[Paper Review] Viscophoretic particle transport

Vahid Khandan, Vincent J.P. Boerkamp+6 more|arXiv (Cornell University)|Dec 22, 2022

Microfluidic and Bio-sensing Technologies4 citations

TL;DR

This study experimentally demonstrates viscophoretic particle transport in microfluidic devices with steep, stable viscosity gradients, revealing drift velocities far exceeding theoretical predictions. Using high-resolution fluorescence microscopy and correlation analysis, the authors quantify viscophoresis and demonstrate a novel method for nanoparticle trapping via synergistic viscophoresis and diffusiophoresis.

ABSTRACT

Viscosity is a fundamental property of liquids and determines the diffusivity of suspended particles. A gradient in viscosity leads to a gradient in diffusivity, yet it is unknown whether such a gradient can lead to directed transport of particles. In this work, we generate a steep, stable viscosity gradient in a microfluidic channel and image the resulting transport of suspended nanoparticles at the single-particle level using high-resolution microscopy. We observe high viscophoretic drift velocities that significantly exceed theoretical predictions. In addition, we utilize viscophoresis for a new type of particle trap. We provide a first quantification of a transport phenomenon that is of importance in any system and any application exhibiting viscosity gradients, for example in separation using membrane technology as well as in inter- and intracellular biomolecular transport.

Motivation & Objective

• To experimentally investigate viscophoretic transport of nanoparticles driven by viscosity gradients at the single-particle level.
• To resolve the Itô-Stratonovich dilemma in transport with space-dependent diffusivity by measuring actual drift velocities.
• To develop a microfluidic platform capable of generating and sustaining steep, stable viscosity gradients for precise particle manipulation.
• To explore the interplay between viscophoresis and diffusiophoresis for novel particle concentration and trapping strategies.
• To quantify viscophoretic drift velocities and compare them with theoretical predictions based on different stochastic process interpretations (α = 0, 1/2, 1).

Proposed method

• Fabricated microfluidic devices with PDMS and borosilicate glass, using oxygen plasma surface activation and silane-based functionalization to create negatively charged surfaces that repel fluorescent nanoparticles.
• Engineered a dual-inlet system with syringe pumps to generate stable, steep viscosity gradients by balancing flow rates inversely proportional to viscosity ratios.
• Employed a home-built TIRF microscopy setup with 402 nm, 520 nm, and 680 nm lasers and sCMOS camera for high-speed imaging at 200–500 fps with 180–360 nm effective pixel resolution.
• Applied pair correlation function (pCF) analysis with Fast Fourier Transform to quantify mean square displacement (MSD) and extract directional diffusion coefficients in eight angular sectors.
• Calculated drift velocity from the center displacement of the pCF at increasing lag times, enabling precise velocimetry of individual nanoparticles.
• Used 8 nm, 28 nm, and 110 nm carboxylated fluorescent beads to investigate size-dependent viscophoretic effects across different diffusivity regimes.

Experimental results

Research questions

• RQ1 What is the magnitude of viscophoretic drift velocity for nanoparticles in a stable, steep viscosity gradient, and how does it compare to theoretical predictions?
• RQ2 Which stochastic process (Itô, Stratonovich, or isothermal) best describes viscophoretic transport in a space-dependent diffusivity field?
• RQ3 How do particle size and diffusivity variation across the gradient influence the net drift velocity?
• RQ4 Can viscophoresis be combined with diffusiophoresis to achieve active particle trapping and concentration in microfluidic systems?
• RQ5 What experimental evidence supports the existence of viscophoretic transport at the single-particle level, given its prior inference only through ensemble measurements?

Key findings

• Viscophoretic drift velocities for nanoparticles significantly exceed theoretical predictions, particularly for larger particles (e.g., 110 nm), indicating strong non-equilibrium transport effects.
• The observed drift velocities are inconsistent with the Itô (α = 0) or Stratonovich (α = 1/2) interpretations, supporting the isothermal model (α = 1) as the most appropriate for colloidal systems.
• A stable, steep viscosity gradient was achieved in microfluidic channels with a 2 μm × 5 μm cross-section, enabling precise control and observation of particle motion.
• The combination of viscophoresis and diffusiophoresis enabled effective trapping and concentration of nanoparticles in a defined region of the microchannel, demonstrating a new microfluidic principle.
• High-resolution pCF analysis revealed directional diffusion anisotropy, with increased MSD in the direction of decreasing viscosity, confirming net transport toward lower viscosity.
• Drift velocities were quantified at the single-particle level with high temporal resolution (up to 500 fps), enabling accurate measurement of both diffusion and directed motion.