J. Phys. Chem. C 2007, 111, 17473-17477
17473
ARTICLES Synthesis of Porous Magnetic Hollow Silica Nanospheres for Nanomedicine Application J. Zhou,† W. Wu,†,‡ D. Caruntu,† M. H. Yu,† A. Martin,† J. F. Chen,*,‡ C. J. O’Connor,† and W. L. Zhou*,† AdVanced Materials Research Institute, UniVersity of New Orleans, New Orleans, Louisiana 70148, and Key Laboratory for Nanomaterials, Ministry of Education, Beijing UniVersity of Chemical Technology, Beijing 100029, China ReceiVed: May 28, 2007; In Final Form: June 26, 2007
A porous magnetic hollow silica nanosphere (MHSN) is a new nanostructured drug carrier for increasing drug loading capability. Keeping the magnetic nanoparticles in the hollow core will limit the toxicity and degradation in a biosystem. In this paper, we report a synthesis of porous MHSNs by sol-gel method. CaCO3/ Fe3O4 composite particles were first fabricated by embedding Fe3O4 nanoparticles into CaCO3 using the rotating packed bed (RPB) method. Tetraethoxysilane (TEOS) was then added as precursor to form a silica (SiO2) layer on the surface of CaCO3/Fe3O4 composite particles. Hexadecyltrimethylammonium bromide (CTAB) and octane act as second templates for the formation of porous silica shells. After removing the surfactants by calcination and etching away the CaCO3 particles, porous MHSNs with magnetite (Fe3O4) nanoparticles inside the cores were formed. The pore size can be tuned by adjusting the amount of the cationic surfactant absorbed on the surface of the composite particles to form self-assembled nanochannels. Ibuprofen was loaded on or into the porous MHSNs, and the drug encapsulation and release were investigated. A slow release was observed for the porous MHSNs, which demonstrated MHSNs are potential carriers for controlled releasing in nanomedicine application.
1. Introduction Magnetic nanoparticles with properties of superparamagnetism have been used in magnetic bioseparation, drug delivery, magnetic resonance imaging contrast enhancement, and hyperthermia treatment of cancer.1-4 Pure magnetic nanoparticles are not very useful in practical applications since they are prone to aggregation and rapid biodegradation when they are directly exposed to a biological system. To overcome these limitations, the magnetic nanoparticles are usually used in the form of coreshell structures or composite nanoparticles.5-7 Most of the coreshell magnetic carriers are dense and consist of cores of superparamagnetic nanoparticles and functionalized shells (silica,8 oleic acid,9 and polystyrene,10 etc.), while composite particles are composed of superparamagnetic nanoparticles and a dispersing matrix (mesoporous silica,11-12 starch, and methoxypoly(ethylene glycol),9 etc.). These shells or matrices usually possess biocompatibility and high surface reactivity and can be conjugated to or trap biomolecules and drugs. Among all of the abovementioned shells and matrices, mesoporous silica is a promising structured material. Its high and well-documented biocompatibility is good for practical applications of magnetic nanoparticles in magnetically guided drug delivery and tumor targeting. * To whom correspondence should be addressed. Telephone: (504) 280-1068 (W.L.Z.); 86-10-64446466 (J.F.C.). Fax: (504) 280-3185 (W.L.Z.); 86-10-64446466 (J.F.C.). E-mail:
[email protected] (W.L.Z.);
[email protected] (J.F.C.). † University of New Orleans. ‡ Beijing University of Chemical Technology.
Hollow structures have attracted much attention due to their various applications, ranging from fillers for low dielectric constant materials and light-weight composites to microencapsulation for controlled release and drug delivery.13-15 Many templates are used to prepare hollow nanostructures such as nanoparticles,16 emulsion droplets,17 lyotropic phases with a multilamellar vesicular structure,18 vesicular solution,19 organic polymeric sphere,20 and inorganic-organic double templates,21 etc. Using nanoparticle templates16,21,22 is an efficient way to achieve hollow nanostructures. In the past, we have synthesized magnetic hollow silica nanospheres using CaCO3 nanospheres and polystyrene beads as templates via sol-gel method.23-25 The magnetic particles are generally embedded in the shells or coated on the surface. Unlike our former methods, Fe3O4/CaCO3 composite nanoparticles were prepared by using a rotating packed bed (RPB) reactor first. The composite nanoparticle was then used as a template and tetraethoxysilane (TEOS) as a precursor to fabricate magnetic hollow silica nanospheres (MHSNs). Since the mesoporous silica has been proven to be a very promising nanostructured material for trapping biomolecules and drugs, therefore, pores were also introduced in the silica shells of MHSNs using hexadecyltrimethylammonium bromide (CTAB) and octane as templates. To investigate their nanomedicine applications, ibuprofen was loaded into the porous MHSNs to study their drug loading and releasing capabilities. 2. Experimental Section 2.1. Synthesis of Fe3O4/CaCO3 Composite Nanoparticles. In a typical experiment, 3500 mL of 5.4 wt % Ca(OH)2
10.1021/jp074123i CCC: $37.00 © 2007 American Chemical Society Published on Web 08/11/2007
17474 J. Phys. Chem. C, Vol. 111, No. 47, 2007
Zhou et al.
Figure 3. TEM bright-field (a) and dark-field (b) images, showing Fe3O4 nanoparticles embedded in the CaCO3 particles.
Figure 1. Schematics of rotating packed bed (RPB) reactor: (1) CO2 gas inlet; (2) rotator; (3) packing; (4) distributor; (5) casing and routeway of circulating water; (6) outlet of suspension of Ca(OH)2 and Fe3O4; (7) CO2 gas outlet; (8) inlet of suspension of Ca(OH)2 and Fe3O4.
Figure 2. FESEM (a) and TEM (b) images of CaCO3 composite particles with Fe3O4 nanoparticles embedded inside.
suspension was added to the RPB reactor (Figure 1) with a rotating speed of 750 rpm and circulated at 400 L/h. After 10 min, 75 mL of a methanolic dispersion of Fe3O4 nanoparticles (15 mg/mL) was added to the system, and then CO2 gas was introduced at a rate of 100 L/h. Superparamagnetic Fe3O4 nanoparticles were prepared by a method based on the hydrolysis of chelate metal alkoxide complexes at elevated temperature in solutions of diethylene glycol (DEG), which was published elsewhere.26 CO2 and Ca(OH)2 contacted at a converse direction and reacted in the reactor. The suspension was circulated continuously during the process. When the pH value of the suspension reached 7.4, CO2 gas was stopped and the reaction was ended. The temperature of the suspension was kept below 25 °C by circulating water during the entire reaction (about 70 min), which was monitored by a thermometer. The obtained suspension was filtrated and dried at 70 °C in vacuum for 6 h to achieve Fe3O4/CaCO3 composite nanoparticles. Then a solgel method was used to prepare magnetic hollow silica nanospheres. 2.2. Synthesis of Porous Magnetic Hollow Silica Nanospheres. A 3 g amount of as-prepared Fe3O4/CaCO3 composite nanoparticles was dispersed ultrasonically into a mixture of 60 mL of ethanol (A APer Acohol and Chemical Co.) and 40 mL of distilled water for 30 min in a beaker and then mechanically stirred for 30 min in a three-neck flask. Then 0.74 g of CTAB (99+%; Johnson Matthew Co.) and 3.30 mL of octane (98%; Aldrich Chemical Co., Inc.) were added to the above suspension, and the obtained mixture was dispersed ultrasonically for another 20 min.27-29 After adding 34 mL of ammonia (30%; J.T. Baker), 3.7 mL of TEOS (99.9%; Johnson Matthew) was dropped into the reaction mixture within 10 min to reach the pH value of 11. The system was stirred at 500 rpm for 2 h at room
Figure 4. TEM image (a) and EDS spectrum (b) of porous MHSNs. The inset is a HREM image, showing pores on the hollow silica shell.
temperature. The obtained suspension was aged for 6 h at room temperature and then filtrated. The cake was dried at 50 °C for 6 h and calcined at 550 °C for 5 h to remove the surfactant template. After calcination, the product was immersed in diluted acetic acid (J. T. Baker) solution (HAc:H2O ) 1:15) for 5 h to remove the CaCO3 cores. After three washes, two times with distilled water and one time with alcohol, and drying at 75 °C for 18 h, the MHSNs with different pore sizes were obtainedand were further used for drug loading and releasing experiments. 2.3. Drug Loading. A 75 mg amount of porous MHSNs was dispersed into 25 mL of ibuprofen (MP Biomedical, USP grade) hexane solution (EM Sciences, GR) with a concentration of 1.8 mg/mL, and the obtained mixture was stirred for 48 h. For the drug release experiments, phosphate buffer solution (pH ) 7.4) was prepared by diluting 79 mL of 0.1 M KOH containing 1.36 g of KH2PO4 to 200 mL. Then the ibuprofen phosphate buffer solutions with different concentrations were prepared and measured by UV spectroscopy. The standard curve of ibuprofen
Hollow Silica Nanospheres for Applied Nanomedicine
J. Phys. Chem. C, Vol. 111, No. 47, 2007 17475
Figure 5. Magnetic properties of the MHSNs: (a) temperature dependence of magnetization at ZFC and FC conditions; (b) hysteresis loops of the MHSNs at 5 and 300 K.
in phosphate buffer solution was drawn, and the linear equation is
Y ) 0.246 + 0.04627X
(1)
For drug release, 70 mg of porous MHSNs containing ibuprofen was dispersed into 50 mL of phosphate buffer solution. The mixture was stirred, and 1 mL samples were extracted at different time intervals. These sample solutions were diluted to 5 mL with pure phosphate buffer solution and centrifuged at 3500 rpm for 15 min. The upper clear solutions were analyzed by UV spectroscopy, and their concentrations can be calculated by the linear equation. Therefore, the drug release curve can be obtained. 2.4. Characterization. A JEOL 2010 transmission electron microscope (TEM) and Carl Zeiss 1530 VP field emission scanning electron microscope (FESEM) were used to examine the size and the morphology of the Fe3O4/CaCO3 composite nanoparticles and the porous MHSNs. EDAX energy dispersive spectroscopy (EDS) was applied to determine the composition of the nanoparticles. A Philips X’pert-MPD diffractometer (Xray diffraction (XRD)) equipped with a graphite monochromator and Cu KR (1.5418 Å) radiation was employed to analyze the crystal structure. A Quantum Design MPMS-5S superconducting quantum interference device (SQUID) magnetometer was used to measure the magnetic properties. N2 adsorption-desorption isotherms were measured at 77 K on a micromeritics ASAP 2010 analyzer. An UV-vis-near-IR spectrophotometer was used to determine absorbency of the ibuprofen solution. 3. Results and Discussion To obtain a better coating of one substance on another, a lower reactant concentration was used aiming at decreasing reaction rate. The preparation of the Fe3O4/CaCO3 composite nanoparticles is a three-phase reaction (gas-liquid-solid) in
Figure 6. Hysteresis loops of the MHSNs at 5 and 300 K before (a) and after (b) calcinations.
Figure 7. Nitrogen sorption isotherm for the MHSNs.
which the transfer of CO2 is a key factor for controlling the reaction rate. In fact, the flow rate of CO2 was only one-sixth of the normal speed of preparing CaCO3 nanoparticles with the RPB method.21,23,24 When RPB is taken as a reactor for preparing spherical CaCO3 nanoparticles, the micromixing strength of the reactants can be enhanced. Accordingly, the size, size distribution, and reaction rate of the CaCO3 nanoparticles are improved. Parts a and b of Figure 2 are FESEM and TEM images of Fe3O4/CaCO3 composite nanoparticles, respectively.
17476 J. Phys. Chem. C, Vol. 111, No. 47, 2007
Zhou et al.
Figure 8. Pore size distribution for the MHSNs sample calculated using BJH method.
The composite nanoparticles are of spherical shape with size ranging from 60 to 90 nm and tend to stick together because of the nature of nanosized particles. Parts a and b of Figure 3 are bright- and dark-field images of the as-synthesized composite nanoparticles. By selecting electron spots from the Fe3O4 diffraction rings for imaging, the Fe3O4 nanoparticles are clearly seen in the dark-field image of Figure 3b. The composite nanoparticles are mostly composed of CaCO3 with Fe3O4 nanoparticles embedded. Few Fe3O4 nanoparticles were observed on the surface of CaCO3 particles. A longer time (2 h) was needed to form homogeneous silica shells on Fe3O4/CaCO3 composite nanoparticles. In fact, it includes two processes: hydrolysis and polymerization. The cationic surfactant CTAB and octane were added as templates during the formation of porous MHSNs in order to tune the pore size.27-29 The MHSNs with different pore sizes in their shells can be achieved after the calcination at high temperatures to remove the organic templates. After etching away CaCO3 by using diluted acetic acid,23,24 hollow silica nanospheres containing magnetic nanoparticles were obtained. Figure 4a shows the porous MHSNs with most of the Fe3O4 nanoparticles in the core. It is apparent from the TEM micrograph that, during the reaction, Fe3O4/CaCO3 composite nanoparticles aggregated and formed nanoparticle clusters, resulting in the formation of bigger silica particles. The EDS analysis shows that, except C and Cu peaks from the TEM grid, only Fe, O, and Si were observed, as shown in Figure 4b. No Ca peak was found, implying that CaCO3 was completely etched away after the reaction. The inset in Figure 4a is a HREM image of CTABinduced porous MHSNs, showing porous structure in the shell. Some ordered pores, which are the form of the higher density porous structure, were also observed. The ordered pores could increase the sites for drug loading later on. The magnetic properties of the MHSNs were measured by SQUID. Zero-field-cooled (ZFC) and field-cooled (FC) magnetization data measured in the temperature range of 5-300 K are shown in Figure 5a. In the ZFC measurement, the initial field was set to zero when cooling the sample from 300 to 5 K. A field of 100 Oe was applied, and the magnetization was measured as the sample was heated from 5 to 300 K. In FC
Figure 9. Drug release of ibuprofen from the porous MHSNs showing slow release.
measurement, a field of 100 Oe was applied as the sample was cooled from 300 to 5 K and the magnetization was measured as the sample was heated from 5 to 300 K in the field of 100 Oe. The ZFC curve shows a maximum at 98.5 K, which is the blocking temperature (TB) of the magnetic nanoparticles in the MHSNs. Such behavior is characteristic of superparamagnetism and is due to the progressive blocking of the magnetic moment of nanoparticles when decreasing the temperature. The MHSNs exhibit superparamagnetism and ferromagnetism above and below the blocking temperature, respectively. To ensure these features, the field-dependent hysteresis loops of the MHSNs were measured at temperatures both below and above the blocking temperature, as shown in Figure 5b. The hysteresis loop at 5 K shows a saturation magnetization of 1.4 emu/g (at the field of 10 000 Oe) and a coercivity of 173 Oe, which confirms that the MHSNs are ferromagnetic below the blocking temperature. The absence of coercivity in the hysteresis loop at 300 K indicates a superparamagnetic behavior. The porous MHSNs have superparamagnetic properties, which means that they are attracted by a magnetic field but retain no residual magnetism when the magnetic field is removed at room temperature. The SQUID measurement was also used to investigate the effect of calcination on magnetic properties. The results show that there is no obvious change in the magnetic properties in this system. TB decreased from 106 to 86 K due to the increase of the concentration of the magnetic nanoparticles in the MHSNs; however, they still exhibit the superparamagnetism above the blocking temperature and ferromagnetism below the blocking temperature which can be further proven by the hysteresis loops, as seen in Figure 6. The nitrogen adsorption-desorption isotherms and BarrettJoyner-Halenda (BJH) pore size distributions (PSD) of the MHSNs are shown in Figure 7. The isotherms of the sample with CTAB and octane (sample 2) have sharper inflections than that without adding a surfactant (sample 1), indicating uniform pore size formation during capillary condensation as shown in Figure 7. Besides, the sharpest peak, seen in Figure 8, indicates a quite uniform pore size distribution for sample 2 which means that surfactant-induced templates were formed during the
TABLE 1: BET Surface Area and Pore Texture Parameters of the Samples sample 1 2
surfactant
pore size diam (nm)
BET surface area (m2/g)
drug-load amt (%)
CTAB and octane
1.5 3.7
91.9995 242.8824
8.7 14.21
Hollow Silica Nanospheres for Applied Nanomedicine synthesis and were also successfully removed during the calcination and rinsing process. To explore the drug carrying capacity of the porous MHSNs, ibuprofen, a typical anti-inflammatory drug, was introduced into the pores of the MHSNs. The uptake amount of ibuprofen, assessed by UV analysis, is ca. 8.7% for the MHSNs with a pore size of 1.5 nm and 14.21% for those with a pore size of 3.7 nm (Table 1); these results indicate that the loaded amount of ibuprofen varies with the change of the pore size and surface area. The release test was performed in 50 mL of pH 7.4 phosphate buffer solution, as seen in Figure 9. It was observed that about 15% of the ibuprofen was released from the MHSNs with a pore size of 3.7 nm in the first half-hour, and 42% was released after 24 h. In the case of MHSNs with pore size of 1.5 nm (without adding any surfactant), however, more than 80% of the ibuprofen was released in the first 0.5 h, which means that most of ibuprofen was loaded on the surface of the silica hollow spheres since the drug adsorbed on the surface of the MHSNs released faster. In addition, the porous MHSNs with pore size of 3.7 nm demonstrate slower release rates than those with a pore size of 1.5 nm, since a certain amount of ibuprofen entered into the pores of silica hollow spheres through the nanochannels. 4. Conclusions Porous MHSNs have been synthesized using CaCO3/Fe3O4 composite nanoparticles prepared by RPB method and a cationic surfactant as template in the sol-gel system. The pore size of the MHSNs can be tuned by adjusting the amount of the surfactant in this system. No CaCO3 and surfactant were left after weak acid etching and calcination. The fabrication of pores on the hollow shells increased drug loading capability and built nanochannels in the silica shells to link the hollow core and the outside for controlled release. SQUID results show that the porous MHSNs still maintain the superparamagnetic behavior, and the calcination at 550 °C did not result in any change of the magnetic properties. Keeping magnetite nanoparticles in the core will prevent the interaction between nanoparticles and the biosystem and limit the toxicity and degradation of the magnetic nanoparticles. In addition, the drug storage and release properties were also investigated. The ibuprofen can be loaded into the MHSNs through the pores, and the amount of the ibuprofen loaded and the rate of the drug release were improved by introducing pores in the MHSNs. A slow release of ibuprofen from the porous MHSNs was observed, which indicates that the porous MHSNs containing magnetic particles in their cores have high potential applications in controlled and target releasing. Acknowledgment. We gratefully acknowledge the support of this work by the AMRI through the DARPA grant no.
J. Phys. Chem. C, Vol. 111, No. 47, 2007 17477 HR0011-07-1-0032. We also thank Dr. G. Caruntu for helping collect the XRD data. A.M. especially is grateful for the support of the National Science Foundation (NSF grant no. DMR0243977). W.L.Z. and J.-F.C. acknowledge partial support from China NSF grant no. 20325621. References and Notes (1) Yang, H. H.; Zhang, S. Q.; Chen, X. L.; Zhuang, Z. X.; Xu, J. G.; Wang, X. R. Anal. Chem. 2004, 76, 1316. (2) Haukanes, B. I.; Kvam C. BioTechniques 1993, 11, 60. (3) Tanaka, T.; Matsunaga, T. Anal. Chem. 2000, 72, 3518. (4) Levy, L.; Sahoo, Y.; Kim, K. S.; Bergey, E. J.; Prasad, P. N. Chem. Mater. 2002, 14, 3715. (5) Liu, X.; Ma, Z.; Xing, J.; Liu, H. J. Magn. Magn. Mater. 2004, 270, 1. (6) Santra, S.; Tapec, R.; Theodoropoulou, N.; Dobson, J.; Hebard, A.; Tan, W. H. Langmuir. 2001, 17, 2900. (7) Bruce, I. J.; Taylor, J.; Todd, M.; Davies, M. J.; Borioni, E.; Sangregorio, C.; Sen., T. J. Magn. Magn. Mater. 2004, 284, 145. (8) Vestal, C. R.; Zhang, Z. J. Nano Lett. 2003, 3, 1739. (9) Kim, D. K.; Mikhaylova, M.; Zhang, Y.; Muhammed, M. Chem. Mater. 2003, 15, 1617. (10) Sieben, S.; Bergemann, C.; Lubbe, A.; Brockmann, B.; Rescheleit, D. J. Magn. Magn. Mater. 2001, 225, 175. (11) Tartaj, P.; Gonzalez-Carreno, T.; Serna, C. J. Langmuir 2002, 18, 4556. (12) Bourlinos, A. B.; Simpoulos, A.; Boukos, N.; Petridis, D. J. Phys. Chem. B 2002, 105, 7432. (13) Avelin, J.; Sihvola, A. J. Elec. 2002, 56, 19. (14) Li, Z. Z.; Wen, L. X.; Shao. L.; Chen, J. F. J. Controlled Release 2004, 98, 245. (15) Sato, Y.; Kawashima, Y.; Takeuchi, H.; Yamamoto, H. Eur. J. Pharm. Biopharm. 2004, 57, 235. (16) Dejugnet, C.; Sukhorukov, G. B. Langmuir 2004, 20, 7265. (17) Schacht, S.; Huo, Q.; Voigt-Martin, I. G.; Stucky, G. D.; Schuth, F. Science 1996, 273, 768. (18) Kramer, E.; Forster, S.; Goltner, C. G.; Antonetti, M. Langmuir 1998, 14, 2027. (19) Hentze, H. P.; Raghavan, S. R.; Mckelvey, E. W.; Kaler, E. W. Langmuir 2003, 19, 1069. (20) Lu, Y.; Mclellan, J.; Xia. Y. Langmuir 2004, 20, 3464. (21) Chen, J.-F.; Ding, H.-M.; Wang, J.-X.; Shao. L. Biomaterials 2004, 25. 723. (22) Lai, C.-Y.; Trewyn, B. G.; Jeftinija, D. M.; Lin, V. S.-Y. J. Am. Chem. Soc. 2003, 125, 4451. (23) Shao, L.; Caruntu, D.; Chen, J. F.; O’Connor, C. J.; Zhou, W.L. J. Appl. Phys. 2005, 97, 10Q908-1. (24) Wu, W.; DeCoster, M. A.; Daniel, B. M.; Chen, J. F.; Yu, M. H.; Caruntu, D.; O’Connor, C. J.; Zhou, W. L. J. Appl. Phys. 2006, 9, 08H1041. (25) Wu, W.; Yu, M. H.; Caruntu, D.; O’Connor, C. J.; Chen, J. F.; Zhou, W. L. J. Magn. Magn. Mater. 2007, 311, 578. (26) Caruntu, D.; Caruntu, G.; Chen, Y. C.; O’Connor, C. J.; Goloverda, G.; Kolesnichenko, V. L. Chem. Mater. 2004, 16, 5527. (27) Lu, Y.; Yang, Y.; Sellinger, A.; Lu, M.; Huang, J.; Fan, H.; Haddad, R.; Lopez, G.; Burns, A. R.; Sasaki, D. Y.; Shelnutt, J.; Brinker, C. J. Nature 2001, 410, 913. (28) Wang, D.; Zhou, W. L.; McCaughy, B. F.; Hampsey, J. E.; Ji, X.; Jiang, Y. B.; Xu, H.; Tang, J.; Schmehl, R. H.; O’Connor, C.; Brinker, C. J.; Lu, Y, AdV. Mater. 2003, 15 (2), 130. (29) Lu, Y. Angew. Chem., Int. Ed. 2006, 45, 7664.
