Photosensitive Spherical Polymer Brushes: Light-Triggered Process of Particle Repulsion

Abstract

We report on a light-triggered process at which repulsive interactions between microparticles with a polyelectrolyte (PE) brush coating can be remotely controlled. The spherical polyelectrolyte brushes are loaded with photosensitive azobenzene containing surfactant which can undergo reversible photo-isomerization from trans to cis state. The surfactant hydrophilicity is altered by illumination with light of an appropriate wavelength, at which a dynamic exchange of the more surface-active trans isomer in comparison to the more water soluble cis isomer with the PE brush generates a concentration gradient of the cis isomers near a solid surface where the particle is sedimented. In this way, each spherical brush produces its local lateral diffusioosmotic flow pointing outside in a radial direction resulting in mutual long-range repulsive interactions. We demonstrate that a PE layer has a higher tendency to absorb surfactant in comparison to plain silica particles, yielding a larger flow strength. This correlation holds true up to a critical intensity, where the dynamic exchange is adsorption limited with respect to trans isomers and especially pronounced for the PE-coated particles.

1. Introduction

Spherical polyelectrolyte brushes (SPBs) consist of polyelectrolyte chains covalently grafted to a curved surface of nanoparticles or microparticles [1,2]. One of the ways to synthesize SPBs is the “grafting from“ approach, where first, small initiator molecules are attached to a particle surface followed by the growing of polymer chains directly out of the surface [3,4,5,6,7,8,9,10,11]. When the brush is placed in water most of the counterions are trapped within the shell, building up high osmotic pressure (osmotic limit) and thus resulting in a swelling of the brush [12,13,14]. In addition, the ionizable groups on a polymer offer a possibility to modify SPBs a posteriori by electrostatic coupling with surfactants, proteins, or nanoparticles or by using them as nanoreactors for the generation of metallic and oxidic nanoparticles [15,16,17,18,19,20,21,22,23,24]. During the uptake of, for instance, surfactants, the native counterions of the brush are released owing to an overall gain in entropy [25,26,27,28,29,30]. This may result in shrinkage of the brush due to the drop in the osmotic pressure or even the overcharging of the polymer shell [31,32,33]. In this way, SPBs can be considered to be a smart/responsive material, since they can drastically react to variations in environmental conditions (the presence of surfactants, proteins, and so on) or even tiny changes in pH or in the ionic strength of the solution [34,35,36,37,38,39]. However, SPBs can do even more; namely, they can pump liquid around, which is shown in our paper for the first time.

Here, we report on spherical polyelectrolyte brushes attached to silica micro-sized particles loaded with photosensitive surfactant. We demonstrate that such functionalized brushes can be made to be active in response to an interaction potential when exposed to irradiation with light of an appropriate wavelength. This takes place owing to the generation of local-light-driven diffusioosmotic (l-LDDO) flow by the particle itself, so that the radially directed flow results in long-range diffusioosmotic (DO) repulsion between the particles, the strength and duration of which can be easily controlled by the intensity of the applied light [40]. The photosensitive surfactant consists of cationic head connected through a spacer of six CH₂ groups to an azobenzene molecule having a butyl tail [41]. Owing to the presence of azobenzene in the hydrophobic tail, depending on the wavelength of the applied irradiation, the surfactant can be brought reversibly into two distinct states: a more hydrophobic trans one and a more hydrophilic cis one [42]. We have demonstrated that, depending on the isomer state, the surfactant molecules prefer to be either within charged confined geometry (trans isomers) such as pores in colloidal particles [43], the charged interior of microgel [44,45], planar polyelectrolyte brushes [46,47], linear polyelectrolyte chains [48,49], and other types of surfaces (plain glass and Teflon, hair, nails, sand, and so on) [50,51], while cis isomers readily leave and prefer to state in the bulk. The reversible uptake and release of the surfactant molecules builds up the concentration gradient of surrounding isomers, and in the case of objects sedimented at the solid/liquid interface, the corresponding gradient in osmotic pressure establishes fluid motion [52]. An ensemble of such particles function via diffusioosmotic flows establishing the range, the strength, and even the sign of interaction potential and may form either a sparse or a densely packed 2D crystalline-like pattern [53].

Using the example of a poly(methacrylic acid) (PMAA) spherical brush, we have found that the dynamic exchange of the isomers within the brush depends strongly on the light intensity having its critical value at 4 mW/cm², where it is defined by three major points: limited diffusivity of the isomers in the PE layer during light illumination, reduced trans–cis isomerization kinetics inside the PE layer, and trans isomer adsorption reduction towards the PE layer. All these factors result in a reduction in the effective strength of DO flow although the SPBs have a high absorbance tendency in comparison to plain colloids.

The paper is structure as follows: (1) we describe the triggering of the synthesized spherical brushes’ photo-responsiveness by loading them with azobenzene containing a surfactant; (2) we study the generation of the local-light-driven diffusioosmotic flow by the brushes and how the strength of the flow depends on irradiation parameters; (3) we compare the behavior of the spherical brushes with porous and plain colloids.

2. Experimental Part

Acetic acid, 3-aminopropyltriethoxysilane (APTS), 2-bromo-2-methylpropionyl bromide (BIBB), ethyl acetate, copper(I)bromide (99.99%), dichloromethane (DCM), hydrochloric acid, N,N,N′,N″,N″-pentamethyl diethylenetriamine (PMDETA), tert-butyl methacrylate (tBuMA), toluene, trifluoroacetic acid (TFA), and sodium hydroxide are purchased from Sigma-Aldrich and used without further purification.

Non-porous silica particles (4.00 ± 0.14 μm) and porous particles (4.9 ± 0.5 μm are purchased from microParticles GmbH (Germany) and dialyzed five times against 5 L Millipore water.

Poly(methacrylic acid)-coated particles SiO₂@PMAA (Figure 1f) are synthesized as described elsewhere (for more details, see Supporting Information, Section S1) [54]. In short, the surface initiator is first grafted onto the particle surface to allow for a subsequent “grafting from” polymerization. The amino functionalization is performed with 3-aminopropyltriethoxysilane (APTS) in acidic water and then reacted in toluene with 2-bromo-2-methylpropionyl bromide (BIBB) to attach the initiator group on the surface. For polymerization, we utilize surface-initiated ATRP with Cu(I)Br as the catalyst and PMDETA as the ligand in a methanol–water mixture (V_MeOH/V_H2O = 4:1), keeping the particle concentration low to avoid the fusion of polymer chains from different particles. Polymerization is performed with a high monomer and Cu to initiator molar ratio, and the monomer concentration is set to 122 mM of tert-butyl methacrylate (tBuMA) per reaction batch. Anionic polyelectrolyte brushes are obtained by the polymerization of tBuMA and subsequent hydrolysis in DCM/TFA (trifluoroacetic acid) to obtain poly(methacrylic acid) brushes (SiO₂@PMAA). After the reaction, polymer-coated SiO₂ particles are transferred into water by dialysis, and charged by adjusting the solution pH to 9. A schematic representation of the synthesis is depicted in Figure S1, Supporting Information. The molecular parameters of the spherical polymer brushes are grafting density of ca. 0.5 nm⁻² and molecular weight of 7·10⁴ g/mol (details of the calculation are presented in Section 1, Supporting Information). The measurements are performed at pH = 6.

Light responsive surfactant. The azobenzene containing trimethyl-ammonium bromide surfactant (C₄-Azo-OC₆-TMAB) is synthesized as described elsewhere [48]. The surfactant (Figure 1f) consists of a spacer of 6 methylene groups between the positively charged trimethylammonium bromide head group and the azobenzene unit with butyl tail attached. A stock solution of surfactant is adjusted to the concentration of 10 mM and diluted to the required sample concentrations ranging from 0.1 to 2 mM. The aqueous dispersion of all particles is mixed with surfactant stock solution and Millipore water at different concentrations and kept at least 1 day for equilibration. A flow chamber with a volume of 40 µL is used in order to provide a closed environment. All samples are kept in the dark or in red light to prevent unwanted photo-isomerization.

Sample preparation for AFM height investigation to estimate brush height. The polydimethylsiloxan (PDMS) film with embedded colloidal SiO₂@PMAA is prepared by cast molding using commercial Sylgard 184 (Dow Corning). This is performed by mixing elastomer and curing agent in a ratio of 10:1, pouring the mixture onto the polymer colloid spheres, and curing it for 24 h at RT. The films are then detached from the substrate and deposited backwards on the glass slide.

Atomic force microscopy measurements. The AFM measurements are carried out using an NTEGRA AFM (NT-MDT) operating in intermittent contact mode. Commercial tips (Nanoworld-Point probe) with a resonance frequency of 320 kHz and a spring constant of 42 N/m are used.

Optical imaging. Image acquisition is performed in video-recording mode at a monitoring speed of 1 frame per second with an Olympus IX73 equipped with a commercial double monochromatic LED light source (Thorlabs; for blue, λ = 455 nm, M455 L2-C1; for red, λ = 625 nm, M625 L2-C), where images are monitored with a Hamamatsu Camera ORCA-Flash 4.0 LT. The samples are kept in the dark to prevent uncontrolled isomerization. The motion of silica particles on a glass substrate is studied under illumination from above with monochromatic light of λ = 455 nm (blue) and λ = 625 nm (red) wavelength. The intensity of irradiation is measured at the sample position ranging from 0.1–50 mW/cm².

Particle motion is tracked using the Mosaic Single Particle Tracking plugin for ImageJ (Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, MD, USA) [55]. Motion analysis and calculation has been implemented in Matlab.

Thermogravimetric analysis (TGA) is performed using thermal analysis TGA/DTA L81 to 1500 °C (Linseis). In a typical measurement, previously dried samples of SiO₂ and SiO₂@PMAA are heated in a synthetic air environment over a temperature range of 20–800 °C and a heating rate of 10 °C/min.

Zeta potential is calculated from the electrophoretic mobility measured using commercial zetasizer (Nano-ZS, Malvern Instruments, Ltd., Malvern, UK). Prior samples are exposed for 12 min under irradiation with light of the desired wavelength (I = 10 mW/cm²), followed by sonication for 10 s.

UV-Vis spectroscopy (Cary 5000 UV-Vis-NIR spectrophotometer, Agilent Technologies, USA) is used to measure the amount of surfactant absorbed by the particles.

Calculation of the amount of adsorbed surfactant: samples are prepared in a manner similar to that previously described, where the surfactant and particle concentration is adjusted to 2 mM and 0.01 mg/mL, respectively. The sample is illuminated by light (λ = 455 nm) with an intensity of 10 mW/cm² for 15 min, followed by centrifugation for 1 min. The collected supernatant is diluted 100 times and measured using a UV-Vis spectrometer. The concentration of adsorbed surfactant is calculated via:

$$c_{\mathrm{ads}}=c_{\mathrm{surf}}-c_{\mathrm{UV}-\mathrm{Vis}}·100$$

(1)

with c_ads, c_surf, and c_UV-Vis as the adsorbed, adjusted, and measured surfactant concentration. C_UV-Vis is calculated using the Beer–Lambert law. Details for the calculation are presented in the Supporting Information, Section S4.

SEM micrographs are recorded with a commercial ZEIS ULTRA PLUS-40-61 microscope.

3. Results and Discussion

Spherical brushes consisting of silica core and poly(methacrylic acid) (PMAA) tethered chains (Figure 1f) are synthesized using an ATRP reaction as described elsewhere (Figure S1, Supporting Information) [54]. The molecular weight of the PMAA chains is fixed to 70·10³ g/mol according to the synthesis parameters as reported for similar brushes (PMMA) on planar interfaces [56]. To calculate the grafting density we utilize two approaches. By measuring the dry thickness of the scratched brush (~36 nm) on the colloid (see Section S2, Supporting Information) utilizing AFM for both scratching and cross-section analysis, we estimate the grafting density of the brush to be (0.39 + 0.01) nm⁻². The estimation of the grafting density using gravimetric analysis yields the value (0.58 + 0.20) nm⁻² (Section S2, Supporting Information). Both methods used in our study have several internal error sources, and can be used only to produce an estimate of the grafting density. In the case of the AFM method, the thickness measurements are atomically precise, but the measurements are conducted only on a small number of particles and are thus lacking in the statistics of the system. In the case of gravimetric analysis, a significant contribution to the measurement error is made by the unknown amount of water stored at the oxide surface of the plain particles and within the polyelectrolyte layer of the spherical brushes. However, both methods provide grafting density values that are reasonable for the ATRP process [56].

The supramolecular complex between the spherical brush and the oppositely charged cationic surfactant is formed when the particles are exposed to a surfactant solution for a certain loading time, similar to the situation with a planar poly(methacrylic acid) (PMAA) brush [46,47]. The surfactant molecules penetrate the spherical PMAA brush and form the complex by electrostatic and hydrophobic interactions, replacing the native counterions within the brush interior [57]. This can be observed using the fluorescent signal emitted from the spherical brush and not visible at the plain colloids (compare Figure 1b,d). Here, we make use of the complex formation between a negative charge of fluorescein dye and a cationic surfactant. This complex effectively penetrates into the PE layer, resulting in enhanced emission. Depending on the amount of the adsorbed surfactant, one can also achieve an overcharging of the spherical brush from negatively to positively charged at concentrations of ca. 0.3 mM (CMC_trans = 0.5 mM) as measured from the Z-potential (Figure S4, Supporting Information). These results are in good agreement with the previous study, where it is shown that the complexation of the CTAB surfactant with a poly(styrene sulfonate) spherical brush takes place much below CMC [20].

Since the surfactant is photo-responsive, one can trigger its desorption out of the brush using light. Indeed, when the spherical brushes are irradiated with blue (λ = 455 nm) or UV (λ = 365 nm) light (τ_irr = 10 min, I = 10 mW/cm²), the surface potential consecutively decreases for all surfactant concentrations, being more pronounced for UV light (Figure S4, Supporting Information). Similar behavior is also observed for charged microgels containing PAA chains and loaded with the azobenzene containing surfactant [58].

The light-triggered reversible adsorption/desorption of the surfactant results in the generation of the concentration gradient around the particles, which leads to a diffusioosmotic flow similar to the microgels (Figure 2) [44]. Here, SiO₂@PMAA are exposed to an aqueous solution of azobenzene containing surfactant (c_azo = 2 mM) and placed in a closed chamber in the optical microscope. Without irradiation (imaging proceeds under red light (λ = 625 nm), with no surfactant photo-isomerization), the sedimented particles move randomly near a surface. Depending on the surfactant concentration one can find the particles in different states: forming aggregates in case of charge compensation and as a single specimen at low and large amounts of surfactant in solution. The number of aggregates decreases with increasing surfactant concertation owing to the overcharging of the brush and electrostatic repulsion. When the irradiation with blue light (λ = 455 nm) is switched on, the particles start to repel each other until they reach an equilibrium position (Figure 3c,e), as in the case of porous particles [51]. During the whole irradiation time, the particles are essentially trapped at their position, with residual fluctuations due to Brownian motion. The separation distance can be as high as several diameters of the colloid.

After switching off the light, the long-range diffusioosmotic repulsion basically vanishes, and the particles move freely, undergoing thermal motion (Figure 2a,d,f). To quantify the extend of lateral repulsion, we calculate the nearest particle distance (NPD) as a function of the irradiation time, where NPD is defined as the smallest average distance to a neighboring particle (Figure 2a).

To investigate the flow strength as a function of applied intensity at a fixed wavelength (455 nm), we fit example data displayed in Figure 3a directly after switching the blue light with a single exponential decay function:

$$NPD=NP{D}_0+\left(NP{D}_{\mathrm{e}}-NP{D}_0\right)·\exp \left(-\Gamma ·t\right)$$

(2)

where NPD_e and NPD₀ are the nearest particle distance at the equilibrated and initial state, respectively, t the time, and Γ the characteristic separation rate. Doubling the intensity leads to an almost doubled separation rate, indicating that the DO is linearly proportional to the light intensity up to a critical intensity of 4 mW/cm² (Figure 3b). Beyond this intensity, the separation rate is effectively constant. Data measured for SiO₂@PMAA are compared with data for porous (P-SiO₂) and non-porous (SiO₂) silica colloids of similar size but otherwise unmodified. The porous particles have a pore size of 6 nm and thus a much larger surface area compared to plain colloids. This results in a large amount of surfactant being absorbed by porous particles compared to plain SiO₂ and spherical brushes (SiO₂@PMAA). The separation takes place within seconds or minutes, depending on the light intensity or the particle type (Figure 3b). All particles show the same trend; with increasing intensity, an increase in the separation can be observed in the following order: P-SiO₂ > SiO₂@PMAA > SiO₂, but without any pronounced velocity limitation at high intensities for SiO₂ and P-SiO₂ (Figure 3b).

We explain the formation of the local diffusioosmotic flow (l-LDDO) for PE-coated particles by the dissimilar photoisomerization rate of the surfactant inside the PE brush and in the bulk, which creates an excess of cis isomer concentration Δc_C near the colloids in comparison to the bulk solution. We have demonstrated that the DO flow velocity scales linearly with Δc_C(t), where the dependence of Δc_C on time can be estimated by solving the equation [51]:

$$\frac{d\Delta c_{\mathrm{C}}}{dt}~I·\left[k_{\mathrm{TC},\mathrm{I}}·c_{\mathrm{T},\mathrm{I}}-\left(k_{\mathrm{TC}}·c_{\mathrm{T},\mathrm{B}}-k_{\mathrm{CT}}·c_{\mathrm{C},\mathrm{B}}\right)\right]$$

(3)

where k_TC,I and k_TC are the rate constants of trans−cis isomerization inside the microgel and bulk solution, respectively, k_CT the rate constant of cis−trans isomerization in the bulk, c_T,I the trans isomer concentration at the interface, c_T,B and c_C,B the concentrations of trans and cis isomers in the bulk, and I the intensity of the light. Here, we interpret the PE brush layer as a diffusive interface capable of absorbing a certain amount of trans isomers, larger than in the case of the plain particles (SiO₂), but less than in the case of the porous colloids (P-SiO₂) (see Figure 4a, where the adsorbed amount of the surfactant is shown for three different particles) (for calculations, see Section 4, Supporting Information). When exposed to blue light, more surfactants are released (c_released ~ c_dark-c_blue) from the P-SiO₂ (c_released = 0.03 mM) than from the SiO₂@PMAA (c_released = 0.02 mM) and fewer from the plain particles (c_released = 2∙10⁻⁴ mM). Furthermore, normalizing the adsorbed surfactant for the P-SiO₂ and the SiO₂@PMAA against the SiO₂ shows that the P-SiO₂ absorbs ~33 and the SiO₂@PMAA ~12.5 times more surfactant in comparison to the SiO₂ (dark state, Figure 4b), but under blue illumination (455 nm) the ratio is reduced for the P-SiO₂ to ~29 and for the SiO₂@PMAA to ~9, where the normalized separation Γ/Γ_SiO2 (red data, Figure 4b) is congruent for the P-SiO₂ ~28.6 but not for the SiO₂@PMAA to be ~3 and therefore a factor 3 smaller than the value for adsorbed c_ads/c_ads,SiO2. Additionally, data in Figure 3b indicate a linear dependence between the DO flow strength and the surfactant storage capacity for porous silica particles (Equation (3)). For the SiO₂@PMAA, in principle, all counterions can be replaced by the surfactant molecules, which leads to a strong collapse of the polyelectrolyte brush [46,47]. Thus, technically, a high amount of surfactant molecules can be stored within the PE, and much faster separation compared to plain silica particles should be expected. However, the data yield only three times faster separation (relative particle separation in Figure 4b). Thus, what is causing the pronounced exchange limitation of the isomers even at a high surfactant storage capability?

This can be visualized by considering the exchange and isomerization kinetics of the isomers, which are in cross-correlation from three possible major interactions of the isomers with the PE brush. Indeed, first, the adsorption rate of the trans isomer to the polymer (shown on an example of p(NIPAM-MAA) microgels) is 22% slower than to plain silica interfaces (k_a,MG ~ 7 L∙(mol∙s)⁻¹ vs. k_a,glass = 9 L∙(mol∙s)⁻¹ [49,58]. Additionally, the photo-isomerization rate of the surfactant in the aggregate (micelle, restricted geometry) is lower in comparison to non-aggregated molecules owing to steric hindrance [59]. The isomerization rate reduction for blue light (455 nm) is reported for PE-surfactant complexes for single chains [49] (k_TC,PE = 3∙10⁻³ cm²∙(mW∙s)⁻¹) or charged microgels [58] (k_TC,MG = 2∙10⁻³ cm²∙(mW∙s)⁻¹) in comparison to the bulk solution value (k_TC,B = 8∙10⁻³ cm²∙(mW∙s)⁻¹), showing a reduction of [1 − (k_TC,PE /k_TC,B)] = 0.62. Note that the surfactant molecules in PE layer are in the aggregated state. Finally, a slower diffusivity of the isomers within the PE layer may contribute to the limitation of the isomer exchange. For instance, the generated cis isomers localized near the core (see Figure 4c) (1) have a longer diffusion distance across the PE layer than those located at the PE/solution interface (2). For the isomers at location (1), the diffusion time may be equal or larger compared to the rate of cis-to-trans isomerization thus hindering the cis isomer release out of the polymer interior. In comparison, at the position (2), the diffusion time is smaller and cis isomers have enough time to be released into the bulk solution to contribute to the local-LDDO flow. Note that the release of surfactant molecules will be compensated by adsorbing counterions, and a swelling of the brush may extend the length of the brush layer, thereby also increasing the diffusion time. These processes become more pronounced with increasing intensity and may explain the marked separation limitation beyond a critical intensity of 4 mW/cm^², which is similar to the recently reported value of 3 mW/cm^² for microgels, versus no separation limitation of plain and porous silica particles [51].

The above discussed points effectively reduce the dynamic exchange of isomers between the PE layer and the bulk; therefore, the DO flow ~ Δc_C does not scale linearly with surfactant storage capacity for the SiO₂@PMAA, while for the SiO₂ interfaces, one observes a linear relation, $\frac{d\Delta c_{\mathrm{C}}}{dt}~c_{\mathrm{T},\mathrm{I}}$.

The extent of the DO flow also depends on the surfactant concentration (Figure 3c) and ionic strength (Figure 3d). The dependence of the DO flow on salt concentration is explained by the scaling of the DO velocity with the squared value of the Debye length: as the ionic strength is increased, the EDL thickness decreases and the DO flow vanishes [52,60]. The surfactant concentration dependence can be understood from cross-correlating the dynamic exchange ability of trans and cis isomers. Assuming the simplest form of adsorption isotherm, the Langmuir relation [50], the rate of trans adsorption increases with increasing trans isomer concentration in bulk solution, while the rate of cis desorption is proportional to the amount of adsorbed cis isomers. Since a higher trans isomer adsorption rate yields a higher amount of adsorbed trans isomers at the interface, more surfactant molecules can be isomerized per time, and therefore more cis isomers are temporarily located at the interface and are quickly desorbed into the bulk solution. Thus, the dynamic exchange between trans and cis isomers is a function of the total surfactant concentration, where an increase in surfactant concentration yields an increase in the amount of exchanged isomers and ultimately a higher induced cis isomer gradient during light illumination. However, for the SiO₂@PMAA particles, the dynamic exchange ability is more complex than at the plain interface; as described above, a similar trend of flow-strength increase with surfactant concentration is observed (Figure 3c).

4. Conclusions

We report on remotely controlled long-range repulsive interactions between microparticles with a polyelectrolyte (PE) brush coating. The mechanism of DO repulsion is based on the generation of a local-light-driven diffusioosmotic flow, when spherical brushes are exposed to light of an appropriate wavelength. To trigger photoactivity in the spherical polyelectrolyte brushes, azobenzene containing surfactant is loaded within the PE layer. The PMAA brush readily absorbs cationic photo-sensitive surfactant in its more hydrophobic trans state and effectively releases more hydrophilic cis isomers, resulting in the permanent exchange of the isomers during illumination. The excess of the cis isomers near the SPB results in an osmotic pressure gradient and the subsequent formation of the diffusioosmotic flow pointing radially out of the particles. Thus, each spherical brush produces its local lateral diffusioosmotic flow, resulting in hydrodynamic interactions and mutual long-range repulsion. We have also thoroughly analyzed the behavior of the porous particles and spherical brushes in comparison to the plain silica particles exposed to the same irradiation and solution conditions. We have found that the amount of the adsorbed surfactant for porous particles and spherical brushes is ~33 and ~12.5 times larger than for the plain colloids, respectively. This is reflected also in a more effective generation of the local DO flow. Indeed, for the porous particles we measured a ca. 29 times larger normalized separation Γ/Γ_SiO2 under irradiation with blue light. However, in the case of the spherical brushes, the DO flow extent increases only by three times (in comparison to the plain particles). We explained this by the fact that the dynamic exchange of the isomers under irradiation with blue light is limited by a reduced trans isomer adsorption tendency together with a reduced trans–cis isomerization rate inside the brush; both processes are weaker in comparison to pure porous silica interfaces. This indicates that the DO flow is limited at a critical light intensity of 4 mW/cm². In summary, photosensitive surfactant is a simple and powerful tool to effectively separate polymer particles without applying any mechanical stresses. This might be an effective application for the separation of unwanted aggregation, which is an often-faced critical phenomenon in colloid and surface science.