Surface-Sensitive Fractioning of Flowing Colloidal Suspensions Sedimented at a Photochemically Active Wall

Abstract

A colloidal particle exposed to nonequilibrium inhomogeneities in the chemical-composition of the solution responds by phoresis with a magnitude dictated by its surface properties (adsorption of solutes relative to the solvent) and the strength of the inhomogeneity gradient. Fractioning by surface-properties for a suspension of sedimented colloidal particles flowing within a microfluidic device can then be achieved by exposing it to chemical-composition gradients normal to the wall. One such method is developed by employing a wall imbibed with a photo-sensitive surfactant; under suitable illumination, it undergoes reversible isomerization and turns the wall into a source of chemical inhomogeneities by isomers-exchange with the solution. Proof-of-concept experiments, complemented by theoretical modeling, demonstrate the feasibility of the proposed method in several examples of particles mixtures. The results highlight its potential in terms of high performance, simple technical requirements, and suitability for dealing with equally-sized-but surface different-particles.

Keywords: azobenzene containing surfactant; chemical activity; fractioning; light‐driven diffusio‐osmosis; microfluidics; microparticle hovering; phoresis.

Figure 1

Illustration (not at scale) of the experimental setup. a) Schematic representation of a porous PVPBMA interface which absorbs the azobenzene containing surfactant. b) AFM height image of PVPBMA in 3D view together with a possible scheme of the pore structure (bottom). c) Dispersion of compact and porous microparticles and solution of azobenzene containing surfactant are injected into the rectangular microfluidic chamber. Yellow rectangle displays the chemical structure of azobenzene‐containing trimethyl‐ammonium bromide surfactant (full name 6‐[4‐(4‐Hexylphenylazo)‐phenoxyl‐butyl‐trimethylammoniumbromide, C₄‐Azo‐OC₆ TMAB). Once particles fully sediment at the bottom layer (PVPBMA or glass as reference) a pressure‐driven fluid flow is switched on. The motion and trajectory of passively transported particles is recorded via video microscopy. The “plain” label indicates compact (nonporous) particles. In the region under illumination (the blue shades), the wall is active due to the photoisomerization of the surfactant and the consequent exchange cis out–trans in with the solution; once in this region, the particles experience the phoretic hovering and, thus, a change in their drift velocity by the ambient flow.

Figure 2

a,b) Scheme of the velocity‐boosting effect on compact and porous SiO₂ microparticles sedimented at glass and polymer interface, respectively. Blue and black arrows represent the lift of compact and porous silica microparticles by LDDO mechanism, respectively, while yellow arrow represents the additional lift boosting from polymer surface. c,d) Mean drift velocity (line) and corresponding standard deviation (filled transparent area) as a function of time for compact (c) and porous (d) SiO₂ microparticles. e) Release rate of surfactant molecules (orange solid line) and mean velocity of compact SiO₂ particles on polymer (orange dashed line) and glass (black solid line = release rate, black dashed line = mean velocity) surface. A clear correlation between the release of surfactant by the wall and the increase in the drift velocity is noticeable. The blue rectangles indicate the illuminated (λ = 455 nm) region, while the time stamps refer to the moment when the recording started (5 s before the light is turned on). f–h) Mean drift velocity of particles under illumination (after 15 s of illumination, t = 20–25 s) displayed as a function of the light intensity at λ = 365 nm (f); λ = 455 nm (g) and λ = 490 nm (h) for compact (triangles) and porous (circles) SiO₂ microparticles above a glass (gray symbols) or a PVPBMA polymer (orange symbols) wall, respectively. The means are computed over an average sample size of n particles per frame of d) glass wall: n = 209 ± 11, PVPBMA wall: n = 189 ± 11; e) glass wall: n = 158 ± 18, PVPBMA wall: n = 123 ± 23. or the data in (f–h) the corresponding average sample size per frame is summarized in Table S3 (Supporting Information). Measurements were done by illuminating only a specific region.

Figure 3

Snapshots (from Videos S7 and S8, Supporting Information) of optical view of the compact (of size D = 4 µm) and porous particles (of size D = 3 µm) before illumination (t = 0 s) and at time t = 9 s after the modulated light spot with moderate intensity (λ = 365 nm, I = 10 mW cm⁻²) was turned on. The focal plane is maintained at the t = 0 position, and thus the elevation of the particles in the rectangular‐shaped illuminated area results in out‐of‐focus imaging. The external flow is switched off. Measurements were done by illuminating only a specific region.

Figure 4

a) Theoretically calculated density of cis‐monomers (color coded) and streamlines of its gradient, i.e., the phoretic velocity response of a particle. The results correspond to a patch of radius 100 R _p, which is similar to the size of the illuminated area in the experimental setup, and to repulsive interactions between the particle and the cis‐isomers. Note that near the wall the streamlines are basically vertical over the whole surface of the patch, i.e., the phoretic velocity is to a very good approximation along the direction of the sedimentation velocity. b) Theoretically calculated (see Theoretical Methods Section and Section S4, Supporting Information) elevation heights (under illumination) above a glass (gray symbols) or a PVPBMA (orange symbols) for plain (circles) or porous (squares) silica particles.

Figure 5

a,b) Mean velocity (thick lines) and the corresponding standard deviation (transparent area) as a function of time for compact SiO₂ microparticles above a glass wall (a) and a PVPBMA active wall (b). The particles have different surface functionalization, being either unmodified, thus with a hydroxyl‐ (─OH) group, or modified with an amino‐group (─NH₂). The blue area marks the time interval during which the light (λ = 490 nm) is on. Only the data in the last 5 s of illumination (t = 20–25 s) is used in the calculation of the average mean velocity and standard deviation shown in (c) and (d). c,d) Summary of the average velocity for example pairs of “weakly chemically active “ microparticles with different surface modification in the case of motion above a (c) glass and (d) PVPBMA wall, respectively. SiO₂ = silica, PS = polystyrene, compact = plain = no surface modification (─OH group), Aminogroup = ─NH₂, ternary functionalized amines = ─NR₃ ⁺, Octyldecylgroup = C18, Carboxylgroup = ─COOH. The corresponding raw data (time resolved average velocity) is shown in the Figures S18–S20 (Supporting Information). Measurements were done by illuminating the region globally. Particle diameter is 5 µm.

Figure 6

Summary of velocity difference from the data in Figure 5c,d as a function of the example pairs of “weakly chemically active “ microparticles with different surface SiO₂ = silica, PS = polystyrene, plain = no surface modification (─OH group), Aminogroup = ─NH₂, ternary functionalized amines = ─NR₃ ⁺, Octyldecylgroup = C₁₈, Carboxylgroup = ─COOH. The gray symbols are results in the case of an inactive glass wall, and the yellow symbols correspond to the results in the case of a chemically active PVPBMA wall. Measurements were done by illuminating the region globally. Particle diameter is 5 µm.

Figure 7

Mean velocity of porous silica microparticles (P‐SiO₂) as a function of intensity (λ  =  455 nm) on glass (blue data) and polymer (orange data) surface at different particle sizes: a) D  =  3 µm, b) D  =  5 µm, c) D  =  20 µm. d) Relative mean velocity (U _PVPBMA/U _glass) as a function of intensity (λ = 455 nm). Illustration of measured trend in d for e) the particle mass and f) intensity dependence. Thus, the velocity of different sized porous particles (D  =  3, 5, and 20 µm) is recorded as a comparison between glass and PVPBMA in Videos S12–S14 (Supporting Information) at local blue light illumination (λ  =  455 nm, I  = 254 mW cm⁻²). Measurements were done with local light illumination. Video S12 (D = 3 µm), Video S13 (D = 5 µm), Video S14 (D = 20 µm).