Sorting of binary active-passive mixtures in designed microchannels

1. Problem Statement

The paper addresses the challenge of separating and sorting passive particles from active-passive mixtures at the microscale. While active-passive mixtures are ubiquitous in biological systems (e.g., bacteria interacting with dead cells or cargo), controlling the transport and separation of passive components within these non-equilibrium systems remains difficult.

Specifically, the authors investigate how to exploit the unique properties of active matter (self-propelled particles) to enhance the transport of passive particles and achieve efficient sorting. The goal is to design a microchannel geometry that acts as a semi-permeable membrane, allowing passive particles to pass while blocking active ones, thereby separating the two species based on their motility and interaction with the environment.

2. Methodology

The study employs computational simulations using Molecular Dynamics (MD) to model a two-dimensional binary mixture confined in a designed microchannel.

• System Composition:
  • Active Particles (Type 1): Modeled as "run-and-tumble" particles (mimicking bacteria like E. coli or microalgae like C. reinhardtii). They follow overdamped Langevin translational dynamics and stochastic rotational dynamics characterized by a tumbling rate ($\alpha$).
  • Passive Particles (Type 2): Modeled as Brownian particles with no self-propulsion.
  • Geometry: The particles are confined in a microchannel with periodic boundary conditions in the x-direction and walls in the y-direction. The central region contains funnel-like obstacles with gaps specifically sized to allow only passive particles to pass (active particles are too large to fit through the gaps when oriented correctly).
• Simulation Parameters:
  • Dynamics: The system solves the full Langevin equation (including inertial effects at short timescales) but operates in the overdamped regime.
  • Variables: The study systematically varies the tumbling rate ($\alpha$) of the active particles and the angle of the obstacles ($\theta_{obs}$) relative to the channel axis.
  • Metrics: Key metrics include the fraction of passive particles in the bottom compartment ($\langle F_{bottom} \rangle$), the time constant of the sorting process ($\tau_d$), the fraction of open gaps, and the transport properties (diffusion coefficient and drift velocity) of passive particles.
• Analysis Tools:
  • Kinetic Modeling: The sorting dynamics are fitted to a first-order kinetic model to extract rate constants ($k_+$ and $k_-$).
  • Advection-Diffusion (AD) Model: A minimal 1D AD model is constructed to rationalize the separation mechanism, incorporating position-dependent diffusion ($D(y)$) and drift velocity ($v(y)$).
  • Gap Analysis: The "clogging area" is defined geometrically to calculate the probability of gaps being open or blocked by active particles.

3. Key Contributions

• Optimization of Sorting Efficiency: The paper identifies a specific regime of tumbling rates and obstacle angles that maximizes the separation efficiency of passive particles.
• Mechanism of Active Pumping: The authors demonstrate that the sorting is not merely due to steric exclusion or differences in diffusion coefficients. Instead, it is driven by an active-pumping mechanism where the active bath induces a net downward drift (advection) of passive particles in the top compartment.
• Role of Geometry vs. Activity: The study decouples the effects of channel geometry and active particle dynamics, showing that while geometry dictates the available flow paths, the activity level determines the "pumping" strength and gap availability.
• Validation via Minimal Models: The complex simulation results are successfully rationalized using a simple Advection-Diffusion model, highlighting the dominance of the advective term over the diffusive term in driving the separation.

4. Key Results

A. Sorting Dynamics and Kinetics

• The sorting process follows first-order kinetics.
• There exists an optimal range for the tumbling rate ($\alpha \approx 1 \times 10^{-4}$ to $2 \times 10^{-4}$) where the separation efficiency is maximized.
• At the optimal parameters (specifically $\theta_{obs} = 75^\circ$ and $\alpha \approx 1.25 \times 10^{-4}$), the system achieves a steady-state passive particle concentration in the bottom compartment of $\langle F_{bottom} \rangle \approx 0.92$. This represents a ~30% improvement over passive-passive mixtures (where only steric interactions exist, yielding $\approx 0.70$).
• The response time (time to reach steady state) is minimized in the same parameter range.

B. The Role of Gaps and Clogging

• Gap Availability: The fraction of open gaps ($P_{total}$) shows an inflection point at the optimal tumbling rate.
  • At low $\alpha$ (high persistence), active particles accumulate in concave regions and clog gaps for long periods.
  • At high $\alpha$ (low persistence), active particles reorient too frequently to effectively push passive particles, and gaps are open but the "push" is weak.
  • The optimal $\alpha$ balances the frequency of active-particle interactions with the duration of gap blockage.
• Geometry Dependence: While the steady-state separation fraction is largely independent of the obstacle angle ($\theta_{obs}$), the speed of sorting increases with $\theta_{obs}$. The angle $75^\circ$ provides the fastest dynamics.

C. Transport Properties: Diffusion and Advection

• Enhanced Diffusion: The active bath significantly enhances the diffusion coefficient of passive particles in the top compartment compared to the bottom (thermal) compartment.
• Induced Drift (Advection): Crucially, the interaction between active particles and the asymmetric confinement (funnels) generates a net downward drift velocity ($v_{top} < 0$) for passive particles.
• Dominance of Advection: The minimal Advection-Diffusion model reveals that the high sorting efficiency is primarily driven by this negative active drift, not just by the difference in diffusion coefficients. Without the drift term, the model fails to reproduce the high separation efficiency observed in simulations.

D. Physical Mechanism

The peak performance occurs when the persistence length of the active particles ($l_p$) is comparable to the length of the obstacles ($L$). This balance ensures:

1. Active particles travel far enough to effectively "kick" passive particles downward.
2. Active particles reorient frequently enough to avoid permanently clogging the gaps, maintaining a flow path for passive particles.

5. Significance and Implications

• Fundamental Physics: The work provides a clear mechanistic understanding of how confinement and activity interact to generate non-equilibrium transport phenomena (active pumping) in binary mixtures. It highlights the importance of the interplay between persistence length and obstacle geometry.
• Technological Applications: The findings offer a blueprint for designing microfluidic devices for:
  • Cargo Delivery: Efficiently transporting passive cargo (e.g., drugs or nanoparticles) using active biological agents.
  • Bioremediation: Separating specific micro-contaminants from complex biological mixtures.
  • Micro-separators: Creating passive, energy-efficient sorting mechanisms powered solely by the internal energy of the active bath.
• Future Directions: The authors suggest future work involving hydrodynamic interactions and experimental realization using Chlamydomonas reinhardtii and colloidal beads, as well as exploring synchronized active pumping to further enhance transport.

In summary, the paper demonstrates that by tuning the tumbling rate of active agents and the geometry of microchannels, one can create an active-pumping mechanism that efficiently sorts passive particles, achieving separation efficiencies far superior to passive systems.