2750
Langmuir 2005, 21, 2750-2753
Distribution of CdSe Quantum Dots within Swollen Polystyrene Microgel Particles Using Confocal Microscopy Melanie Bradley,* Natasha Bruno, and Brian Vincent School of Chemistry, University of Bristol, Cantock’s Close, Bristol BS8 1TS, U.K. Received November 1, 2004. In Final Form: January 28, 2005 CdSe quantum dots (QDs) are semiconducting nanoparticles that fluoresce when stimulated by visible light. This property has been exploited in their use as tracer particles in biomedical applications. In this study, confocal microscopy has been used to determine the distribution of QDs within polystyrene microgel particles, dispersed in an organic solvent. It was found that the extent of microgel swelling affected the penetration of the QDs into the particles. Only when the microgel particles were swollen to their maximum extent were the QDs able to penetrate into the central core region of the particles.
Introduction Microgel particles are widely studied because of their potential applications in such areas as controlled delivery systems, catalyst supports, and pollution control. Such particles swell when placed in a good solvent for the polymer network. The extent of swelling is controlled by the amount of cross-linking monomer present in the microgel particles. Much of the research on microgel particles has focused on their physical properties, such as the variation of their size with changing environmental conditions.1 Far fewer studies have been conducted to fully characterize the physical structure of microgel particles, in particular, the pore-size distribution. By choosing nanoparticles of different sizes, their differing extents of penetration into the microgel polymer network may be explored. There are also practical reasons for studying the penetration of nanoparticles into microgel particles. For example, previous studies of the encapsulation of differentsized quantum dots (QDs) in microgel particles resulted in the development of multiplexed optical coders for use in biological imaging.2,3 In that work, the embedding process involved swelling the particles in a suitable solvent mixture containing the QDs. A second approach, aimed at incorporating QDs into polymer particles was based on infiltrating polystyrene (PS) beads, containing extensive networks of mesopores, with QDs. This technique is dependent on the hydrophobic attraction between the QDs and channel walls in the PS particle. The QDs penetrate into the pores by diffusion.4 Another route for the encapsulation of QDs within polymer particles has been to polymerize emulsion droplets of monomer containing dispersed QDs.5 QDs have also been covalently bonded into polymer films and particles; this latter technique involves precoating the QDs with polymerizable ligands.6,7 * To whom correspondence should be addressed. E-mail:
[email protected]. (1) Saunders: B. R.; Vincent, B. Adv. Colloid Interface Sci. 1999, 80, 1. (2) Han, M.; Gao, X.; Su, J. Z.; Nie, S. Nat. Biotechnol. 2001, 19, 631. (3) Rosenthal, S. J. Nat. Biotechnol. 2001, 19, 621. (4) Gao, X.; Nie, S. Anal. Chem. 2004, 76, 2406. (5) Yang, X.; Zhang, Y. Langmuir, in press. (6) O’Brien, P.; Cummins, S. S.; Darcy, D.; Dearden, A.; Masala, O.; Pickett, N. L.; Ryley, S.; Sutherland, A. J. Chem. Commun. 2003, 2532. (7) Hirai, T.; Saito, T.; Komasawa, I. J. Phys. Chem. B 2001, 105, 9711.
The benefit of using a porous host structure, over a nonporous particle, is a greater penetration of the QDs into the particles and an associated increase in the fluorescence intensity, compared to systems where the QDs are effectively just adsorbed in the monolayer around the particles. In this work, we aim to extend the scope of this earlier work in an attempt to determine the actual distribution of cadmium selenide (CdSe) QDs dispersed in polystyrene microgel particles. Two methods were used to produce these particles. The first was by swelling PS microgel particles in a suitable solvent mixture containing the QDs. The second method involved the direct polymerization of styrene droplets, containing dispersed CdSe nanoparticles. The effect of the swelling properties of the PS microgel particles on the distribution of CdSe was investigated using confocal microscopy. Experimental Section Chemicals. Selenium powder in tri-n-octylphosphine (TOP), cadmium acetate in TOP, tri-n-octylphosphine oxide (TOPO), hexadecylamine (HDA), toluene, chloroform, and propanol (all ex Aldrich) were used as received. Poly(vinyl alcohol) (PVA), with an average molar mass of 115 kg mol -1 and a degree of hydrolysis of ∼87% (BDH), and benzoyl peroxide (both ex Fluka) were also used as received. Nitrogen (oxygen free) was supplied by BOC Gases. Styrene and divinylbenzene (ex Aldrich) were purified, prior to use, by distillation. Dr Paul Mulvaney (University of Melbourne) kindly supplied particles of CdSe capped with ZnS. Synthesis of Polystyrene Microgel Particles. Polystyrene (PS) microgel particles were prepared by suspension polymerization. The aqueous phase and the oil phase were prepared separately. A 0.5 wt % aqueous solution of PVA was prepared. The oil phase was comprised of 10 wt % styrene, 0.1-0.5 wt % divinylbenzene (DVB), and 0.1 wt % of a thermally activated initiator, benzoyl peroxide. The oil phase and the aqueous, continuous phase were emulsified together using a Silverson L4RT high-speed shear mixer, operating at 6500 rpm for 15 min. The polymerization reaction was then carried out in a sealed reaction vessel at 70 °C, under an atmosphere of N2, with constant stirring, for ∼24 h. Synthesis of CdSe Quantum Dots. The preparation of CdSe quantum dots has been described previously in the literature.8 To this end, two precursor solutions were first prepared: (i) selenium powder in TOP and (ii) cadmium acetate in TOP. Trin-octylphosphine oxide (TOPO), hexadecylamine (HDA), and the Se/TOP solution were heated to 280 °C, under argon. The Cd/ (8) Mekis, I.; Talapin, D. V.; Kornowski, A.; Haase, M.; Weller, H. J. Phys. Chem. B 2003, 107, 7454.
10.1021/la047322r CCC: $30.25 © 2005 American Chemical Society Published on Web 03/05/2005
Distribution of CdSe QDs within PS Microgel Particles
Langmuir, Vol. 21, No. 7, 2005 2751
Figure 1. 1 wt % cross-linked PS microgel particles in solvent mixtures containing (a) 100% propanol, (b) 5% chloroform and 95% propanol, (c) 60% chloroform and 40% propanol, and (d) 100% chloroform (vol/vol). TOP solution was then injected at 250 °C, over ∼1 min, resulting in the growth of the CdSe nanoparticles. Adding cold toluene quenched further growth of the particles. Precipitation and several washings in methanol resulted in purified CdSe particles. Penetration of CdSe Nanoparticles into PS Microgel Particles. The CdSe QDs were dispersed in chloroform. Freezedried microgel particles (0.005 g) were redispersed in propanol (5 mL total volume) and a chosen composition of chloroform (containing the QDs). The dispersions were left for several hours to allow the penetration of the nanoparticles into the microgel particles to occur. The mixed dispersion was then gently heated to evaporate the chloroform. Microscope slides of the particles were then prepared. Direct Polymerization of CdSe into PS Microgel Particles. Dried CdSe QDs were dispersed in styrene monomer containing 1 wt % DVB before the monomer emulsion was prepared. The oil phase appeared orange as a result of the presence of the QDs. Suspension polymerization was then conducted, as outlined above. Characterization. A Nikon Optiphot optical microscope, fitted with Nikon 320 and 340 objective lenses, was used to observe and measure the change in PS microgel diameter resulting from solvent (chloroform) swelling of the particles. The length scale on the microscope was calibrated using a 50 × 2 µm graticule. Images were collected using the program Euresys MultiCam (Easygrab 2.6). Ultraviolet-visible (UV-vis) absorption spectra of the CdSe quantum dots dispersed in chloroform were obtained at 20 °C using a Hewlett-Packard 8453 spectrometer. Transmission electron microscopy (TEM) images were obtained using a JEOL Jem 1200 Ex. Confocal microscopy and ultramicrotomy were used to view the CdSe QDs within the PS microgel particles. Confocal microscope images (Carl Zeiss LSM-Pascal) of the fluorescence from the QDs dispersed throughout the microgel particles were collected using a He-Ne laser (λ ) 543 nm) as the excitation source. Confocal microscopy was used because it is a nondestructive, high-resolution technique that can be used to image optical sections that can be accumulated to produce a depth profile through a particle. Samples of PS microgel particles including quantum dots within their matrix were prepared for ultramicrotomy by setting them in Spurr’s resin.9 Thin sectioning was performed with an MTXL ultramicrotome using a diamond knife. Images of the sections were obtained using TEM.
Results and Discussion Swelling of PS Microgel Particles in Solvent Mixtures. Optical microscopy was used to follow the swelling properties of the cross-linked PS microgels in mixtures of propanol (a poor solvent for PS) and chloroform (a good solvent for PS). The results for the 1 wt % crosslinked particles are shown in Figure 1, as a function of increasing chloroform concentration. As the particle swells, its net refractive index decreases, becoming closer to that of the solvent. This results in less light scattering, and hence, less contrast between the particles and the solvent and the optical microscope images clearly illustrate this (Figure 1). (9) Spurr J. Ultrastruct. Res. 1969, 26, 31.
Figure 2. (a) TEM image of the dried CdSe quantum dots. (b) Absorbance spectra of the CdSe quantum dots in chloroform.
The extent of particle swelling is determined by the cross-linker concentration. Upon swelling with chloroform (100 vol %), the 1 wt % cross-linked microgel increases in diameter by 38% and the 5 wt % cross-linked microgel by 25%. Characterization of the CdSe Particles. The synthesized CdSe QD particles were characterized by TEM and UV-vis absorption spectroscopy. The results are shown in Figure 2. The size of the CdSe QDs, determined by TEM, is ∼5 nm, and the particles appear to be reasonably monodisperse (Figure 2a). The principal absorption maximum for the CdSe particles is at 520 nm (Figure 2b). Penetration of CdSe Nanoparticles into PS Microgel Particles. Confocal microscopy was used to investigate the distribution of CdSe nanoparticles that had been encapsulated within preformed PS microgel particles by the particle penetration technique, as a function of the extent of the microgel swelling. In Figure 3, a depth profile is shown for a (representative) 1 wt % cross-linked PS microgel particle, swollen with 5 vol %
2752
Langmuir, Vol. 21, No. 7, 2005
Bradley et al.
Figure 3. Confocal microscope images of sections through a 1 wt % cross-linked PS microgel containing CdSe QDs. The microgel particles were swollen in solvent mixtures of 5% chloroform (containing dispersed CdSe particles) and 95% propanol (vol/vol).
Figure 4. (a) Confocal microscope images of sections through a 1 wt % cross-linked PS microgel containing CdSe QDs. The microgel particles were swollen with a 60 vol % solution of chloroform, containing the dispersed CdSe particles. (b) Confocal microscope images of sections through a 5 wt % cross-linked PS microgel containing the CdSe particles. The microgel particles were swollen in solvent mixtures of 60% chloroform (containing the dispersed CdSe particles) and 40% propanol (vol/vol).
chloroform, in the presence of the QDs. A fluorescent “shell” around the edge of the spherical microgel particle is evident. Similar results were obtained for the 5 wt % crosslinked microgel particles. A fluorescence intensity profile across the midsection of the particle was used to calculate the infiltration depth of the QDs into the microgel particles, for both cross-linker concentrations. It was calculated that the penetration of the QDs was 39 and 24% of the radius for the 1 and 5 wt % cross-linked PS microgel particles, respectively. This correlates with the increased swelling of the 1 wt % cross-linked microgel particles, compared with the 5 wt % microgel particles. However, the fact that, in both cases, the QDs only penetrate into the peripheries of the microgel particles suggests that the effective “pore size” decreases toward the center of the microgel particles. It was anticipated that, by increasing the percentage of chloroform, hence increasing the microgel swelling, deeper penetration of the QDs into the microgel particles might be achieved. Figure 4 shows the results of swelling the 1 and 5 wt % cross-linked microgel particles with 60 vol % chloroform, in the presence of the CdSe particles. It is evident from these images that the QDs have, in this case, infiltrated further into the microgel particles. Focusing on Figure 4a (1 wt % cross-linked microgel particles), the QDs seem now to have penetrated the entire particle. Hence, at this extent of swelling, the pores in the center of the PS particle are now sufficiently large to accommodate the QDs, as well as those at the periphery. Figure 4a should be contrasted with Figure 4b, which shows a similar distribution of CdSe for the 5 wt % crosslinked particles. In the latter case, there appears again to be an impenetrable core region, indicated by the absence of fluorescence in this region. This may be due to a higher cross-linker density toward the center of the particle. It would seem that nanoparticle penetration experiments of this kind are very effective in revealing information concerning the internal structure and porosity of microgel particles dispersed in a solvent (mixture), under various conditions. Porosity information cannot be achieved for these types of systems using conventional methods, such as the Brunauer-Emmett-Teller (BET) method, as
these require a dry powder in which the microgel pores will be collapsed. Having knowledge of the internal structure of microgel particles is important for their applications, particularly in their use as reaction templates.10-12 Antonietti et al.10 have used polystyrene sulfonate microgel particles as nanoreactors for the reduction of a gold salt (Au1) to gold particles (Au0). Their work showed that one of the variables that affected the shape of the gold particles formed was the microgel particle cross-linker concentration. They found that the elastic, restraining forces imposed by the gel matrix restricted the growth of the gold particles. TEM images of the morphology of the gold particles formed gave an indication of the porosity and internal structure of their microgel particles. Their images resemble those shown here in Figure 4a, although their reaction conditions greatly influenced the morphology obtained and therefore are not a true representation of the internal structure of a microgel particle. Nie et al. have actively been investigating the generation of microbeads for optical coding.2,4 In their first report on this topic, they infiltrated PS microgel particles with quantum dots using the swelling technique, and in a solvent mixture containing 5 vol % chloroform. They estimated that the QDs were located in the outer 25% of the bead radius, and the results reported here are consistent with this. More recently, their focus has been on increasing the intensity of fluorescence from each bead by making the distribution of QD more uniform. To achieve this, they generated mesoporous PS beads. They found that mesoporous PS beads doped with QDs are more than 1000 times brighter and 5 times more uniform in fluorescence intensity than nonporous beads of the same size (and have QDs mainly located toward the outer radius of the particle). However, we show that it is possible to achieve the same distribution of QDs in PS microgel particles by simply tuning the extent of microgel swelling. (10) Antonietti, M.; Grohn, F.; Hartmann, J.; Bronstein, L. Angew. Chem., Int. Ed. Engl. 1997, 36, 2080. (11) Yao, H.; Takada, Y.; Kitamura, N. Langmuir 1998, 14, 595. (12) Spanka, C.; Clapham, B.; Janda, K. D. J. Org. Chem 2002, 67, 3045.
Distribution of CdSe QDs within PS Microgel Particles
Figure 5. (a) TEM image of a cross section (50 nm slice) of a 1 wt % cross-linked PS microgel containing the CdSe particles. (b) Higher-magnification TEM image focusing on an area containing CdSe particles within the cross section.
Distribution of CdSe Nanoparticles in Suspension-Polymerized PS Particles. Direct suspension polymerization of styrene monomer droplets, containing the dispersed CdSe QDs, was also carried out. Although the QDs dispersed well in the monomer and PS particles containing the QDs were successfully obtained, when these dispersions were subjected to UV light, they did not fluoresce. Hence, in this case, ultramicrotomy was used to probe the interior of the microgel particles. Figure 5a shows the incorporation of the QDs within the PS particles. It appears as if the QDs are aggregated. The higherresolution image (Figure 5b) confirms this picture. It seems likely that this aggregation occurs during the polymerization process; it could be caused by a depletion aggregation mechanism as the polystyrene MW and concentration build up during polymerization. To overcome the fluorescence quenching problem, particles of CdSe capped with ZnS were used and the suspension polymerization process was repeated. Figure 6a shows the confocal microscopy results for the direct polymerization of the CdSe/ZnS QDs into a 1 wt % crosslinked PS microgel particle. It is clear that phase separation of the QDs toward the polymer/water interface has occurred during the polymerization process, with most of the QDs located toward the exterior of the particle. Figure 6b shows the results of swelling this sample with 60 vol % chloroform. As expected, given the results of the swelling
Langmuir, Vol. 21, No. 7, 2005 2753
Figure 6. (a) Confocal microscope image of sections through a 1 wt % cross-linked PS microgel containing CdSe QDs. The QDs were dispersed in the monomer before polymerization. (b) Confocal microscope image after swelling in a solvent mixture of 60% chloroform and 40% propanol (vol/vol).
experiments, it is possible to redistribute the QDs throughout the entire microgel particle. Conclusions The confocal microscopy results have shown that nanoparticle penetration experiments may be used to explore the internal “pore” structure of large microgel particles, as a function of their cross-linker density and degree of swelling. It would seem that, at a given solvent composition, the pore size decreases toward the center of the microgel particles. It would also seem that the nanoparticle “penetration” method into (sufficiently swollen) polymer microgel particles is a much better method for producing polymer beads containing (stable) QDs, distributed over the whole volume space of the particles, than in situ suspension polymerization of monomer droplets, containing dispersed QDs, since aggregation of the latter is difficult to avoid during the polymerization process. Acknowledgment. We wish to thank Rachel Doherty (Bristol University) for help with the synthesis of the QDs, Dr Paul Mulvaney (Melbourne University) for providing the ZnS-capped QDs, and Dr Paul Bartlett (Bristol University) for the help using the confocal microscope. We gratefully acknowledge the support of the EPSRC (through the IMPACT Faraday Partnership) (Grant No. GR/R90086/01). LA047322R
