Direct Thiol–Ene Photocoating of Polyorganosiloxane Microparticles

Dec 9, 2013 - Raman spectroscopy was applied to directly follow the ...... Asymmetrical Flow Field-Flow Fractionation Macromolecules 2001, 34, 8347–...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/Langmuir

Direct Thiol−Ene Photocoating of Polyorganosiloxane Microparticles Christian Kuttner,*,† Petra C. Maier,† Carmen Kunert,† Helmut Schlaad,‡ and Andreas Fery*,† †

Department of Physical Chemistry II, University of Bayreuth, Bayreuth 95440, Germany Department of Colloid Chemistry, Max Planck Institute of Colloids and Interfaces, Potsdam 14424, Germany



S Supporting Information *

ABSTRACT: This work presents the modification of polyorganosiloxane microparticles by surface-initiated thiol−ene photochemistry. By this photocoating, we prepared different core/shell particles with a polymeric shell within narrow size distributions (PDI = 0.041−0.12). As core particle, we used highly monodisperse spherical polyorganosiloxane particles prepared from (3-mercaptopropyl)trimethoxysilane (MPTMS) with a radius of 0.49 μm. We utilize the high surface coverage of mercaptopropyl functions to generate surface-localized radicals upon irradiation with UVA-light without additional photoinitiator. The continuous generation of radicals was followed by a dye degradation experiment (UV/vis spectroscopy). Surface-localized radicals were used as copolymer anchoring sites (“graftingonto” deposition of different PB-b-PS diblock copolymers) and polymerization initiators (“grafting-from” polymerization of PS). Photocoated particles were characterized for their morphology (SEM, TEM), size, and size distribution (DLS). For PS-coated particles, the polymer content (up to 24% in 24 h) was controlled by the polymerization time upon UVA exposure. The coating thickness was evaluated by thermogravimetric analysis (TGA) using a simple analytical core/shell model. Raman spectroscopy was applied to directly follow the time-dependent consumption of thiols by photoinitiation.



condensation of organosilanes in water.21,22 These hybrid silica particles are essentially polyorganosiloxanes, which inherit their functionality from the respective low-molecular-weight precursor. Polyorganosiloxanes are commonly used in a myriad of applications23 and for colloidal structures. By sequential condensation or combination of different hydrolizable organosilanes in aqueous dispersion in the presence of surfactant, amphiphilic polyorganosiloxane nanospheres24,25 or nanocapsules25−27 with different core/shell architectures can be synthesized. Jungmann et al. reported that the ionic part of such structures can be loaded with dye28 or employed as passive nanoreactors29 for the synthesis and stabilization of noble metal colloids. In this study, we show the modification of polyorganosiloxane particles prepared from (3-mercaptopropyl) trimethoxysilane (MPTMS) by surface-initiated thiol−ene photochemistry without additional photoinitiator. Both “grafting-onto” deposition of preformed polybutadiene-based copolymers (PB-bpolymer) and “grafting-from” polymerization are robust techniques for thin-film coatings.30−33 These complementary techniques were adapted to modify particles with polystyrene (PS and PB-b-PS) as a hydrophobic model polymer shell. For “grafting-from” polymerization, the thickness of the coating was controlled by the duration of the light exposure (up to 90 nm

INTRODUCTION Many technical applications utilize composite materials which require specific interfacial properties between their constituents for good performance.1,2 With regard to mechanical stability of composites, reinforcement of polymeric materials with silica particles (or fibers) as filler exhibits higher energy absorption during impact fracture.3,4 This would allow for increased toughness with negligible loss of tensile strength. Here, interfacial stability demands the surface of the filler to provide chemical compatibility and high adhesion. Adhesion largely depends on molecular interfacial structure as well as molecular interfacial interactions.5 The filler/matrix adhesion can be controlled by functionalization of the filler surface by physisorptive6 or chemisorptive processes. For the latter, low-molecular-weight modifications (e.g., by silane sizings)7 are common practice.8 Polymeric modifications are desirable for interface engineering9−14 but often avoided in applications due to higher costs and difficult up-scaling. Most polymer modifications require an initial lowmolecular-weight functionalization to introduce anchoring or initiation sites at the surface.4,15−18 Hence, the most frequently applied methodology for functional particles in the micrometer size range is the silylation of particles prepared by the Stöber method using tetraesters of silicic acid in alcohols.19 Recently, Fuchs and Avnir reported the preparation of polymer-modified silica particles by entrapment of amphiphilic copolymers within the forming silica particles.20 As an alternative to pure silica, Lee and co-workers presented an one-step preparation method of monodisperse hybrid silica particles based on the self© XXXX American Chemical Society

Received: October 14, 2013 Revised: December 5, 2013

A

dx.doi.org/10.1021/la4039864 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

laser (633 nm). Raman scattering was detected by a peltier-cooled CCD camera (−70 °C, Synapse) situated behind a 1800 grooves/mm grating spectrometer. The thickness of thin films on particles was calculated from thermogravimetric analysis (TGA) (N2 flow, heating rate 5 °C/min, TGA/SDTA 851e, Mettler Toledo). Dynamic light scattering (DLS) was performed using a goniometer setup (ALVLaservertriebsgesellschaft, Langen, Germany) with an Ar ion laser (532 nm, Coherent Verdi-V2) and with a HeNe laser (633 nm, JDS Uniphase 1145/P) as light source. The correlation function was generated using an ALV-5000/E multiple-τ digital correlator and afterward analyzed by the method of cumulants.35 Toluene served as an index matching bath. Measurements were performed at low particle concentrations at scattering angles between 30° and 110° and were repeated at least three times.

within 24 h). We evaluated the polymer content by thermogravimetric analysis (TGA) and applied a core/shell model to estimate the coating thickness on the particles. The resulting morphologies were characterized by electron microscopy techniques (SEM and TEM) and compared to uncoated particles. The generation of radicals during the UV initiation process was followed by dye degradation experiments (UV/vis spectroscopy).



EXPERIMENTAL PART

Materials. All chemicals (p. a. grade) were purchased from SigmaAldrich unless mentioned differently: styrene (99.9%), purified from stabilizers by filtration through basic alumina column; (3-mercaptopropyl) trimethoxysilane (MPTMS, 95%); p-(dimethylamino)azobenzene (dimethyl yellow, Fluka). Solvents, used as received: ethanol, toluene (99.8%), and cyclohexane (99.5%). Further reagents: ammonium hydroxide (25%, Fluka). Deionized water (DI) was obtained from a water purification system (Milli-Q Advantage A10, Millipore). The different polybutadiene-b-polystyrene (PBx-b-PS360, x = 43, 339, 663, and 854; subscripts denoting the average number of repeating units) diblock copolymers were synthesized via anionic polymerization in tetrahydrofuran (THF) at low temperatures in the presence of alkoxides according to standard procedures.31,34 Synthesis of Polyorganosiloxane Particles. A volume of 7.5 mL of MPTMS was added into 750 mL of water at 40 °C under vigorous stirring (KPG, 600 rpm, 1−2 h) until the turbid solution caused by oil droplets turned completely transparent. Then, ammonium hydroxide (NH4OH, 3 mL, 25 vol %) was added to the transparent mixture solution. After 5 min of stirring, the reaction progressed for 24 h without stirring at 40 °C. After completion of the reaction, the solution was diluted by large amounts of ethanol and allowed to cool down to room temperature. The resulting precipitate was collected by centrifugation (3000 rpm, 20 min, RT) and washed three times upon ultrasonication (5 min, 40 °C) using ethanol to remove nonreacted material. The obtained particles were stored in ethanol as suspension (15 mg/mL). Thiol−Ene Photocoating of Polyorganosiloxane Particles. Approximately 30 mg of particles (solids content of 2 mL ethanol suspension) were added to cyclohexane solutions containing 5 vol % of styrene (“grafting-from” method) or 2 wt % of PB-b-PS diblock copolymer (“grafting-onto” method), in closed vials (total volume 21 mL) under Ar atmosphere. Polymerizations were carried out upon irradiation with UVA light (Höhnle UV F 400F, 400 W, blue filter: 320 nm < λ < 420 nm; see Figure S1 in the Supporting Information) for 24 h for “grafting-onto” and for different durations (3, 6, 12, 18, 24 h) for “grafting-from”. The temperature was kept below 30 °C. Polymergrafted particles were sequentially centrifuged (see above) and washed with toluene upon ultrasonication. Dye Degradation by Radical Initiation. Approximately 30 mg of particles (solids content of 2 mL ethanol suspension) were added to a 1 M dye solution of p-(dimethylamino)azobenzene (dimethyl yellow) in toluene (total volume 18 mL). The solution was irradiated with UVA light (Höhnle UV F 400F, see above) for 24 h upon vigorous stirring. At given times, aliquots of 1 mL samples were taken from the particle solution, centrifuged, and the supernatant was analyzed by UV/vis spectroscopy (Lambda 19, Perkin-Elmer). A dye solution with identical composition without particles was used as reference. Before each sample measurement, the quartz cuvette was cleaned six times with toluene and a new baseline was determined. Characterization. Particles were imaged by transmission electron microscopy (TEM) at 80 kV (CEM 902, Zeiss). Deposited on cleaned silicon wafers and glass substrates, particles were investigated by scanning electron microscopy (SEM), by energy dispersive X-ray spectroscopy (EDX) (Leo1530, Zeiss), and by atomic force microscopy (AFM) in tapping mode (Dimension V, Veeco Metrology Group) with AC160TS-W2 cantilevers (300 kHz, 42 N/m) by Olympus. Raman spectra were acquired using a confocal Raman microscope (LabRAM Division, HORIBA Jobin Yvon) equipped with a lens from Olympus (50x, NA = 0.75) and a linear-polarized HeNe



RESULTS AND DISCUSSION Polyorganosiloxane Particles. The applied synthesis of polyorganosiloxane microparticles was based on a sol−gel condensation with MPTMS as precursor.21 In contrast to the Stö ber synthesis,19 the applied sol−gel process utilizes precursor molecules which are fully soluble in water. Therefore, the functional organosilane forms a homogeneous dispersion. We used MPTMS to prepare particles with a high surface coverage of thiol groups. Upon hydrolysis of its three methoxy groups MPTMS forms silanols of high reactivity. NH4OH was added as catalyst to increase the rate of nucleation, as previously reported by Oh et al.22 Formed nuclei grow to spherical particles until the available precursor is consumed. The result of the sol−gel process was a sedimented precipitate. After washing, this precipitate was found to consist of spherical microparticles (see Figure 1). The SEM images

Figure 1. Scanning electron micrographs of uncoated polyorganosiloxane microspheres.

show particles with a smooth surface without visual surface defects as confirmed by AFM topography (see Figure 2a). The size of the particles was evaluated by dynamic light scattering (DLS) using the method of cumulants35 with a hydrodynamic radius Rh of about 490 ± 75 nm depending on the dispersion medium. Figure 2b shows the decay rate Γ versus the squared scattering vector q2 for multiangle measurements in ethanol and toluene. The Rh in ethanol was slightly smaller than in toluene. At first glance, the linear succession of Γ (q2) suggests spherical objects of uniform shape as scatterers. For DLS, the size-dispersity B

dx.doi.org/10.1021/la4039864 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Figure 3. Raman spectrum of polyorganosiloxane microparticles prepared from MPTMS.

free from reactive silanols and exclusively exposes thiolterminated alkyls. Thiol−Ene Photocoating: UV-Initiation. Thiol−ene photocoating utilizes the formation of surface-located thiyl radicals by UVA-light in presence of olefins. Details on the mechanism of thiyl formation are provided in the Supporting Information. Contrary to common free-radical-transfer techniques, this approach does not require additional photoinitiator. The radical species is generated in solution but resides at the particle surface. Surface-localized radicals can serve as anchoring site for preformed solution-borne polymer (“grafting-onto”) or as polymerization initiators (“grafting-from”). As a first step, we evaluated the particles’ capability to form radicals. For this purpose, a dye degradation (decomposition) experiment was performed. Dye solutions with and without polyorganosiloxane particles were irradiated by UVA-light. Subsequently, samples were drawn from both solutions, centrifuged to separate the particles, and the supernatant was analyzed by UV/vis spectroscopy. This allowed for the comparison of the two solutions and to evaluate the influence of the particles on the rate of dye degradation. The applied dye was p-(dimethylamino)azobenzene (dimethyl yellow), a commonly used indicator in the low pH regime, which is stable upon irradiation of UVA-light. After 24 h of UVA irradiation, the molecular degradation (without particles) was below 10%. Figure 4 shows the radical degradation of dimethyl yellow in the presence of polyorganosiloxane particles. The absorbance decrease at 409 nm was used to follow the accelerated degradation of the dye as a result of the formation of radicals at the particle surface. The initial absorbance decrease (at 0 h) of approximately 25% is related to the adsorption of dye molecules to the surface of the particles (see Figure S2, Supporting Information).37 Any further absorbance decrease corresponds to a degradation (and decomposition) of the dye by the generation of thiyl radicals at the surface and the transfer of radicals to vicinal dye molecules. The decay was followed over a time frame of 24 h. The normalized maximum of the absorbance peak decreases almost linearly with time. The generation of new radicals proceeds continuously as long as irradiation takes place. Even though the total concentration of radicals increases progressively with time, the local concentration is kept low due to the spatial separation of the random distribution of active sites. “Grafting-onto” Copolymer Deposition. Thiol−ene photocoating was used to anchor polybutadiene (PB)-based diblock copolymers to polyorganosiloxane particles. Figure 5 shows TEM and SEM images for PB-b-PS-coated particles. After deposition onto flat substrates and evaporation of the

Figure 2. (a)Atomic force micrograph of uncoated polyorganosiloxane particles and (b) dynamic light scattering in ethanol (red) and toluene (green).

(PDI) is a dimensionless number calculated from a simple twoparameter-based cumulant analysis (Γ and μ2) and scaled such that values smaller than 0.05 are attributed to monodisperse particles.35,36 PDI ≡

μ2 Γ2

=

σR h 2 ⟨R h⟩2

(1)

where Γ is the first-order and μ2 is the second-order moment of the exponential description of the autocorrelation function, and σRh is the width of the Gaussian distribution. The PDI was evaluated as a mean of all measured angles. For both media, the PDI was about 0.048 indicating a highly monodisperse sizedistribution (see Figure S5, Supporting Information). Energy dispersive X-ray spectroscopy (EDX) was performed to assess the elementary composition at the particle surface (atom ratio: 49% C, 31.5% Si, 15.5% O, and 4% S). The high carbon content is according to expectations that the surface is occupied by the functional alkyls of MPTMS. The framework of the particles consists of silicon and oxygen (SiOx network). Since the content of sulfur was lower than expected, we performed Raman spectroscopy to evaluate the particle composition in more detail. Raman scattering allows for the discrimination of the functional groups at and close to the surface. Figure 3 shows the characteristic spectrum of polyorganosiloxane particles. For details on the vibrational assignments, see Table S1 in the Supporting Information. Additional to the signals of the SiOx framework, the alkyl chain can be identified. The thiol group shows a pronounced signal at 2566 cm−1. Interestingly, the spectrum does not contain a peak at 3420 cm−1, which could be attributed to free silanol groups. It was expected to detect unreacted silanols due to incomplete cross-linking inside the particles as a result of the steric hindrance by the alkyl chains. The absence of this signal indicates that the particle surface is C

dx.doi.org/10.1021/la4039864 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

that the copolymer was strongly attached to the particle surface. The morphology at the surface can be related to the adsorption pattern of anchor-buoy diblock copolymers, where one block (PB) forms an anchoring layer and the other block (PS) remains free in a diluted and extended state.31 The multiply attached anchor chains form a bottom layer at the surface. The top layer mainly consists of buoys, which in dry state do not form a smooth film but a rough assembly of collapsed domains (see Figure 5c, d). DLS was used to evaluate the size and size-distribution of the coated particles in good solvent (toluene, see Figures S6 and S7 in the Supporting Information). The size-distribution slightly increased but remained within a PDI range of 0.06−0.09, indicating a narrow size-distribution. The hydrodynamic radii of copolymer-coated particles determined by DLS in good solvent can be found in Table 1. In fact, Rh is a measure related to the Table 1. Hydrodynamic Radii of Diblock CopolymerPhotocoated Particles in Toluene Determined by Dynamic Light Scattering uncoated anchor block NPBa Rh, nmb measured increase, nmc calculated increase, nmd

495

PB-b-PS360-coated 43 583 88 94

339 599 104 110

663 617 122 123

854 631 136 130

a

Polymerization degree of anchor block. bHydrodynamic radii. Each value has an apparent accuracy of approximately ±6 nm, based on the linear fits of the diffusion coefficients. For details, see Figure S6 in the Supporting Information. cHydrodynamic radius increase (Rh 495 nm), compared to uncoated state. dTheoretical increase (Rh,coated 495 nm), using eq 2 with α = 0.75 (PB) and β =1 (PS).

Figure 4. Absorbance decay of dimethyl yellow dye in toluene: (a) UV/vis spectra for different times of UVA exposure, and (b) absorbance decay at 409 nm, normalized by reference spectra without microparticles.

diffusional properties of a hypothetical hard sphere with a surrounding solvation layer. For coated particles, larger Rh values were obtained in comparison to uncoated particles. The increase of Rh in good solvent can be correlated to the copolymer composition by a rough estimation: α β R h,coated ≈ R h,uncoated + (Nanchor + Nbuoy )l

(2)

where N is the degree of polymerization, l is the monomer unit length (≈0.25 nm), and the exponents α and β reflect the respective chain conformation. An exponent of 1 describes a terminal-attached fully stretched chain, whereas a value of 0.5 correlates to a coiled conformation of an ideal chain. In good solvent, it can be expected that the buoy chain (PS) is well solvated and stretches due to the steric repulsion by neighboring buoys (brush conformation). The anchor chain (PB) is multiply attached to the surface and to chain segments of the same or neighboring anchor blocks. Since the adsorption was performed in a nonselective solvent, where both blocks are well solvated, the anchors do not form collapsed globules but retain in conformations of larger volume. The anchoring network of inter- and intracross-linked chains can be expected to exhibit an exponent between 0.5 and 1. Even though this predication is only a rough estimate, it shows a scaling similar to the measured increase of Rh for PBx-b-PS360 coatings. Therefore, the size of the solvated core/shell particles can be tailored by the choice of the buoy-anchor copolymer. The outmost shell of the coated particles consists of a layer of buoy chains (PS), which could be substituted by a functional polymer block by use of the corresponding diblock copolymer (PB-b-polymer).

Figure 5. Diblock copolymer-photocoated microparticles: TEM of (a) PB43-b-PS360 and (b) PB663-b-PS360 coatings; SEM of (c) PB43-b-PS360 and (d) PB854-b-PS360 coatings.

solvent, thin connections were found between particles assemblies (see Figure 5a−c). These connections ranged up to several hundred nm in length and could be identified as copolymer of the coating. To verify that the visible copolymer between the particles was not physisorbed but surface-tethered, five additional washing cycles (centrifugation and ultrasonication in toluene, which is a good solvent for both blocks) were performed. The washing solution was free from extracted polymer. The morphology did not change, which concludes D

dx.doi.org/10.1021/la4039864 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Figure 6. Transmission electron micrographs of particles after photocoating of polystyrene (“grafting-from”) for increasing polymerization times.

Figure 7. Thermogravimetric analysis of polymer-coated microparticles: (a) Comparison of different polymerization times and uncoated polyorganosiloxane particles. The sample pure silica denotes a commercial batch (PSi-0.2 and PSi-0.8, Discovery Scientific Inc.) of standard TEOSbased silica particles as reference. (b) Increasing polymer content (mass loss) for longer polymerization times with estimated coating thickness (error bars in red, right axis).

“Grafting-from” Polymerization. As a complementary approach, thiol−ene photocoating was used for “grafting-from” polymerization on polyorganosiloxane particles. The formation of surface-localized radicals can be utilized to initiate the growth of polymer at the surface. To follow the photocoating, a series of samples with identical monomer content (5 vol % styrene) was prepared and characterized. Figure 6 shows TEM images of this series in dry state. Coated particles showed connections between particles, similar to the findings for the copolymer coatings. Again, the samples were exposed to five additional washing cycles (centrifugation and ultrasonication in toluene, which is a good solvent for PS) to verify that the polymer was tethered to the surface. Compared to uncoated particles, the modified particles look like hexagonal-like structures in the dry state. This shape is a drying effect and resulted from the soft shells being partly fused in the swollen state. Upon drying, the shells collapse onto the core while thin polymer connections remain at the contact points between the closely packed particles. With increasing polymerization time these features become more pronounced. SEM images show similar morphologies and an increased surface roughness compared

to uncoated particles (see Figure S4 in the Supporting Information). To evaluate the apparent size of the polymer-coated particles, DLS in good solvent (toluene) was measured. The mean hydrodynamic radii were in a narrow range of 637 ± 6 nm (see Figure S8 in the Supporting Information). A tendency toward higher PDI values for longer polymerization times was found, but a PDI < 0.13 still indicates narrow distributions (see Figure S9 in the Supporting Information). Also, longer polymerizations (18 and 24 h) showed slightly smaller Rh values within a deviation of 11 nm. This apparent contraction could be a result of an increased chain density in the coating layer. Based on the initiation method, the present polymerization of surface-tethered chains is not an instantaneous event like in classical free-radical polymerizations with initiation by radicaltransfer from the solution. As shown by the dye degradation experiment, the UVA-light induced formation of thiyl radicals proceeds sporadically and continuously. After initiation and upon chain growth, these active sites are translocated away from the surface reducing the chance for termination of newly formed radicals at the surface. Ideally, this would allow for the coexistence of many active chains at high spatial density and E

dx.doi.org/10.1021/la4039864 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Table 2. Coating Thickness and Polymer/Thiol Content of PS-Photocoated Polyorganosiloxane Particles polymerization time, h

0

3

6

12

18

24

polymer content wshell, %a coating thickness t, nmb relative PS content, %c relative thiol content, %c

0 0 0 100

11.5 41 ± 7 37.9 41.4

18.0 65 ± 10 49.8 28.3

21.7 78 ± 13 78.5 6.4

23.6 85 ± 14 97.1 2.3

24 86 ± 14 100 1.3

Mass loss difference (TGA) of sample to uncoated polyorganosiloxane particles with an apparent accuracy of approximately ±0.5%. For details, see Figure S10 in the Supporting Information. bCalculated using eq 4 with a core particles radius of r = 484 nm (DLS, Rh in ethanol). cDetermined by Raman spectroscopy; see Figure 8. a

wshell. The following equation was used to calculate the coating thickness of polystyrene on polyorganosiloxane particles.

their simultaneous unperturbed growth. It can be expected that both grafting density and mean chain length increase successively with time. Depending on the time of initiation, the resulting chains may exhibit a high dispersity in length. Even though thiol−ene photocoating is generally a free-radical process, the confinement of radicals to the surface enables the deposition to proceed in a controlled manner. Thermogravimetric analysis (TGA) was done to quantify the amount of covalently attached polymer at the particle surface. Therefore, we compared the thermogravimetric behavior of polymer-coated to uncoated polyorganosiloxane particles. As a result of the high carbon content (as evaluated by EDX), polyorganosiloxane particles behave differently from regular pure silica particles upon heating.4,38,39 Pure silica particles exhibit high temperature stability with a mass loss of below 5% up to 800 °C (see Figure 7a). Polyorganosiloxane particles are temperature-labile and start to decompose at 240 °C. At 800 °C a total mass loss of about 45% was detected, which is related to the organic-based fraction of the particles. The more stable SiO x framework accounts for the remaining mass. In comparison, polymer-coated particles started to lose mass already at 200 °C. Since all “grafting-from” samples were prepared from the same batch of uncoated polyorganosiloxane particles, a higher mass loss is directly correlated with a higher amount of coated polymer. From the differences in mass loss, the polymer content of the respective sample was determined. Figure 7b presents the polymer content of particles modified by “grafting-from” of polystyrene without additional photoinitiator. After 24 h of photocoating, the particles showed an effective polymer content of 24%. The coating thickness t, that is, the thickness of the coating if all chains collapsed onto the core particle to reach the bulk polymer density, can be calculated based on a simple core/shell model (for derivation, see the Supporting Information). ⎛ ⎞ wshell t = ⎜⎜ 3 1 + kρ − 1⎟⎟r 1 − wshell ⎝ ⎠

with

kρ =

t ≈ 0.741rwshell for kρ =

(4)

The resulting coating thicknesses are included in Figure 7b and listed in Table 2. Even though this approach is based on an ideal model and assuming a shell to core density ratio of 2:1, it is useful to represent the trend of the thickness increase for longer polymerization times. The true thickness might deviate from the calculated values, since the particles density is an estimation (ρcore ≈ 2 g/cm3). The correlation of DLS and TGA shows that the polymer content increased with longer polymerization times, while the hydrodynamic radius stayed almost constant throughout the process. Therefore, the effective polymer density of the shell has to increase throughout the time of photocoating, which is according to the nature of the continuous initiation process. In order to further study the “grafting-from” process, we conducted Raman spectroscopy of the polystyrene-coated particles. Figure 8a presents the frequency range of 2400− 3200 cm−1, while the full spectra can be found in Figure S11 in the Supporting Information. The spectra show the additional signals, which can be expected for a polystyrene modification (vibrational assignments, see Table S1 in the Supporting Information). To allow for a direct comparison, all spectra were normalized for the area under the vibrations of aliphatic CH2 at 2800−3000 cm−1 (see Figure S12, Supporting Information). From the normalized spectra, the relative contents of thiol (area ratio of ν(SH) at 2566 cm−1 to aliphatic ν(CH2)) and polystyrene (area ratio of aromatic ν(CH) at 3040 and 3053 cm−1 to aliphatic ν(CH2)) can be evaluated. Figure 8b shows the time-dependent consumption of thiols and coinciding formation of polystyrene. After the first 3 h, the thiol content is decreased to 41.4% and further decreases to 1.3% after 24 h. The rate of photoinitation decreases over time since it depends on the concentration of available thiols. As a consequence thereof, the generation of new growing chains also distinctively slows down. After 18 h of polymerization, neither the coating thickness (see Figure 7b) nor the polystyrene content (see Figure 8b) show significant increases even after additional 6 h of UVA irradiation. This is a clear consequence of the low residual concentration of thiols at the particle surfaces. From this, it can be concluded that the surface chain density (i.e., grafting density) increases throughout the process, as suggested by the proposed initiation mechanism. For the applied UVA-mediated photoinitiation, an apparent thiol half-life of 2.7 ± 0.4 h can be calculated from the exponential decrease in Figure 8b. This is much faster than expected from the degradation of dimethyl yellow (9.5 ± 1.5 h). The difference of the initiation rate could be related to steric reasons or the UVA absorption efficiencies of the formed

ρcore ρshell

2 1

(3)

Here, wshell is the polymer content (shell) evaluated from the TGA mass loss, kρ is the density ratio, ρ is the bulk density of polymer (shell) or the polyorganosiloxane (core), and r is the radius of the uncoated core particles (DLS, 484 nm in ethanol). The bulk density ratio kρ is an important scaling factor in this calculation, defining the core/shell system (polymer on polymer particle kρ ≈ 1, polymer on polyorganosiloxane particle kρ ≈ 2, and polymer on pure silica particle kρ ≈ 2.5). The scaling behavior of the relative coating thickness t/r and the polymer content (mass loss wshell) is shown in Figure S10 in the Supporting Information. Below a polymer content of approximately 30% a simplified estimation can be applied. Here, the t/r ratio increases linearly with the polymer content F

dx.doi.org/10.1021/la4039864 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

using Raman spectroscopy, it was possible to show the successive increase of the relative polystyrene content toward longer polymerization times. Furthermore, we used this approach to directly follow the time-dependent consumption of thiols by the UVA-mediated photoinitiation. In respect to thiol−ene photocoating, both “grafting-onto” copolymer deposition and “grafting-from” polymerization are complementary techniques which allow for the production of polymer-modified microparticles of narrow size-distributions. Both techniques can be adapted to functional copolymers (PBb-polymer) and a broad range of monomers. Therefore, we believe that the presented photocoating method embodies a rapid and versatile pathway to modify polyorganosiloxane surfaces and to prepare tailored polymer-modified core/shell micro- and nanoparticles. Such particles could find application as colloidal filler for composites or as colloidal probes for adhesion/force measurements by AFM.41−46



ASSOCIATED CONTENT

S Supporting Information *

Applied UV irradiation; dye adsorption; SEM of PS-coated particles; DLS data; calculation of coating thickness; simplified estimation; core/shell model for TGA; Raman data; normalization procedure; vibrational assignments. This material is available free of charge via the Internet at http://pubs.acs.org.



Figure 8. Raman analysis of polystyrene-coated particles: (a) spectra of different polymerization times, normalized for the area under the vibrations of aliphatic CH2 at 2800 to 3000 cm−1; (b) time-dependent changes of the relative contents of thiol and polystyrene, calculated from the area ratios of the observed vibrational bands.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

charge-transfer complexes of thiol and unsaturated species (e.g., ene or azo compound; see Supporting Information for details on thiyl formation). This could be a suitable starting point for further studies on the mechanism of UVA photoinitiation.

Notes

The authors declare no competing financial interest.





ACKNOWLEDGMENTS Financial support was given by the European Research Council in the framework of Starting Grant #306686 METAMECH: Template-assisted assembly of METAmaterials using MECHanical instabilities. The authors thank Dr. Andreas Hanisch for the synthesis and characterization of the applied diblock copolymers and Silke Nemeth for help with the UV/vismonitored dye degradation experiments. They are also thankful to Ute Kuhn for help with TGA measurements. C.K. and P.C.M. thank Katja von Nessen for her support with DLS. C.K. appreciates fruitful discussions with Dr. Wolfgang Häfner and Moritz Tebbe.

CONCLUSION In conclusion, we reported the modification of polyorganosiloxane particles by thiol−ene photocoating. Highly monodisperse microparticles particles prepared from MPTMS were characterized for size and size-dispersity (DLS), morphology (SEM), elementary composition (EDX), and functional groups at the surface (Raman spectroscopy). The available thiols at the surface were utilized to generate surface-localized radicals upon irradiation with UVA-light without additional photoinitiator. The particles’ capability to continuously form radicals was evaluated by a dye degradation experiment. These radicals were used as copolymer anchoring sites and polymerization initiators to generate a hydrophobic shell of polystyrene around the polyorganosiloxane core. The “grafting-onto” deposition of anchor-buoy diblock copolymers was shown for a series of PB-b-PS. The resulting morphology at the particle surface was according to anchorbuoy grafted31 and adsorbed films.40 The hydrodynamic radii of the copolymer-coated particles were correlated to the composition of the respective copolymers. The “grafting-from” polymerization resulted in closed polystyrene shells, while the coating thickness was controlled by the polymerization time. After 24 h, a polymer content of 24% was achieved, which was correlated to a coating thickness of approximately 90 nm by use of a simple core/shell model. By



REFERENCES

(1) Brondsted, P.; Lilholt, H.; Lystrup, A. Composite Materials for Wind Power Turbine Blades. Annu. Rev. Mater. Res. 2005, 35, 505− 538. (2) Corum, J.; Battiste, R.; Ruggles, M.; Ren, W. Durability-Based Design Criteria for a Chopped-Glass-Fiber Automotive Structural Composite. Compos. Sci. Technol. 2001, 61, 1083−1095. (3) Sun, L.; Gibson, R. F.; Gordaninejad, F.; Suhr, J. Energy Absorption Capability of Nanocomposites: a Review. Compos. Sci. Technol. 2009, 69, 2392−2409. (4) Zheng, J.-Z.; Zhou, X.-P.; Xie, X.-L.; Mai, Y. W. Silica Hybrid Particles with Nanometre Polymer Shells and Their Influence on the Toughening of Polypropylene. Nanoscale 2010, 2, 2269−2274.

G

dx.doi.org/10.1021/la4039864 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

(5) Zhang, C.; Hankett, J.; Chen, Z. Molecular Level Understanding of Adhesion Mechanisms at the Epoxy/Polymer Interfaces. ACS Appl. Mater. Interfaces 2012, 4, 3730−3737. (6) Gröschel, A. H.; Löbling, T. I.; Petrov, P. D.; Müllner, M.; Kuttner, C.; Wieberger, F.; Müller, A. H. E. Janus Micelles as Effective Supracolloidal Dispersants for Carbon Nanotubes. Angew. Chem., Int. Ed. 2013, 52, 3602−3606. (7) Drown, E.; Almoussawi, H.; Drazal, L. Glass-Fiber Sizings and Their Role in Fiber Matrix Adhesion. J. Adhes. Sci. Technol. 1991, 5, 865−881. (8) Kulkarni, S. A.; Ogale, S. B.; Vijayamohanan, K. P. Tuning the Hydrophobic Properties of Silica Particles by Surface Silanization Using Mixed Self-Assembled Monolayers. J. Colloid Interface Sci. 2008, 318, 372−379. (9) Brown, H. R. Chain Pullout and Mobility Effects in Friction and Lubrication. Science 1994, 263, 1411−1413. (10) Maguire, J. F.; Talley, P. L.; Lupkowski, M. The Interphase in Adhesion: Bridging the Gap. J. Adhes. 1994, 45, 269−290. (11) Cole, P. J.; Cook, R. F.; Macosko, C. W. Adhesion Between Immiscible Polymers Correlated with Interfacial Entanglements. Macromolecules 2003, 36, 2808−2815. (12) Fratzl, P.; Burgert, I.; Gupta, H. On the Role of Interface Polymers for the Mechanics of Natural Polymeric Composites. Phys. Chem. Chem. Phys. 2004, 6, 5575−5579. (13) Zou, H.; Wu, S.; Shen, J. Polymer/Silica Nanocomposites: Preparation, Characterization, Properties, and Applications. Chem. Rev. 2008, 108, 3893−3957. (14) Ndoro, T. V. M; Boehm, M. C.; Mueller-Plathe, F. Interface and Interphase Dynamics of Polystyrene Chains Near Grafted and Ungrafted Silica Nanoparticles. Macromolecules 2012, 45, 171−179. (15) Perruchot, C.; Khan, M. A.; Kamitsi, A.; Armes, S. P.; von Werne, T.; Patten, T. E. Synthesis of Well-Defined, Polymer-Grafted Silica Particles by Aqueous ATRP. Langmuir 2001, 17, 4479−4481. (16) Kim, S.; Kim, E.; Kim, S.; Kim, W. Surface Modification of Silica Nanoparticles by UV-Induced Graft Polymerization of Methyl Methacrylate. J. Colloid Interface Sci. 2005, 292, 93−98. (17) Tsukagoshi, T.; Kondo, Y.; Yoshino, N. Protein Adsorption on Polymer-Modified Silica Particle Surface. Colloids Surf., B 2007, 54, 101−107. (18) Chen, J.-J.; Struk, K. N.; Brennan, A. B. Surface Modification of Silicate Glass Using 3-(Mercaptopropyl)Trimethoxysilane for ThiolEne Polymerization. Langmuir 2011, 27, 13754−13761. (19) Stö ber, W.; Fink, A.; Bohn, E. Controlled Growth of Monodisperse Silica Spheres in Micron Size Range. J. Colloid Interface Sci. 1968, 26, 62−69. (20) Fuchs, I.; Avnir, D. Induction of Amphiphilicity in Polymer at Silica Particles: Ceramic Surfactants. Langmuir 2013, 29, 2835−2842. (21) Lee, Y.-G.; Park, J.-H.; Oh, C.; Oh, S.-G.; Kim, Y. C. Preparation of Highly Monodispersed Hybrid Silica Spheres Using a One-Step Sol−Gel Reaction in Aqueous Solution. Langmuir 2007, 23, 10875− 10878. (22) Oh, C.; Shim, S.-B.; Lee, Y.-G.; Oh, S.-G. Effects of the Concentrations of Precursor and Catalyst on the Formation of Monodisperse Silica Particles in Sol-Gel Reaction. Mater. Res. Bull. 2011, 46, 2064−2069. (23) Cole, M. A.; Bowman, C. N. Evaluation of Thiol-Ene Click Chemistry in Functionalized Polysiloxanes. J. Polym. Sci., Part A: Polym. Chem. 2013, 51, 1749−1757. (24) Jungmann, N.; Schmidt, M.; Maskos, M. Characterization of Polyorganosiloxane Nanoparticles in Aqueous Dispersion by Asymmetrical Flow Field-Flow Fractionation. Macromolecules 2001, 34, 8347−8353. (25) Jungmann, N.; Schmidt, M.; Maskos, M.; Weis, J.; Ebenhoch, J. Synthesis of Amphiphilic Poly(Organosiloxane) Nanospheres with Different Core-Shell Architectures. Macromolecules 2002, 35, 6851− 6857. (26) Shukla, A.; Degen, P.; Rehage, H. Synthesis and Characterization of Monodisperse Poly(Organosiloxane) Nanocapsules with or Without Magnetic Cores. J. Am. Chem. Soc. 2007, 129, 8056−8057.

(27) Qiao, S. Z.; Lin, C. X.; Jin, Y.; Li, Z.; Yan, Z.; Hao, Z.; Huang, Y.; Lu, G. Q. M. Surface-Functionalized Periodic Mesoporous Organosilica Hollow Spheres. J. Phys. Chem. C 2009, 113, 8673−8682. (28) Jungmann, N.; Schmidt, M.; Ebenhoch, J.; Weis, J.; Maskos, M. Dye Loading of Amphiphilic Poly(Organosiloxane) Nanoparticles. Angew. Chem., Int. Ed. 2003, 42, 1714−1717. (29) Jungmann, N.; Schmidt, M.; Maskos, M. Amphiphilic Poly(Organosiloxane) Nanospheres as Nanoreactors for the Synthesis of Topologically Trapped Gold, Silver, and Palladium Colloids. Macromolecules 2003, 36, 3974−3979. (30) Kuttner, C.; Tebbe, M.; Schlaad, H.; Burgert, I.; Fery, A. Photochemical Synthesis of Polymeric Fiber Coatings and Their Embedding in Matrix Material: Morphology and Nanomechanical Properties at the Fiber-Matrix Interface. ACS Appl. Mater. Interfaces 2012, 4, 3485−3492. (31) Kuttner, C.; Hanisch, A.; Schmalz, H.; Eder, M.; Schlaad, H.; Burgert, I.; Fery, A. Influence of the Polymeric Interphase Design on the Interfacial Properties of (Fiber-Reinforced) Composites. ACS Appl. Mater. Interfaces 2013, 5, 2469−2478. (32) Bertin, A.; Schlaad, H. Mild and Versatile (Bio-)Functionalization of Glass Surfaces via Thiol-Ene Photochemistry. Chem. Mater. 2009, 21, 5698−5700. (33) Wickard, T. D.; Nelsen, E.; Madaan, N.; ten Brummelhuis, N.; Diehl, C.; Schlaad, H.; Davis, R. C.; Linford, M. R. Attachment of Polybutadienes to Hydrogen-Terminated Silicon and Post-Derivatization of the Adsorbed Species. Langmuir 2010, 26, 1923−1928. (34) Schacher, F. H.; Yuan, J.; Schoberth, H. G.; Mueller, A. H. E. Synthesis, Characterization, and Bulk Crosslinking of Polybutadieneblock-poly(2-vinyl pyridine)-block-poly(tert-butyl methacrylate) Block Terpolymers. Polymer 2010, 51, 2021−2032. (35) Koppel, D. E. Analysis of Macromolecular Polydispersity in Intensity Correlation Spectroscopy - Method of Cumulants. J. Chem. Phys. 1972, 57, 4814−4820. (36) Gun’ko, V. M.; Klyueva, A. V.; Levchuk, Y. N.; Leboda, R. Photon Correlation Spectroscopy Investigations of Proteins. Adv. Colloid Interface Sci. 2003, 105, 201−328. (37) Tonle, I. K.; Ngameni, E.; Tcheumi, H. L.; Tchieda, V.; Carteret, C.; Walcarius, A. Sorption of Methylene Blue on an Organoclay Bearing Thiol Groups and Application to Electrochemical Sensing of the Dye. Talanta 2008, 74, 489−497. (38) Nassar, E. J.; Nassor, E. C.; de, O.; Avila, L. R.; Pereira, P. F. S.; Cestari, A.; Luz, L. M.; Ciuffi, K. J.; Calefi, P. S. Spherical Hybrid Silica Particles Modified by Methacrylate Groups. J. Sol-Gel Sci. Technol. 2007, 43, 21−26. (39) Meng, Z.; Xue, C.; Zhang, Q.; Yu, X.; Xi, K.; Jia, X. Preparation of Highly Monodisperse Hybrid Silica Nanospheres Using a One-Step Emulsion Reaction in Aqueous Solution. Langmuir 2009, 25, 7879− 7883. (40) Marques, C. M.; Joanny, J. F. Block Copolymer Adsorption in a Nonselective Solvent. Macromolecules 1989, 22, 1454−1458. (41) Kappl, M.; Butt, H. The Colloidal Probe Technique and Its Application to Adhesion Force Measurements. Part. Part. Syst. Charact. 2002, 19, 129−143. (42) Erath, J.; Schmidt, S.; Fery, A. Characterization of Adhesion Phenomena and Contact of Surfaces by Soft Colloidal Probe AFM. Soft Matter 2010, 6, 1432−1437. (43) Erath, J.; Cui, J.; Schmid, J.; Kappl, M.; Campo, A. D.; Fery, A. Phototunable Surface Interactions. Langmuir 2013, 29, 12138−12144. (44) Pretzl, M.; Neubauer, M.; Tekaat, M.; Kunert, C.; Kuttner, C.; Leon, G.; Berthier, D.; Erni, P.; Ouali, L.; Fery, A. Formation and Mechanical Characterization of Aminoplast Core/Shell Microcapsules. ACS Appl. Mater. Interfaces 2012, 4, 2940−2948. (45) Poehlmann, M.; Grishenkov, D.; Kothapalli, S. V. V. N.; Härmark, J.; Hebert, H.; Philipp, A.; Hoeller, R.; Seuss, M.; Kuttner, C.; Margheritelli, S.; Paradossi, G.; Fery, A. On the Interplay of Shell Structure with Low and High-Fequency Mechanics of Multifunctional Magnetic Microbubbles. Soft Matter 2014, 10, 214−226. (46) Kuznetsov, V.; Papastavrou, G. Adhesion of Colloidal Particles on Modified Electrodes. Langmuir 2012, 28, 16567−16579. H

dx.doi.org/10.1021/la4039864 | Langmuir XXXX, XXX, XXX−XXX