Self-Assembly of Ultrabright Fluorescent Silica ParticlesThis work was supported by the US Army Research Office (grant W911NF-05-1-0339)

Fluorescent particles have broad applications in tagging, tracing, and labeling. Fluorescence is typically generated through incorporation of either inorganic or organic fluorescent dyes into the particle s material. While inorganic dyes are typically more stable, their number and compatibility is rather restricted. The large variety of organic dyes makes them attractive for creating fluorescent particles. However, the problems are in the stability and typically high toxicity of these dyes. Incorporation of dyes into a silica matrix seems to be one of most promising approaches because of the excellent sealing ability of silica and wide compatibility of silica with other materials. Numerous attempts to embed organic dyes into silica xerogels and zeolites have been made for a long time. To prevent leakage of the dyes out of the porous matrix, the dyes were covalently bound to the silica matrix. While the photostability of such materials was higher than the stability of pure dyes, it did not prevent bleaching substances, including oxygen, from penetrating inside such composite materials. In the case of xerogels, it is rather hard to use them for labeling. Recently there have been several reports about the incorporation of fluorescent lasing dyes into mesoporous patterned silica films and silica roads. Here we describe a novel synthesis of mesoporous silica particles with encapsulated organic dyes, in which the dyes are physically entrapped inside a silica matrix, through a one-step self-assembly process. We also demonstrate the encapsulation of a mixture of different dyes within single silica particles. The dyes are physically encapsulated inside nanosize channels/tubes in micrometersize particles. Particles of such size can be of particular use in a number of applications, including flow cytometry and the labeling of skincare products. Encapsulation is manifested through virtually no leakage of the dyes out of the particles. High concentrations of the fluorescent dyes can be reached inside the pores without dimerization of the fluorescent molecules. This results in fluorescence that is up to 500 times brighter than the maximum obtained from the same dye dissolved in aqueous solution at its maximum concentration before self-quenching (due to dimerization). By comparing this fluorescence with that observed for the brightest micrometer-size particles assembled from aqueouscompatible quantum dots encapsulated in polymeric particles of similar size ( 1.2 mm) reported recently, one finds that the particles presented here (scaled to the same size) are about 170 times brighter. The fluorescence of the assembled particles is very stable. To synthesize the mesoporous silica particles, the process of self-organization of mesoporous silica through acidic cationic surfactant templating and condensation of a silica precursor was used. The dyes are added at the beginning of the synthesis at concentrations (1–20810 m) that are substantially higher than the dimerization concentration of the fluorescent molecules ( 0.5810 m). It is interesting to note that the dyes are still dimerized in the synthesis solution, while we do not see any dimerization in the final product, the silica particles with encapsulated dyes (dimerization of the dyes can easily be detected by broadening of the fluorescence and, especially, absorbance peaks). It is known that the type of the synthesis used here results in a hexagonal array of pores/channels, which is typical for MCM-41 (or SBA-3 in another classification) material. Small-angle X-ray scattering analysis (SAXS; not shown) shows that we are dealing with a mesoporous material with an interchannel distance of about 4 nm, which is the same as reported previously. 25] The most important feature of our synthesis is the morphology of the assembled material. This morphology consists of particles with specific shapes, mostly “discoids” and “gyroids” with a size on the order of a few micrometers (Figure 1a). Typical dimensions of the discoids/gyroids are 5–10 mm. One can also see some fiber shapes. The fibers, typically, are noticeably larger than the discoids/gyroids. Therefore, it is not difficult to separate them if necessary by either filtering or fluidic separation (Figure 1b, c). The inset in Figure 1a shows a zoomed image of a couple of particles that have a specific “ray” type of defect on their surfaces. Such defects have already been reported and are the manifestation of closed channels/ pores running along the circumference around the axis of symmetry of the particles. Certainly, this is indirect evidence of having closed pores. More direct proof can be obtained by analysis of the diffusion of the dyes out of the particles. It makes sense to compare diffusion out of our particles with diffusion out of particles that are known to have open channels/pores. The synthesis of well-characterized openchannel fibers with the same type of symmetry of channels and about the same microsize (285 mm) as our particles has been reported recently. We used that synthesis but added the same amount of dye (rhodamine 6G, R6G) as was used for the synthesis reported here. As a result, we collected open-channel fibers with the dye inside the channels. The fibers were collected with a cone filter (0.5-mm open pore) and continuously washed for two minutes with distilled (DI) water. Our “closed-channel” particles, the discoids/gyroids, were collected and washed in exactly the same way. After that, the open-channel fibers were put in DI water in a cuv[*] Prof. I. Sokolov Department of Physics, Department of Chemistry, and Center for Advanced Material Processing (CAMP) Clarkson University, Potsdam, NY 13699-5820 (USA) Fax: (+1)315-268-6610 E-mail: [email protected]