticle sedimentation on the bottom electrode of the chamber and pore sizes. b–d Fluorescence microscopy image (scale bar 10 μ\mum), trajectory and diagram of the movement of a rod-like colloid in a porous medium while (b) applying a f=1f = 1 kHz field, (c) a f=10f = 10 kHz field and (d) OLP. e Trajectory of a single particle undergoing sequently the following applied ΔV=8\Delta V = 8 Vpp_{\mathrm{pp}} AC electric field: with f=1f = 1 kHz (yellow, starting point orange square), with f=10f = 10 kHz (orange, starting point green triangle) and after performing OLP (purple, starting point purple star). Inset: zoom in of f=1f = 1 kHz trajectory. f Navigation performance for the three protocols as a function of the applied field. Left axis (dark gray): straightness index, defined as the maximum distance from the starting point normalized by the total path length. Right axis (light gray): Average instantaneous velocity. Error bars represent the standard error of the mean.

For a fixed voltage difference of ΔV=10\Delta V = 10 Vpp_{\mathrm{pp}}, we identify three dynamical states. At low frequency (f∼1f \sim1 kHz, Figure 1bFigure 1 and Supplementary Movie S2) rods self-propel while lying parallel to the substrate in a planar orientation. As in their spherical Janus analogues, propulsion is dominated by asymmetric induced-charge electroosmotic (ICEO) flows around the particle [36, 49, 50], which generate electrohydrodynamic flows that drive the particle with the dielectric face leading [49]. At intermediate frequency (f∼10f \sim10 kHz, Figure 1bFigure 2 and Supplementary Movie S3), rods remain active but now electro-orient under a field-induced torque [51], and is also consistent with our previous work of rod-like colloids under the actuation of electric fields [52, 53]. The resulting field-aligned Janus rods move faster than the planar-oriented rods, due to reduced drag and substrate friction. Interestingly, they also exhibit strong chiral swimming behavior, most likely due to the anisotropy of the coating or the particle shape that magnifies when the particles stand up. At higher frequencies (f∼1f \sim1 MHz, Figure 1bFigure 3 and Supplementary Movie S4), the rods retain their orientation along the applied AC field, but are no longer active due to the decrease of the ICEO flows driving the motion as U∝f−1U \propto f^{-1} [54].

Figure 2. Chirality swimming behavior under AC electric field. a, b Representative trajectories (scale bars: 5 μ\mum) and corresponding rose diagrams of the displacement vector angle for rod-like Janus colloids under (a) f=10f = 10 kHz field and (b) the OLP, both at ΔV=8\Delta V = 8 Vpp_{\mathrm{pp}}. c–e Comparison between f=10f = 10 kHz AC field (green dashed line/triangles) and OLP (purple solid line/stars) for: (c) mean squared displacement (MSD), (d) autocorrelation of the displacement vector angle, and (e) angular velocity ω\omega as a function of the applied field. Error bars represent the standard error of the mean.

When from this oriented and not motile state at f=1f = 1 MHz, we abruptly decrease to f=1f = 1 kHz (Figure 1bFigure 4, ΔV=8\Delta V = 8 Vpp_{\mathrm{pp}}), the particles retain their orientation and display active motion revealing a hidden dynamical state induced by pre-alignment (Figure 1c and Supplementary Movie S5). Hereafter, we refer to this procedure as the Orientation Locking Protocol (OLP). In comparison with the oriented chiral state, they seem to be slower and also less chiral (as we will detail in the next section). This hidden state transition is characterized in Figure 1c, where we quantify the evolution of the population of oriented and active particles. We observe an abrupt increase in the number of oriented particles at f=1f = 1 MHz, while no active particles are detected. This field was applied for a sufficient duration to ensure the particles were electro-oriented (t∼5t \sim5 s). When decreasing the frequency from f=1f = 1 MHz (oriented not active) to f=1f = 1 kHz, the particles remain oriented for all the experimental time windows (t≫60t \gg60 s), and most of the particles become active when oriented (Figure 1c). We hypothesize that this behavior is driven by asymmetric electro-osmotic flows producing a restoring torque upon vertical tilting. This torque prevents the particles from diffusing and losing their alignment [51].

Figure 4. Experiments and simulation of active rods in porous media. a, b Fluorescence microscopy image of the porous medium (scale bar 10 μm10\,\mu\mathrm{m}) and trajectories of rod-like Janus colloids (a) under f=10 kHzf = 10\,\mathrm{kHz} field and (b) after applying an OLP. c, d Simulations of Active Brownian Particles (ABPs) in a reconstructed experimental porous medium (microparticle radius Rmp=1 μmR_{\mathrm{mp}} = 1\,\mu\mathrm{m}) with parameters matching (c) the f=10 kHzf = 10\,\mathrm{kHz} and (d) the OLP, both at ΔV=8 Vpp\Delta V = 8\,\mathrm{V}_{\mathrm{pp}}. The normalized effective displacement is Rmax⁡/L=0.004R_{\max}/L = 0.004 and Rmax⁡/LT=0.174R_{\max}/L_{\mathrm{T}} = 0.174, respectively. Insets: Detail of a single simulated trajectory under each protocol. e, f Comparison between f=10 kHzf = 10\,\mathrm{kHz} field (green triangles) and OLP (purple stars) for 1000 simulated ABPs to reach a threshold distance (Rmax⁡targetR_{\max}^{\mathrm{target}}): (e) Mean First Passage Time (MFPT) and (f) Path length (LTL_{\mathrm{T}}). Insets show the corresponding experimental data. Error bars represent the standard error of the mean, and solid lines are cubic spline interpolations. Error bars are calculated as the standard error of the mean. Solid lines are cubic spline interpolations.

To shed light on the underlying mechanism, we performed further measurements of the electro-orientation of the particles in the bulk experimental conditions under the mentioned OLP. The electric birefringence response (Figure 1d) showed that after switching the field to 1 kHz, the birefringence Δn\Delta n, which is proportional to the degree of alignment with the applied field (Supplementary Note S2, and Supplementary Fig. S2), turns to nearly the same value as the direct protocol. This means that there is no effect of the pre-orientation of the birefringence signal when comparing the OLP and direct frequency application with a final frequency of f=1f = 1 kHz.

These results suggest that the asymmetry of the fluxes generated at the bottom of the chamber are necessary for the pre-orientation not to fade.

The swimming speed vv for all of these behaviors at increasing voltage is depicted in Figure 1e, and all of the active scales as v∝E2v \propto E^2. Faster swimming rods are those in the chiral oriented state Figure 1bFigure 2. The active particles performing OLP from f=1f = 1 MHz to f=1f = 1 kHz exhibit longer persistence lengths LpL_p, than their oriented analogues at f=10f = 10 kHz (Supplementary Information, Supplementary Note S3 and Supplementary Fig. S6). Therefore, after the abrupt frequency switching, the rods retain their orientation, which affects the system dynamics and reveals a hidden dynamical state. This allows for a fast change of dynamical patterns in space, which we will further explore later in the next section.

Chirality switching to navigate porous media

We quantify in detail the dynamical behavior of the oriented-states of the active rod: i) directly actuated at f=10f = 10 kHz (Figure 2a), ii) and the oriented active Janus rod using OLP, changing from f=1f = 1 MHz to f=1f = 1 kHz (Figure 2b), both at the same applied voltage. Upon direct application of f=10f = 10 kHz the particle exhibits a chiral trajectory. However, the locking method induces a dynamical state with a persistent motion and a residual chirality. This difference in behavior is depicted in the rose plot of the angle distribution in Figure 2a and Figure 2b.

In Figure 2c, we plot the mean squared displacement (MSD) for both types of dynamical conditions. When directly applying f=10f = 10 kHz, the MSD shows pronounced oscillations (characteristic of strong chiral motion) and approaches a plateau at long lag times, consistent with confinement to a finite orbit while the particle spins continuously. In contrast, under the OLP protocol, particles behave as an active Brownian particle (ABP) with weak chirality: the MSD exhibits a short-time ballistic regime ∝v2τ2\propto v^{2}\tau^{2} followed by a crossover to an effective diffusive growth, with only a slight modulation at the chirality timescale. The strong oscillations under f=10f = 10 kHz field also show up in the correlation function of the angle (Figure 2d).

We also measure the angular velocity, ω\omega, for various particles of each class (N≈30N \approx30) in Figure 2e. As with the linear velocity measurements, the angular velocity increases proportionally to the square of the field intensity for both cases and reach higher values for chirals colloids as expected. Furthermore, population statistics confirm that there is no preferential handedness for either protocol, resulting in a racemic mixture of trajectories (see Supplementary Note S3). Thus, with this abrupt change in frequency, we can keep the same orientational behavior, but in situ and on demand change the chirality behavior of the active Janus, which is particularly interesting for the exploration of porous media.

To quantify the transport performance of the different on-demand dynamical states in complex environments, we designed porous media using assembled microparticles. Briefly, a droplet of polystyrene spheres suspension with a radius of Rmp=1.0 μR_{\mathrm{mp}} = 1.0\ \mum was dried onto a slide, and the beads were subsequently sintered on the bottom electrode as depicted in Figure 3a (more details about this protocol in Supplementary Note S4, Fig.S4, and Methods). Then, the parallel electrode configuration was assembled identically to previous experiments. When actuating the rods directly at f=1f = 1 kHz (planar motion) they typically get stuck due to their bigger size with respect to the array characteristic constriction width (Figure 3b and Supplementary Movie 6). When applying f=10f = 10 kHz (oriented and highly chiral, Figure 3c and Supplementary Movie 7), the particles effectively probe a smaller confined region than navigating the entire landscape, despite their high swimming speed due to their higher chirality (with a minority of planar particles getting trapped). Finally, when applying the OLP 1 MHz–to–1 kHz (oriented and more persistent), the particles ticles align efficiently with the field, easily escape from confined pores, and swim throughout the porous medium (Figure 3d and Supplementary Movie 8). In Figure 3e, we depict the trajectory of a single rod within the porous array while changing the frequency and actuation sequence. While the particle remains hindered at f=1 kHzf = 1\,\mathrm{kHz} and locally confined despite high speeds at f=10 kHzf = 10\,\mathrm{kHz}, the OLP enables effective obstacle sorting and particle movement.

In order to quantify their navigation efficiency, we calculated straightness index as the ratio of the longest distance from the starting point (Rmax⁡R_{\max}) to the total path length traveled (LTL_{\mathrm{T}}), as well as the mean instantaneous velocity (Figure 3f). Although the particles were faster under a f=10 kHzf = 10\,\mathrm{kHz} field, those under the OLP stood out in the evaluation of the greatest Rmax⁡/LTR_{\max}/L_{\mathrm{T}}. Hence, we can clearly distinguish two regimes: one in which particles remain confined, probing small regions (f=10 kHzf = 10\,\mathrm{kHz}), and another in which particles can readily escape and travel between distant regions (OLP).

Mean-first passage time: a comparison to experiments

To gain a comprehensive understanding of the role of chirality in exploring complex media, we performed numerical simulations to reproduce the experimental environment and extend the parameter space to conditions unattainable in our current system but relevant to future chiral particle design. In Figure 4, we compare experimental observations (Figure 4a and b) and numerical results obtained using the active Brownian particle (ABP) model [55] under two distinct conditions (Figure 4c and d, see Methods for more details). The experimental porous environment was reconstructed by replicating the static obstacles directly from the experimental configurations. Steric interactions between the particles and the obstacle boundaries were explicitly accounted for via elastic collision, ensuring no overlap between the swimmers and the porous matrix. The first set of simulations reproduces the oriented chiral motion state at f=10 kHzf = 10\,\mathrm{kHz} (Figure 4c), using the experimentally measured values v=70 μm s−1v = 70\,\mu\mathrm{m}\,\mathrm{s}^{-1} and ω=25 rad s−1\omega= 25\,\mathrm{rad}\,\mathrm{s}^{-1} as input parameters. The second case mimics the 1 MHz1\,\mathrm{MHz}–to–1 kHz1\,\mathrm{kHz} OLP state (Figure 4d), for which we adopt v=20 μm s−1v = 20\,\mu\mathrm{m}\,\mathrm{s}^{-1} and ω=0.5 rad s−1\omega= 0.5\,\mathrm{rad}\,\mathrm{s}^{-1}. This procedure was repeated for N=1000N = 1000 independent trajectories, with the full particle paths recorded for post-processing (see Supplementary Note S5 for details).

Compared with the experiments, the simulated particles representing chiral motion state (Figure 4c) appear less strongly trapped near the walls. This difference likely arises from the absence of short-range attractive interactions with the wall due to the AC field application. Nevertheless, in the simulations, the particles predominantly remain confined within a well-defined region over time due to their intrinsic chiral motion. In contrast, in the simulations corresponding to the OLP 1 MHz–to–1 kHz field state (Figure 4d), particles escape most efficiently from the porous array, moving more persistently across the landscape.

To characterize the spatio-temporal exploration performance in complex environments, we report the Mean First Passage Time (MFPT) in Figure 4e for simulations and experiments. This metric is defined as the average time required for a swimmer, starting from a random position, to first reach a predefined target displacement, Rmax⁡targetR_{\max}^{\mathrm{target}} [56, 57]. Mathematically, for a single trajectory r(t)r(t), the first passage time is given by τ(Rmax⁡target)=min⁡(t>0:∣r(t)∣≥Rmax⁡target)\tau(R_{\max}^{\mathrm{target}}) = \min\left(t > 0 : \lvert r(t)\rvert\ge R_{\max}^{\mathrm{target}}\right). Further details regarding these and other metrics used in this section can be found in Supplementary Fig. S4a and Supplementary Note S5.

As observed in Figure 4e, the high-chirality swimmer reaches the target distance (Rmax⁡target<6 μm=2v/ωR_{\max}^{\mathrm{target}} < 6\,\mu\mathrm{m} = 2v/\omega) faster due to its larger propulsion speed. However, once the target distance exceeds the characteristic orbital diameter, the MFPT for the f=10f = 10 kHz regime exhibits a sharp upturn towards higher timescales comparable to the persistent swimmer.

timescales comparable to the persistent swimmer. At these scales, the dynamical advantage of higher speed is canceled by the path tortuosity, which effectively traps the particle in a localized area. Consequently, at larger Rmax⁡targetR_{\max}^{\mathrm{target}}, the dynamical effect of the higher speed is cancelled out, and both protocols converge to a similar long-time MFPT.

While the MFPT converges at large distances, the LTL_{\mathrm{T}} reveals a fundamental difference between both strategies (Figure 4f). To quantify this, we analyze the covered distance (LTL_{\mathrm{T}}) to reach Rmax⁡targetR_{\max}^{\mathrm{target}}. As shown in Figure 4f, the high-chirality swimmer consistently travels significantly larger LTL_{\mathrm{T}} to reach the same Rmax⁡targetR_{\max}^{\mathrm{target}}. The continuous spinning induces frequent lateral deflections and wall encounters, forcing the particle to explore the void space redundantly (low straightness index, Rmax⁡/LT≪1R_{\max}/L_{\mathrm{T}} \ll1). In contrast, the persistent swimmer explores the porous landscape more efficiently through its openings despite its lower velocity. Thus, it achieves a ratio of displacement to path length closer to the high persistent motion limit (Rmax⁡/LT→1R_{\max}/L_{\mathrm{T}} \to1). The separation in LTL_{\mathrm{T}} is amplified, precisely when the MFPT values merge as chiral particles practically loop around a point. Although the simulations do not perfectly match all the experimental measurements (insets in 4 e and f), comparable MFPT values and divergence in LTL_{\mathrm{T}} at large Rmax⁡targetR_{\max}^{\mathrm{target}} establish a robust framework. Consequently, we can exploit these simulations as a reliable tool to investigate further transport regimes beyond our current experimental reach.

Using the numerical model, we further examine different random porous media and decouple the effect of swimming speeds and chirality (not possible in the experimental system) by extending the value of vv and ω\omega to a broad range of values (Figure 5). We simulate trajectories with a fixed linear vv and varying ω\omega (Figure 5a), and at fixed ω\omega varying vv (Figure 5b). We summarise our results of measuring the MFPT and Rmax⁡/LT→1R_{\max}/L_{\mathrm{T}} \to1 for these two variables. In Figure 5c, we identify two distinct transport regimes separated by a geometric threshold at v/ω∼2 μmv/\omega\sim2\,\mu\mathrm{m}. In the sub-pore regime (v/ω<2 μmv/\omega< 2\,\mu\mathrm{m}), particles undergo localized circular motion, and increasing linear velocity is counterproductive as it merely increases the redundant path length (LTL_{\mathrm{T}}) without facilitating escape. Therefore, the highest navigation efficiency is consistently obtained for the low-velocity particles. In the escape regime (v/ω>2 μmv/\omega> 2\,\mu\mathrm{m}), the trajectory radius exceeds the pore dimensions, forcing wall encounters that break the chiral confinement and enable long-range navigation. An overview of the navigation performance using the MFPT for Rmax⁡target=10 μmR_{\max}^{\mathrm{target}} = 10\,\mu\mathrm{m} is provided in Figure 5d. Consistent with the previous results, in the subporous regime (v/ω<2 μmv/\omega< 2\,\mu\mathrm{m}), high chirality increases exploration time by confining the particle regardless of its propulsion capacity.

Figure 5. Performance of simulated ABP in microparticle built-up porous arrays. a Trajectories of single ABP with different value of ω\omega and a fixed value of v=40 μm s−1v = 40\,\mu\mathrm{m}\,\mathrm{s}^{-1} for randomly generated porous arrays. b Trajectories of single ABP with different value of vv and a fixed value of ω=5 rad s−1\omega= 5\,\mathrm{rad}\,\mathrm{s}^{-1} for randomly generated porous arrays. c Heatmap of mean Rmax⁡/LTR_{\max}/L_{\mathrm{T}} as a function of the angular velocity ω\omega and the linear velocity vv (NABP=500N_{\mathrm{ABP}=500}). d Heatmap of MFPT for Rmax⁡target=10 μmR_{\max}^{\mathrm{target}} = 10\,\mu\mathrm{m} as a function of the angular velocity ω\omega and the linear velocity vv (NABP=500N_{\mathrm{ABP}=500}).

Our results suggest an interplay between the two magnitudes, revealing a crossover regime: particles with low linear and angular velocities can exhibit smaller MFPTs than those with higher values of both. Results for other values of Rmax⁡targetR_{\max}^{\mathrm{target}} can be found in Supplementary fig:S7 in Supplementary Note S5.

Discussion

The dynamical behavior and exploration strategy pattern of microswimmers is crucial for designing complex environment exploration [11, 58]. In particular, the ability to switch on-demand between persistent and chiral behavior provides a versatile mechanism to reconfigure microswimmer dynamics according to its specific goals and the local environmental constraints.

Here, we have presented a strategy to tune the orientation, swimming speed and chirality of Janus active rods actuated via AC electric fields only by switching the field frequency regime. Our work identifies and exploits the hidden dynamical state of the active rods changing from chiral to persistent motion while keeping a perpendicular orientation with respect to the plane. We therefore explore the navigation features of our system in experiments and simulations in porous media. We reveal a fundamental trade-off: while high chirality favors local exploration, it leads to entrapment. Conversely, the suppressed chirality achieved via the pre-alignment of the rod-like particles enables efficient long-range transport and obstacle avoidance.

This tunability, achieved here via frequency-switching AC electric field, could be a benchmark to design exploration pathways with other types of actuation mechanisms. The ability to switch in-situ between searching (chiral) and traveling (persistent) states mimics the diversity of swimming strategies seen in biological microswimmers, with precise external control in their synthetic analogues. This opens new avenues for applications in targeted cargo delivery within biological tissues, soil remediation, and the engineering of active materials that can self-regulate their transport properties.

Methods

Fabrication of Janus rod-like particles

To fabricate the rod-like active Janus particles, polymeric rods [48] were diluted in MilliQ water and deposited onto a glass slide, previously rendered hydrophilic by air plasma treatment. The suspension is sufficiently diluted to ensure a sub-monolayer where individual rods align with their long axes parallel to the substrate. Upon drying under controlled humidity, the monolayer is sputter-coated (Safematic CCU-010, Switzerland) with 6 nm of Pt. The particles are retrieved from the glass slide by sonicating the substrate in MilliQ water.

Cell preparation and AC electric field experiments

The transparent electrodes are fabricated from 18mm ×\times 18mm No.0 cover glasses (85-115μ\mum -thick, Menzel Gläser, Germany) covered with 3nm of Cr and 10nm of Au using a sputter coater (Safematic CCU-010, Switzerland). The bottom electrode is functionalized with a 10 mg/mL PSS (Sigma Aldrich, USA) solution to minimize particle adhesion to the substrate during the experiments. The solution is allowed to adsorb for 10 minutes before the electrode is rinsed and dried. A water droplet containing the particles is deposited on the bottom electrode within the 9mm circular opening of a 0.12 mm thick sealing spacer (Grace Bio-Labs Secure-Seal, USA). The top and bottom electrodes are connected to a signal generator (National Instruments Agilent 3352X, USA) that applies the AC electric field, varying frequencies from f=1f = 1 kHz to f=1f = 1 MHz and varying the voltage between ΔV=2\Delta V = 2 Vpp and ΔV=10\Delta V = 10 Vpp.

Characterization of ABP dynamics

The Janus particles are imaged in bright field and fluorescence using an inverted microscope Olympus IX71 equipped with 40×\times objective. Image sequences are acquired with a sCMOS camera (Andor Zyla 4.2) at frame rates varying between 10 and 100fps, depending on the swimming speed and concentration of the particles. Particle trajectories were extracted using the TrackMate plugin in FIJI [59]. Subsequently, we measured the velocities of rod-like Janus particles by studying the mean squared displacement for small values of Δt\Delta t

MSD(Δt)=4DtΔt+v2Δt2(1)\mathrm{MSD}(\Delta t) = 4D_t\Delta t + v^2\Delta t^2 \tag*{(1)}

being DtD_t the translational diffusion coefficient and vv the propulsion velocity of the active colloid. Then, the persistence length (LpL_p) which represents the distance it takes for an active colloid to be statistically considered diffusive, is computed by fitting the MSD(Δt)\mathrm{MSD}(\Delta t) for greater values of Δt\Delta t:

MSD(Δt)=4DtΔt+Lp2(DrΔt−1+exp⁡(−DrΔt))(2)\mathrm{MSD}(\Delta t) = 4D_t\Delta t + L_p^2\left(D_r\Delta t - 1 + \exp(-D_r\Delta t)\right) \tag*{(2)}

being DrD_r the rational diffusion coefficient [60].

ABP Particle-based model

The motion of the ABP is modeled using the overdamped Langevin equations [55]. The particle’s position vector r⃗(t)=(x(t),y(t))\vec{r}(t) = (x(t), y(t)) and orientation angle θ(t)\theta(t) evolve over time according to:

r⃗˙(t)=vu^(θ)+2Dtξ⃗(t)(3)\dot{\vec{r}}(t) = v\hat{u}(\theta) + \sqrt{2D_t}\vec{\xi}(t) \tag*{(3)}
θ˙(t)=ω+2Drη(t)(4)\dot{\theta}(t) = \omega+ \sqrt{2D_r}\eta(t) \tag*{(4)}

Where vv is the constant self-propulsion speed and u^(θ)=(cos⁡θ,sin⁡θ)\hat{u}(\theta) = (\cos\theta,\sin\theta) is the unit vector representing the orientation of the particle’s propulsion axis. The term ω\omega is a deterministic angular velocity, causing the particle to execute a persistent circular motion.

The terms ξ⃗(t)\vec{\xi}(t) and η(t)\eta(t) represent uncorrelated Gaussian white noise, accounting for thermal fluctuations. These noise terms have zero average and satisfy ⟨ξi(t)ξj(t′)⟩=δijδ(t−t′)\langle\xi_i(t)\xi_j(t')\rangle= \delta_{ij}\delta(t-t') and ⟨η(t)η(t′)⟩=δ(t−t′)\langle\eta(t)\eta(t')\rangle= \delta(t-t'). The strengths of these fluctuations are determined by DtD_t and DrD_r.

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Code availability statement

The code used in this study is available from the corresponding author upon reasonable request.

Acknowledgements

We thank José Fojo for the help with the first AC field experiment setup. L.A. thanks the French Agence Nationale de la Recherche (ANR-23-CE06-0007-0) and Frontiers of Life (U. Bordeaux) for funding. M.L.J.O. and R.R. thank Grant No. PID2024.161166NB.I00 funded by MI-CIU/AEI/10.13039/501100011033/FEDER/UE for funding. A.C. thanks Grant No. PRE2022-2-101943 funded by MI-CIU/AEI /10.13039/501100011033 and by FSE+ for the financial support.

Author Contributions

Author contributions are defined based on the CRediT (Contributor Roles Taxonomy) and listed alphabetically. Conceptualization: L.A.; Formal analysis: A.C., Funding acquisition: L.A., M.L.J.O., R.R.; Investigation: L.A., A.C.; Methodology: L.A., A.C., C.F.R; Project administration: L.A.; Software: L.A., A.C.; Supervision: L.A., M.L.J.O., R.R.; Validation: L.A., M.L.J.O., R.R.; Visualization: L.A., A.C.; Writing – original draft: L.A., A.C. Writing – review, and editing: L.A., A.C. M.L.J.O., R.R., C.F.R.

Competing Interests

The authors declare no competing interests.

Additional Information

Supplementary Information is available for this paper. Correspondence and requests for materials should be addressed to L.A.

References

  1. [1]Tapomoy Bhattacharjee and Sujit S. Datta, “Bacterial hopping and trapping in porous media,” Nature Communications 10, 2075 (2019).DOI
  2. [2]Christina Kurtzhaler, Suvendu Mandal, Tapomoy Bhattacharjee, Hartmut Löwen, Sujit S Datta, and Howard A Stone, “A geometric criterion for the optimal spreading of active polymers in porous media,” Nature Communications 12, 7088 (2021).DOI
  3. [3]Clemens Bechinger, Roberto Di Leonardo, Hartmut Löwen, Charles Reichhardt, Giorgio Volpe, and Giovanni Volpe, “Active particles in complex and crowded environments,” Rev. Mod. Phys. 88, 045006 (2016).DOI
  4. [4]Matthew F Krummel, Frederic Bartumeus, and Audrey Gérard, “T cell migration, search strategies and mechanisms,” Nature Reviews Immunology 16, 193–201 (2016).DOI
  5. [5]Marco Dentz, Adama Creppy, Carine Douarché, Eric Clément, and Harold Auradou, “Dispersion of motile bacteria in a porous medium,” Journal of Fluid Mechanics 946, A33 (2022).DOI
  6. [6]Howard C. Berg and Douglas A. Brown, “Chemotaxis in escherichia coli analysed by three-dimensional tracking,” Nature 239, 500–504 (1972).pubmed.ncbi.nlm.nih.gov/4563019
  7. [7]Kevin J Modica, Yuchen Xi, and Sho C Takatori, “Porous media microstructure determines the diffusion of active matter: experiments and simulations,” Frontiers in Physics 10, 869175 (2022).DOI
  8. [8]Ali Karimi, Mohammad Yaghoobi, and Alireza Abbaspourard, “Geometry of obstructed pathway regulates upstream navigational pattern of sperm population,” Lab on a Chip 25, 631–643 (2025).DOI
  9. [9]Michael Mihlan, Katharina M Glaser, Maximilian W Epple, and Tim Lämmermann, “Neutrophils: amoeboid migration and swarming dynamics in tissues,” Frontiers in cell and developmental biology 10, 871789 (2022).DOI
  10. [10]Amin Dehkhargahni, Nicolas Waisbord, and Jeffrey S Guasto, “Self-transport of swimming bacteria is impaired by porous microstructure,” Communications Physics 6, 18 (2023).arxiv.org/abs/2201.03059
  11. [11]Metin Sitti, Kathan Ceylan, Wenqi Hu, Joshua Guntilan, Mehmet Turan, Sehyuk Yim, and Eric Diller, “Biomedical applications of untethered mobile milli/microrobots,” Proceedings of the IEEE 103, 205–224 (2015).pubmed.ncbi.nlm.nih.gov/27746484
  12. [12]Ugur Bozuyuk, Paul Wrede, Erdost Yildiz, and Metin Sitti, “Roadmap for clinical translation of mobile microrobotics,” Advanced Materials 36, 2311462 (2024).DOI
  13. [13]Vivien Willems, Alexandre Baron, Daniel Fernandez-Matoz, Gianna Wolfsberg, Jean-Christophe Baret, Eric Dufresne, and Laura Alvarez, “Run-and-tumble dynamics of active giant vesicles,” Soft Matter (2025).DOI
  14. [14]Yeongju Jung, Kangkyu Kwon, Jinwoo Lee, and Seung Hwan Ko, “Untethered soft actuators for soft standalone robotics,” Nature Communications 15, 3510 (2024).DOI
  15. [15]Francesca Oltolina, Ana Peigneux, Donato Colangelo, Nausicaa Clemente, Annarita D’Urso, Guido Valente, Guillermo R Iglesias, Concepcion Jimenez-Lopez, and Maria Prat, “Biomimetic magnetite nanoparticles as targeted drug nanocarriers and mediators of hyperthermia in an experimental cancer model.” Cancers 12, 2564 (2020).DOI
  16. [16]Arturo Moncho-Jordá, Ana B Jódar-Reyes, Matej Kanduč, Alicia Germán-Bellod, Juan M López-Romero, Rafael Contreras-Cáceres, Francisco Sarabia, Miguel García-Castro, Hector A Perez-Ramirez, and Gerardo Odriozola, “Scaling laws in the diffusive release of neutral cargo from hollow hydrogel nanoparticles: Paclitaxel-loaded poly (4-vinylpyridine),” ACS nano 14, 15227–15240 (2020).DOI
  17. [17]Pablo Graván, Juan Antonio Marchal, and Francisco Galisteo-González, “Improving tumor treatment: Cell membrane-coated nanoparticles for targeted therapies,” Materials Today Bio , 101716 (2025).DOI
  18. [18]Bradley J Nelson, Ioannis K Kaliakatsos, and Jake J Abbott, “Microrobots for minimally invasive medicine,” Annual Review of Biomedical Engineering 12, 55–85 (2010).DOI
  19. [19]Simone Gervasoni, Norman Pedrini, Tarik Rifai, Cedric Fis­cher, Fabian C Landers, Michael Mattmann, Roland Dreyfus, Silvia Viviani, Andrea Veciana, Enea Masina, et al., “A human-scale clinically ready electromagnetic navigation system for magnetically responsive biomaterials and medical devices,” Advanced Materials 36, 2310701 (2024).DOI
  20. [20]Moqiu Zhang, Lidong Yang, Haojin Yang, Lin Su, Junnan Xue, Qinglong Wang, Bo Hao, Yihang Jiang, Kai Fung Chan, Joseph Jao Yiu Sung, et al., “A magnetically actuated microcatheter with soft rotatable tip for enhanced endovascular access and treatment efficiency,” Science advances 11, eadv1682 (2025).DOI
  21. [21]Jaideep Katuri, Xing Ma, Morgan M. Stanton, and Samuel Sánchez, “Designing micro- and nanoswimmers for specific applications,” Accounts Chem. Res. 50, 2–11 (2017).DOI
  22. [22]Stefano Palagi and Peer Fischer, “Bioinspired microrobots,” Nat. Rev. Mater. 3, 113–124 (2018).
  23. [23]Nandan Murali, Shashank Bhushan Das, Satyam Yadav, Simran Kaur Rainu, Neetu Shingh, and Soutik Betal, “Advanced biomimetic and biohybrid magnetic micro/nano-machines,” Advanced Materials Technologies 9, 2400239 (2024).DOI
  24. [24]Mustafa Sabri Kilic and Martin Z Bazant, “Induced-charge electrophoresis near a wall,” Electrophoresis 32, 614–628 (2011).DOI
  25. [25]Alicia Boymelgreen and Gilad Yossifon, “Observing electrokinetic janus particle–channel wall interaction using microparticle image velocimetry,” Langmuir 31, 8243–8250 (2015).DOI
  26. [26]Antonio Ramos, Pablo Garcia-Sanchez, and Hywel Morgan, “AC electrokinetics of conducting microparticles: A review,” Current Opinion in Colloid & Interface Science 24, 79–90 (2016).DOI
  27. [27]Giovanni Volpe, Ivo Buttoni, Dominik Vogt, Hans-Jürgen Kümmerer, and Clemens Bechinger, “Microswimmers in patterned environments,” Soft Matter 7, 8810 (2011).DOI
  28. [28]Touvia Miloh, “Dipolophoresis of janus nanoparticles in a microchannel,” Electrophoresis 34, 1939–1949 (2013).
  29. [29]AM Boymelgreen, G Kunti, Pablo Garcia-Sanchez, A Ramos, G Yossifon, and T Miloh, “The role of particle-electrode wall interactions in mobility of active janus particles driven by electric fields,” Journal of Colloid and Interface Science 616, 465–475 (2022).
  30. [30]Juliane Simmchen, Jaideen Katuri, William E Uspal, Mihail N Popescu, Mykola Tasinkevych, and Samuel Sánchez, “Topographical pathways guide chemical microswimmers,” Nature Communications 7, 10598 (2016).DOI
  31. [31]Megan S Dykes, Xiao Zhong, Jiajun Tong, Takuji Adachi, Y a peng Liu, Leif Ristroph, Michael D Ward, Michael J Shelley, and Jun Zhang, “Guiding microscale swimmers using teardrop-shaped posts,” Soft Matter 13, 4681–4688 (2017).
  32. [32]Carolina van Baalen, Stefania Ketzetzi, Anushka Tintor, Israel Gabay, and Lucio Isa, “Gating and tunable confinement of active colloids within patterned environments,” Soft Matter 21, 3850–3858 (2025).arxiv.org/abs/2501.00660
  33. [33]Simona Colabrese, Kristian Gustavsson, Antonio Celani, and Luca Biferale, “Flow navigation by smart microswimmers via reinforcement learning,” Phys. Rev. Lett. 118, 158004 (2017).DOI
  34. [34]Hamid Karani, Gerardo E. Pradillo, and Petia M. Vlahovska, “Tuning the random walk of active colloids: From individual run-and-tumble to dynamic clustering,” Phys. Rev. Lett. 123, 208002 (2019).DOI
  35. [35]Alexander R. Sprenger, Miguel Angel Fernandez-Rodriguez, Laura Alvarez, Lucio Isa, Raphael Wittkowski, and Hartmut Löwen, “Active brownian motion with environment-dependent motility: theory and experiments,” Langmuir 36, 7066–7073 (2020).DOI
  36. [36]Miguel Angel Fernandez-Rodriguez, Fabio Grillo, Laura Alvarez, Marco Rathlef, Ivo Buttinnoni, Giovanni Volpe, and Lucio Isa, “Feedback-controlled active brownian colloids with space-dependent rotational dynamics,” Nature Communications 11, 4223 (2020).DOI
  37. [37]Stefania Ketzetzi, Lorenzo Caprini, Vivien Willems, Laura Alvarez, Hartmut Löwen, and Lucio Isa, “Active colloidal molecules with dynamic configurational freedom,” ACS Nano 19, 29430–29439 (2025).DOI
  38. [38]L. Alvarez, M. A. Fernandez-Rodriguez, A. Alegria, S. Arrese-Igor, K. Zhao, M. Kröger, and Lucio Isa, “Reconfigurable artificial microswimmers with internal feedback,” Nature Communications 12 (2021), 10.1038/s41467-021-25108-2.DOI
  39. [39]Danne M van Roon, Giorgio Volpe, Margarida M Telo da Gama, and Nuno A M Araújo, “The role of disorder in the motion of chiral active particles in the presence of obstacles,” Soft Matter 18, 6899–6906 (2022).DOI
  40. [40]Danne M. van Roon, Giorgio Volpe, Margarida M. Telo da Gama, and Nuno A. M. Araújo, “Three-dimensional model for surface accumulation of chiral and nonchiral microswimmers,” Phys. Rev. E 111, 045405 (2025).DOI
  41. [41]Chung Wing Chan, Daihui Wu, Kaiyao Qiao, Kin Long Fong, Zhiyu Yang, Yilong Han, and Rui Zhang, “Chiral active particles are sensitive reporters to environmental geometry,” Nature Communications 15, 1406 (2024).DOI
  42. [42]Stephen Ebbens, Richard A.L. Jones, Anthony J. Ryan, Ramin Golestanian, and Jonathan R. Howse, “Self-assembled autonomous runners and tumblers,” Phys. Rev. E 82, 6–9 (2010).pubmed.ncbi.nlm.nih.gov/20866681
  43. [43]Hartmut Löwen, “Chirality in microswimmer motion: From circle swimmers to active turbulence,” The European Physical Journal Special Topics 225, 2319–2331 (2016).DOI
  44. [44]Allan M. Brooks, Syeda Sabrina, and Kyle J. M. Bishop, “Shape-directed dynamics of active colloids powered by induced-charge electrophoresis,” Proc. Natl. Acad Sci. USA 115, E1090–E1099 (2018).DOI
  45. [45]Steven van Kesteren, Laura Alvarez, Silvia Arrese-Igor, Angela Alegria, and Lucio Isa, “Self-propelling colloids with finite state dynamics,” Proceedings of the National Academy of Sciences 120, e2213481120 (2023).DOI
  46. [46]Metin Sitti, “Physical intelligence as a new paradigm,” Extreme Mechanics Letters 46, 101340 (2021).pubmed.ncbi.nlm.nih.gov/35475112
  47. [47]Agrim Gupta, Silvio Savarese, Surya Ganguli, and Li Fei-Fei, “Embodied intelligence via learning and evolution,” Nature communications 12, 5721 (2021).DOI
  48. [48]Carla Fernández-Rico, Taiki Yanagishima, Arran Curran, Dirk GAL Aarts, and Roel PA Dullens, “Synthesis of colloidal su-8 polymer rods using sonication,” Advanced Materials 31, 1807514 (2019).DOI
  49. [49]Sumit Gangwal, Olivier J. Cayre, Martin Z. Bazant, and Or-lin D. Velev, “Induced-charge electrophoresis of metallodielectric particles,” Phys. Rev. Lett. 100, 058302 (2008).pubmed.ncbi.nlm.nih.gov/18352441
  50. [50]Chenhui Peng, Israel Lazo, Sergij V. Shiyanovskii, and Oleg D. Lavrentovich, “Induced-charge electro-osmosis around metal and janus spheres in water: Patterns of flow and breaking symmetries,” Phys. Rev. E 90, 051002 (2014).DOI
  51. [51]Paloma Arenas-Guerrero, Ángel V Delgado, Antonio Ramos, and María L Jiménez, “Electro-orientation of silver nanowires in alternating fields,” Langmuir 35, 687–694 (2018).DOI
  52. [52]José Fojo, Rodolfo Subert, Laura Rodríguez-Arco, Modesto T López-López, Marjolein Dijkstra, Carla Fernández-Rico, and Laura Alvarez, “Field-driven reversible networks from colloidal rods,” Soft Matter 21, 4596–4605 (2025).DOI
  53. [53]Ana Cazorla, Sergio Martín-Martín, Ángel V Delgado, and María L Jiménez, “Electro-optics of confined systems,” Journal of Colloid and Interface Science 658, 52–60 (2024).DOI
  54. [54]Xingfu Yang, Sophia Johnson, and Ning Wu, “The impact of stern-layer conductivity on the electrohydrodynamic flow around colloidal motors under an alternating current electric field,” Adv. Intell. Syst. 1, 1900096 (2019).DOI
  55. [55]Giorgio Volpe, Sylvain Gigan, and Giovanni Volpe, “Simulation of the active brownian motion of a microswimmer,” American journal of physics 82, 659–664 (2014).DOI
  56. [56]Maryam Khatami, Katrin Wolff, Oliver Pohl, Mohammad Reza Ejtehadi, and Holger Stark, “Active brownian particles and run-and-tumble particles separate inside a maze,” Scientific reports 6, 37670 (2016).DOI
  57. [57]Yuguang Yang, Michael A Bevan, and Bo Li, “Efficient navigation of colloidal robots in an unknown environment via deep reinforcement learning,” Advanced Intelligent Systems 2, 1900106 (2020).arxiv.org/abs/1906.10844
  58. [58]Giorgio Volpe and Giovanni Volpe, “The topography of the environment alters the optimal search strategy for active particles,” Proceedings of the National Academy of Sciences 114, 11350–11355 (2017).DOI
  59. [59]Jean-Yves Tinevez, Nick Perry, Johannes Schindelin, Genevieve M Hoopes, Gregory D Reynolds, Emmanuel Laplantine, Sebastian Y Bednarek, Spencer L Shorte, and Kevin W Eliceiri, “Trackmate: An open and extensible platform for single-particle tracking,” Methods 115, 80–90 (2017).DOI
  60. [60]Maximilian Bailey, Alexander Sprenger, Fabio Grillo, Hartmut Löwen, and Lucio Isa, “Log it: How to fit an active brownian particle’s mean squared displacement with improved parameter estimation,” Physical Review E (2022).DOI

Paper details

vlnc:W_3mzeeb8a
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents