Reconfigurable Optofluidic Barriers for 3D Particle Control via Thermally Induced Flows

E. Ruiz-Reina1, F. Schmidt2, C. D. González-Gómez3, M. Sulliger2, R. A. Rica-Alarcón4, J. Ortega-Arroyo2, R. Quidant2
1Multiphysics Modeling School, Department of Applied Physics II, University of Málaga
2Department of Mechanical and Process Engineering, ETH Zurich
3Department of Applied Physics, University of Granada
4Research Unit ‘Modeling Nature’ (MNat), Department of Applied Physics, University of Granada
Publié en 2025

Microfluidic systems traditionally rely on fixed physical barriers to manipulate fluids and particles. While effective, these static structures lack the flexibility required for dynamic or multifunctional applications. In this work, recently published in Nature Photonics [1], we present a reconfigurable optofluidic platform that uses structured light and photothermal conversion to generate virtual thermal barriers within microfluidic chambers. These barriers enable real-time, three-dimensional control over fluid flow and particle trajectories, offering a versatile alternative to traditional microfluidic architectures. We used COMSOL Multiphysics to simulate the coupled heat transfer and fluid dynamics that arise from localized laser-induced heating. The simulations were essential to understand and optimize the interplay between thermo-osmosis, thermophoresis, and natural convection in the formation of virtual barriers. The Heat Transfer in Fluids and Laminar Flow physics interfaces were coupled via the Nonisothermal Flow multiphysics interface. Additionally, we employed the Particle Tracing for Fluid Flow interface to model the motion of colloidal particles under the influence of thermal gradients. To replicate the experimental setup, we modeled a microfluidic chamber with gold nanorod-coated surfaces and introduced localized heat sources corresponding to laser foci. We implemented custom boundary conditions to simulate thermo-osmotic slip velocities and included thermophoretic forces in the particle tracing module. The simulations were validated against experimental data obtained via digital holographic microscopy and optical diffraction tomography. Our results show that optofluidic barriers can be dynamically reconfigured to perform key microfluidic operations such as steering, splitting, merging, and trapping of particles. Furthermore, by arranging thermal pillars in periodic patterns, we demonstrated size-based particle sorting akin to deterministic lateral displacement (DLD) assays. The COMSOL simulations were instrumental in optimizing the spatial configuration and thermal input required for efficient sorting. Applications and Impact: This technology opens new avenues for adaptive lab-on-chip devices, where multiple fluidic functions can be integrated and reconfigured in real time. Potential applications include: - Chemical synthesis, where dynamic control over mixing and reaction zones enhances efficiency and selectivity. - Biomedical diagnostics, enabling precise manipulation of cells and biomolecules without physical contact. - Single-cell analysis, where reconfigurable traps allow long-term observation and isolation. - Environmental sensing, through selective particle sorting and concentration. - Soft robotics and programmable matter, where fluidic control can be embedded into responsive systems. The ability to simulate complex thermal-fluid-particle interactions with COMSOL Multiphysics was key to engineering multifunctional microfluidic environments with high spatial and temporal resolution. This approach demonstrates how virtual barriers can surpass the limitations of physical structures, offering a flexible and scalable solution for next-generation microfluidic technologies.