Controlling the collective transport of large passive particles with suspensions of microorganisms

TL;DR

This work demonstrates how to control the collective transport of hundreds of large passive particles with phototactic microalga using directional light stimuli in suspensions of Chlamydomonas reinhardtii to trigger bioconvection rolls capable of macroscale transport.

Abstract

A promising approach to transport cargo at the microscale lies within the use of self-propelled microorganisms, whose motion entrains that of passive particles. However, most applications remain limited to just a few passive particles of similar size as the microorganisms, since the transport mechanism relies on the interaction between individual swimmers and single particles. Here, we demonstrate how to control the collective transport of hundreds of large passive particles with phototactic microalga. Using directional light stimuli in suspensions of Chlamydomonas reinhardtii, we trigger bioconvection rolls capable of macroscale transport. Passive particles an order of magnitude larger than the microalgae are either attracted or repelled by the rolls depending on their density. Using experiments and simulation, we rationalize these bioconvective flows and describe how to harness them for cargo transport, with future applications in targeted drug delivery and decontamination.

Figures (5)

• Figure 1: Experimental setup - Photo-bioconvective flows displace large beads. (a) Schematic image of the setup. A square chamber of width $\qty{9}{\mm}$ and height $H$ is placed on top of a red light LED panel. The experiment is recorded with a camera placed above the chamber. Two blue light LED bands are placed on each side of the chamber. (b) Cross-sectional view of the chamber. The chamber of height $H$ is filled with a suspension of algae Chlamydomonas reinhardtii and polyethylene (PE) beads. (c) Sketch of the recirculating convection rolls that appear when algae accumulate in a region of the chamber. Passive particles in range of those rolls are either attracted or repelled depending on their relative density with the medium. Dense beads are drawn in black and light beads in white. (d-g) Repulsion of denser beads of diameter $d_b=\qty{50}{\um}$, located on the bottom surface, in a chamber of height $H=\qty{310}{\um}$ (top view). As negatively phototactic algae accumulate at the lateral wall, beads denser than the fluid are pushed away, forming a front (see https://seminaris.polytechnique.fr/share/s/wmNYoFtoTnya5FZ). (h-k) Attraction of lighter beads of diameter $d_b=\qty{460}{\um}$, located on the top surface, in a chamber of height $H=\qty{930}{\um}$ (top view). Both lateral LEDs are then switched on, forming a region of highly concentrated algae which draws in beads lighter than the fluid (see https://seminaris.polytechnique.fr/share/s/aQJpTfqWfPjMjm3). Both experiments (d-k) are conducted with an initial optical density OD$_i$ of 10, i.e. $c_i=\qty{3e7}{\cells\per\mL}$.
• Figure 2: Bead dynamics within bioconvective flows. (a) Experimental trajectories of beads of diameter $d_b = \qty{110}{\um}$, in a chamber of height $H=\qty{310}{\um}$, pushed away from the chamber lateral wall by phototactic C. reinhardtii accumulating at the top boundary, $\mathrm{OD}_i = 10$ (top view). (b) Simulated evolution of the local density $\rho$ in the chamber, alongside the flow field it generates. Red regions correspond to zones of high velocity and illustrate the formation and outward propagation of bioconvective rolls. The successive positions of a simulated bead of diameter $d_b = \qty{50}{\um}$ sitting on the floor of the chamber is indicated by black disks. (i): $t=\qty{10}{\s}$. (ii): $t=\qty{100}{\s}$. (iii): $t=\qty{480}{\s}$. See https://seminaris.polytechnique.fr/share/s/qbYKmyknoKSa9CG. (c) Position $y$ normal to the wall (magenta), and velocity $v$ (teal) of three representative beads of diameter $d_b = \qty{50}{\um}$, initially located at increasing distances from the wall. Experimental data are shown as solid lines while simulated data with a $t_\mathrm{lag}\simeq2\,$min are displayed as dashed lines. Top: A bead, initially close to the wall ($y_i=\qty{125}{\um}$), is rapidly accelerated by the algae-induced bioconvective flow, reaching a peak velocity of approximately $\qty{10}{\micro\metre\per\second}$, before decelerating and entering a steady regime at $\qtyrange{0.5}{1}{\micro\metre\per\second}$. Experiment and simulations are in excellent agreement. Middle: A bead starting farther away ($y_i=\qty{545}{\um}$) is accelerated more gradually and reaches a lower peak velocity. Experiment and simulations are in very good agreement. Bottom: A bead initially far from the wall ($y_i=\qty{880}{\um}$) moves at much lower speeds. Experiment and simulations are in good agreement (see main text for discussion). (d) Instantaneous simulated bead velocity $v_y^{\mathrm{sim}}$ as a function of the instantaneous experimental velocity $v_y^{\mathrm{exp}}$, for beads initially located within 1200 of the wall for an initial optical density $\rm{OD}_i=10$. Their initial position $y_i$ is color-coded from purple ($y_i=107\um$) to yellow ($y_i=1200\um$). The black line shows $v_y^{\mathrm{sim}} = v_y^{\mathrm{exp}}$.
• Figure 3: Denser beads form a front when repelled from dense algal regions. (a) Profiles of algal concentration $\mathrm{OD}(y)$ near the lateral wall ($y=0$) for different initial optical densities OD$_i$. The profiles are taken $t=\qty{10}{\minute}$ after the blue LED is switched on. The position of the front of beads $y_\mathrm{front}$ is denoted for each curve with a filled colored symbol. Simulated profiles, in dashed lines, are also superimposed with their respective front position which is calculated as the position where the vertical fluid velocity is maximal (hollow symbols). (b) Temporal evolution of the experimental algal concentration profile $\mathrm{OD}(y)$ near the wall ($y=0$) for an initial optical density OD$_i=10$. Initially, there is no light stimulus so the profile is roughly flat. After the blue LED is switched on at $t=0$, the algae begin to accumulate at the wall. The position of the beads front $y_\mathrm{front}$ is denoted at each time with a colored star. See Supplementary Figure 13 for a comparison between the experimental and simulated profiles. (c) Temporal evolution of the position of the beads front $y_\mathrm{front}$ for different initial optical densities OD$_i$, in the experiments (symbols) and in the numerical simulations without (dashed lines) or with adaptation (dotted lines). (d) Front velocity $v_\mathrm{front}$ scales linearly with $\mathbf{u}_{\rm photo} \bar{c}_i(1-\bar{c}_i)$, both in the simulations (small symbols) and in the experiments (large symbols with errorbars). Simulations were done with a normalized initial concentration $\bar{c}_i=c_i/c_{\rm max}$ ranging from 0.05 to 0.3, corresponding to $c_i=\qtyrange{1.1e7}{1.0e8}{\cells\per\mL}$ or OD$_i=\qtyrange{3.8}{34}{}$, and a phototactic velocity $u_\mathrm{photo}=\qtyrange{10}{80}{\um\per\s}$. Solid line: $v_\mathrm{front}=1.3\,\bar{c}_i (1-\bar{c}_i) u_{\rm photo}$.
• Figure 4: Surface cleaning metrics. (a) Temporal evolution of the bead surface fraction on the bottom surface $\Psi_b$, normalized by its initial value $\Psi_b^0=\Psi_b(t=0)$ before the blue light is switched on. $\Psi_b$ is measured in a 6x0.9 region, corresponding to 67x10% of the whole chamber, located at the wall opposite of the light stimulus (see green rectangle in left inset, a typical image of beads in the entire 9x9 chamber at $t=0$). Right inset: Characteristic cleaning time $\tau$, calculated from the exponential fits $\Tilde{\Psi}_b=\Tilde{\Psi}_0\exp((t_0-t)/\tau)$. (b) Cleaning efficiency $\varepsilon$ of the beads by the algal bioconvection rolls, as a function of the initial optical density OD$_i$. $\varepsilon=1$ means that every bead was successfully swept away. $\varepsilon$ is measured by comparing the bead surface fraction, in the region located between the wall and 200 ahead of the final bead front position, between $t=0$ and after the roll has vanished. (c) Temporal evolution of the cleaned surface $\tilde{S}_\mathrm{clean}(t)$ across the whole chamber. $\tilde{S}_\mathrm{clean}(t)$ corresponds to $S_\mathrm{clean}$ subtracted by the initial value $S_\mathrm{clean}^0=S_\mathrm{clean}(t=0)$ before the blue light is switched on. A region is considered cleaned if there are strictly less than 2 beads in a 200x200 region. (d) Total cleaned surface $\tilde{S}_\mathrm{clean}^\mathrm{tot}$ across the entire chamber, as a function of the initial optical density OD$_i$. $\tilde{S}_\mathrm{clean}^\mathrm{tot}=\langle \tilde{S}_\mathrm{clean}(t)\rangle_{t=10..15\min}$. Insets: Snapshots of the beads in the entire 9x9 chamber at $t=20\,$min.
• Figure 5: Large-scale transport of large particles. (a)-(c) An "algae cannonball" cleans the chamber. Time-lapse of an experiment with 50 beads in a dense C. reinhardtii suspension at $\mathrm{OD}_i = 7$ (top view). Both lateral LEDs are switched on, creating a concentrated algal plume far from the walls. Denser beads are pushed away from the plume as it moves forward, see https://seminaris.polytechnique.fr/share/s/BQz7jc2bp3dYFqm. In less than an hour, a region stretching over 4x5 is almost perfectly cleaned from beads. (d)-(f) Transporting a raft of beads. Time-lapse of an experiment with 230 beads in a dense C. reinhardtii suspension at $\mathrm{OD}_i = 10$ (top view). By carefully adjusting LED intensities, the algal plume and bead raft are gradually moved, displacing the raft by almost 4 mm over two hours. The latter collects neighboring beads as it moves, see https://seminaris.polytechnique.fr/share/s/fqeBDiCN7Q8rsxY. (g) Temporal evolution of the cleaned surface $\tilde{S}_\mathrm{clean}(t)=S_\mathrm{clean}(t)-S_\mathrm{clean}^0$ by the 'algae cannonball' shown in (a)-(c). Orange dashed line: affine fit $\tilde{S}_\mathrm{clean}=\gamma t +S_0$ with $\gamma=\qty{0.2}{\square\mm\per\min}$ and $S_0=\qty{-1.2}{\square\mm}$. (h) Temporal evolution of the bead surface fraction on the top surface $\Psi_b$ for the beads raft shown in (d)-(f), normalized by its initial value $\Psi_b^0=\Psi_b(t=0)$ before the blue light is switched on. $\Tilde{\Psi}_b$ is measured in three different regions of 7^2: a target region to clean (green), a target region to transport the beads to (blue) and a control region (red) as reference. Their respective locations are shown in insets.