7492
Langmuir 2008, 24, 7492-7499
Engineering of SiO2-Au-SiO2 Sandwich Nanoaggregates Using a Building Block: Single, Double, and Triple Cores for Enhancement of Near Infrared Fluorescence Shuping Xu, Shay Hartvickson, and Julia Xiaojun Zhao* Department of Chemistry, UniVersity of North Dakota, Grand Forks, North Dakota 58202 ReceiVed February 13, 2008. ReVised Manuscript ReceiVed April 6, 2008 We have developed a simple and flexible chemical method to synthesize orderly metallic nanoaggregates using a designed SiO2-Au core-shell building block. The number of the building blocks in a nanoaggregate is tunable from one to three. These metal nanostructures can generate an enlarged localized electromagnetic field through surface plasmon resonance and enhance the optical signals of the photoactive molecules within this electromagnetic field. Aggregates of metallic nanoparticles provide a higher signal enhancement than well-dispersed nanoparticles combined. The level of signal enhancement is determined by the number of building blocks in a nanoaggregate. The signal enhancement of the nanoaggregates has been verified with a near-infrared (NIR) dye. In the NIR region, biological samples have low background signals and deeper penetration of radiation. The application of these NIR enhanced metal nanostructures will open a significant approach for sensitive detection of biological samples.
1. Introduction The need for sensitive determinations of trace amounts of analytes has driven the rapid development of various novel nanomaterials. Photoactive nanomaterials, such as quantum dots (QDs),1,2 dye-doped nanoparticles,3 and gold or silver nanoparticles,4–7 are some of the most promising signaling reagents for achieving high detection sensitivity. These nanomaterials provide direct signals for the determination of trace analytes. However, the signal intensity of these nanomaterials is intrinsic and limited by their maximum value. To raise the limit of their intrinsic intensities, some form of signal amplification is needed. One alternative is photonic resonance enhancement. Noble metal nanostructures can generate an enlarged localized electromagnetic field through surface plasmon resonance and enhance the optical signals of the photoactive molecules within this electromagnetic field.8–10 The principle of the localized surface plasmon resonance (LSPR) of metallic nanostructures has been well studied in the past several years.8–10 At the nanoscale, collective oscillations of metallic free electrons are limited by the nanostructure boundaries and thus form surface plasmon waves along the interface. When the nanomaterial interface is irradiated by an * Corresponding author. Tel: 701-777-3610. E-mail:
[email protected]. (1) Chan, W. C. W.; Nie, S. M. Science 1998, 281, 2016–2018. (2) Bruchez, M.; Moronne, M.; Gin, P.; Weiss, S.; Alivisatos, A. P. Science 1998, 281, 2013–2016. (3) (a) Bagwe, R. P.; Yang, C.; Hilliard, L. R.; Tan, W. Langmuir 2004, 20(19), 8336–8342. (b) Tan, W.; Wang, K.; He, X.; Zhao, J. X.; Drake, T.; Wang, L.; Bagwe, R. P. Med. Res. ReV. 2004, 24(5), 621–638. (c) Wang, L.; Yang, C.; Tan, W. Nano Lett. 2005, 5, 37–43. (d) Zhao, X.; Tapec-Dytioco, R.; Tan, W. J. Am. Chem. Soc. 2003, 125(38), 11474–11475. (4) Daniel, M.-C.; Astruc, D. Chem. ReV. 2004, 104(1), 293–346. (5) Ghosh, S. K.; Pal, T. Chem. ReV. 2007, 107(11), 4797–4862. (6) Jain, P. K.; Huang, X.; El-Sayed, I. H.; El-Sayed, M. A. Plasmonics 2007, 2(3), 107–118. (7) Kazuo, W.; Dietrich, M.; Niklas, N.; Hans-Joachim, F. Chem. ReV. 2006, 106(10), 4301–4320. (8) (a) Stuart, D. A.; Haes, A. J.; Yonzon, C. R.; Hicks, E. M.; Van Duyne, R. P. IEEE Proc. Nanobiotech. 2005, 152(1), 13–32. (b) Haes, A. J.; Haynes, C. L.; McFarland, A. D.; Schatz, G. C.; Van Duyne, R. P.; Zou, S. MRS Bull 2005, 30(5), 368–375. (c) Willets, K. A.; Van Duyne, R. P. Annu. ReV. Phys. Chem. 2007, 58, 267–297. (9) Xia, Y.; Halas, N. J. MRS Bull. 2005, 30(5), 338–348. (10) Lakowicz, J. R. Plasmonics 2006, 1(1), 5–33.
incident light beam, the surface plasmon wave resonates with the optical wave at an optimized condition, resulting in the greatest absorption of the incident light. As a result, an enlarged localized electromagnetic field is manifested around the nanostructures, providing extra energy for signaling reagents present within this electromagnetic field. One remarkable example of this effect is surface enhanced Raman scattering (SERS). Using Au or Ag nanomaterials, SERS can enhance scattering signals by up to 1010-fold.11–13 The energy level of the electromagnetic field strongly depends on the metallic plasmon property of the nanomaterials.14 This property is determined by several factors, including characteristics of the metal noumenon (size, shape, structure and dielectric constant), the surrounding medium (dielectric constant), the incident light (direction and wavelength), and so forth.8–10 Among these, the shape, size, and structure of the metal nanomaterials are critical to achieve controllable plasmonic materials and powerful surface-enhanced matrices.15,16 Aggregates of metallic nanoparticles generate higher signal enhancement than individual nanoparticles combined. Theoretical stimulations have demonstrated that the edge of nanostructures in general and the junction area between two nanoparticles in an aggregate exhibit a stronger localized electromagnetic field than other areas.17–19 Due to such an effect, research efforts have been focused on the development of various metallic nanoag(11) Mulvaney, S. P.; Keating, C. D. Anal. Chem. 2000, 72, 145R–157R (12) Campion, A.; Kambhampati, P. Chem. Soc. ReV. 1998, 27, 241–250. (13) (a) Krug, J. T.; Wang, G. D.; Emory, S. R.; Nie, S. J. Am. Chem. Soc. 1999, 121, 9208–9214. (b) Emory, S. R.; Nie, S. J. Phys. Chem. B 1998, 102, 493–497. (c) Emory, S. R.; Haskins, W. E.; Nie, S. J. Am. Chem. Soc. 1998, 120, 8009–8010. (14) (a) Itoh, T.; Hashimoto, K.; Ikehata, A.; Ozaki, Y. Appl. Phys. Lett. 2003, 83, 5557. (b) Itoh, T.; Hashimoto, K.; Ikehata, A.; Ozaki, Y. Chem. Phys. Lett. 2004, 389, 225–229. (15) Wiley, B.; Sun, Y.; Chen, J.; Cang, H.; Li, Z.-Y.; Li, X.; Xia, Y. MRS Bull. 2005, 30(5), 356–361. (16) Jana, N. R.; Pal, T. AdV. Mater. 2007, 19(13), 1761–1765. (17) Stockmann, M.; Shalev, V. M.; Moskovits, M.; Botet, R.; George, T. F. Phys. ReV. B 1992, 46, 2821–2830. (18) Lal, S.; Taylor, R. N.; Jackson, J. B.; Westcott, S. L.; Nordlander, P.; Halas, N. J. J. Phys. Chem. B 2002, 106(22), 5609–5612. (19) Quidant, R.; Zelenina, A. S.; Nieto-Vesperinas, M. Appl. Phys. A-Mater. 2007, 89(2), 233–239.
10.1021/la8004757 CCC: $40.75 2008 American Chemical Society Published on Web 06/12/2008
SiO2-Au-SiO2 Sandwich Nanoaggregates
Langmuir, Vol. 24, No. 14, 2008 7493
Scheme 1. Schematic Diagram of Synthesis of SiO2-Au-SiO2 Sandwich Nanoaggregates
gregates.20,21 Despite the success of physical and electrical methods that require expensive instruments, such as electronbeam lithography,22–24 chemical self-assembly methods have shown great potential for economical and simple fabrication of metallic nanoaggregates. Traditionally, metal colloids have been aggregated by adding the proper chemicals, for example, salts (NaCl, KCl) and surfactants,5,21,25,26 or applying a beam of UV-visible (UV-vis) light or laser to induce an accumulation.27–29 Colloidal self-aggregation at a two-phase interface has also been frequently used.20a However, these methods cannot control the shape and size of the nanoaggregates, resulting in a mixture of various irregular nanoaggregates. Although these nanoaggregates can enhance the surface plasmon, it is difficult to precisely control the extent of enhancement. Thus, the development of controllable and orderly metallic nanoaggregates using chemical methods remains a challenge. In the present study, we have found that SiO2-Au core-shell nanoparticles are highly effective building blocks for precise fabrication of nanoaggregates. Others have used building blocks modified with functional molecules, such as single-strand DNAs, antibodies or antigens, and ligands, for this purpose.30–34 The (20) (a) Hu, J.; Zhao, B.; Xu, W.; Fan, Y.; Li, B.; Ozaki, Y. J. Phys. Chem. B 2002, 106(25), 6500–6506. (b) Hu, J.; Zhao, B.; Xu, W.; Fan, Y.; Li, B.; Ozaki, Y. Langmuir 2002, 18(18), 6839–6844. (21) Yang, Y.; Shi, J.; Tanaka, T.; Nogami, M. Langmuir 2007, 23, 12042– 12047. (22) Su, K.-H.; Wei, Q.-H.; Zhang, X.; Mock, J. J.; Smith, D. R.; Schultz, S. Nano Lett. 2003, 3(8), 1087–1090. (23) Gunnarsson, L.; Rindzevicius, T.; Prikulis, J.; Kasemo, B.; Kall, M.; Zou, S.; Schatz, G. C. J. Phys. Chem. B 2005, 109(3), 1079–1087. (24) Jain, P. K.; Huang, W.; El-Sayed, M. A. Nano Lett. 2007, 7(7), 2080– 2088. (25) Creighton, J. A.; Blatchford, C. G.; Albrecht, M. G. J. Chem. Soc. Faraday Trans. 1979, 75, 790–797. (26) Siiman, O.; Bumm, L. A.; Callaghan, R.; Blatchford, C. G.; Kerker, M. J. Phys. Chem. 1983, 87, 1014–1023. (27) Niidome, Y.; Hori, A.; Takahashi, H.; Goto, Y.; Yamada, S. Nano Lett. 2001, 1(7), 365–369. (28) Kawasaki, M.; Hori, M. J. Phys. Chem. B. 2003, 107(28), 6760–6765. (29) (a) Bjerneld, E. J.; Murty, K.; Prikulis, V. G. K.; Ka¨ll, M. J. ChemPhysChem 2002, 1, 116–119. (b) Bjerneld, E. J.; Svedberg, F.; Ka¨ll, M. Nano Lett. 2003, 3(5), 593–596. (30) In Nanoparticles: Building Blocks for Nanotechnology (Nanostructure Science and Technology); Rotello, V., Ed.; Springer, New York, 2003; Chapter 7. (31) Zhang, H.; Hussain, I.; Brust, M.; Cooper, A. I. AdV. Mater. 2004, 16(1), 27–30. (32) Alivisatos, A. P.; Johnsson, K. P.; Peng, X.; Wilson, T. E.; Loweth, C. J., Jr.; Schultz, P. G. Nature 1996, 382, 609–611. (33) Storhoff, J. J.; Mirkin, C. A. Chem. ReV. 1999, 99, 1849–1862.
specific recognition reactions of the functional molecules linked multiple blocks together to form orderly nanoaggregates. Instead of using functionalized building blocks to promote precise assembly of nanoaggregates, we used a more versatile stabilizer in this study. As the concentration of the stabilizer was changed, aggregation of the SiO2-Au core-shell building blocks changed in a precise manner. These nanoaggregates were then immobilized with an outer silica matrix to produce very stable single, double, and triple building block clusters (Scheme 1). The photosignal enhancement of the nanoaggregates was verified by doping a NIR fluorescent probe molecule in the outer layer of the silica matrix. A NIR label was chosen for two reasons. NIR dyes usually show intrinsically low fluorescence emission characteristics and thus provide a good model to assess plasmon resonance enhancement. Moreover, the NIR region is a particularly favorable window for biological sensing and detection. Biological samples have low absorption, autofluorescence, and scattering, resulting in low background signals and deeper penetration of NIR radiation.35 The application of NIR enhanced metal nanostructures to the NIR spectral region would be a significant development methodology for the determination of biological samples.
2. Experimental Section 2.1. Materials. Tetraethylorthosilicate (TEOS), polyoxyethylene(10)isooctylphenylether[TritonX-100,4-(C8H17)C6H4(OCH2CH3)10OH], and methyl sulfoxide (DMSO) were purchased from Acros Organics. Sodium citrate, gold(III) chloride trihydrate (HAuCl4.3H2O, 99.9+%), hydroxylamine hydrochloride (98%, A.C.S grade), 3-aminopropyltriethoxysilane (APTS, 95%), sodium borohydride (>98%), polyvinylpyrrolidone molecule (PVP-10, average molecular weight of 10 kg/mol), and NIR 797 isothiocyanate (1′-bis(4-sulfobutyl)11-(4-isothiocyanatophenylthio)-3,3,3′,3′-tetramethyl-10,12-trimethyleneindotricarbocyanine monosodium salt) were purchased from Sigma-Aldrich Inc. Ammonia (28-30%, GR) was purchased from EM Industries Inc. 1-Hexanol (99+%) was purchased from Alfa Aesar. Potassium carbonate (K2CO3 · 1.5H2O, A.C.S grade), cyclohexane (HPLC grade), and ethanol (95%) were purchased from Fisher Scientific. MilliQ water (18.6 Ωcm-1) was used to make aqueous solutions. (34) Loweth, C. J.; Caldwell, W. B.; Peng, X.; Alivisatos, A. P.; Schultz, P. G. Angew. Chem. Int. Ed. 1999, 38, 1808–1812. (35) Umar, M. IEEE Eng. Med. Biol. Magn. 2004, 23(4), 58–66.
7494 Langmuir, Vol. 24, No. 14, 2008 2.2. Synthesis of Prebuilding Blocks of SiO2-Au Nanoparticles. First, the SiO2 cores were synthesized by using a reverse microemulsion method.3a To adsorb Au nanoparticles on the SiO2 surface, 50.0 µL of 3-aminopropyltrimethoxysilane (APTS) was added to the microemulsion to provide amino groups on the surface of SiO2 nanoparticle. The size of the SiO2 cores was adjusted by using different amounts of water. After effective washing, the SiO2 nanoparticles were resuspended into 40.0 mL of ethanol as a stock solution. The nanoparticle concentration was 3.8 × 1011 particles/ mL. Second, Au nanoparticles 4 ( 1 nm in diameter were synthesized on the basis of a literature method.36 The detailed procedure is described in the Supporting Information. The Au nanoparticle solutions were kept in a refrigerator at 4.0 °C before use. Third, 1.0 mL of stock SiO2 nanoparticle solution was dropwise added into 40.0 mL of Au nanoparticle solution with vigorous stirring. The Au nanoparticles were adsorbed on the SiO2 cores through electrostatic force after 6.0 min reaction. The surplus Au nanoparticles were separated by centrifuging at a speed of 6500 rpm for 12.0 min. The supernatant was carefully removed. The purplish red precipitate was SiO2-Au nanoparticles. Here, the particle suspended solution showed a dark red color. The prebuilding blocks were not stable. The precipitate was then resuspended into 10.0 mL of water in an ultrasonic bath for further growth of Au. 2.3. Synthesis and Stabilizing of the SiO2-Au Core-Shell Building Blocks. The unstable Au nanoparticles on the SiO2 core surface were grown in the gold growth solution to form a Au shell. The gold growth solution consisted of 1.0 mL of 1.0 × 10-2 g/mL chloroauric acid and 25.0 mg of K2CO3 in 90.0 mL of water. Under vigorous stirring, the solution turned transparent and colorless. Then, 10.0 mL of SiO2-Au nanoparticle aqueous solution (containing about 3.8 × 1011 particles) was added into the gold growth solution. The reaction started when 0.5 M of hydroxylamine hydrochloride was slowly added. The color of the mixture was first clear pink and then turned to purple and blue, and finally dark green, indicating an Au shell was produced. The total consumed hydroxylamine hydrochloride was 1.0 mL. To stabilize the SiO2-Au core-shell particles, PVP (0.1 g/mL) was added to the above solution. After an overnight stirring, the surplus PVP was removed by centrifuging at a speed of 3500 rpm for 15.0 min. Finally, the SiO2-Au core-shell nanoparticles were resuspended into 10.0 mL of EtOH as the stock building block solution. 2.4. Synthesis of the Sandwich Nanoaggregates. One, two, three, and multiple SiO2-Au building blocks congregated during the formation of the silica shell. A 2.5 mL portion of SiO2-Au building blocks was diluted to 10.0 mL using ethanol, and 0.12 mL of water, 4.0 µL of TEOS, and 1.0 mL of ammonia (29%) were added into the above solution. The SiO2-Au particles spontaneously aggregated during the process of formation of the silica shell. After 1 h reaction, the sample was centrifuged at a speed of 3500 rpm for 15.0 min. Finally, the particles were washed by ethanol at least three times. The thickness of the SiO2 shell was dependent on the amount of TEOS. 2.5. Purification of the Single, Double, and Triple Building Block Core Nanoaggregates. The separation and purification of the nanoaggregates were conducted by adjusting the centrifuge speeds. The single-, double-, triple-core and polycore aggregates were obtained from the precipitants at the centrifuge speeds of 1200-2000, 500-800, and below 500 rpm. Three rounds of centrifugation were preferred to obtain purified products. 2.6. Preparation of Dye-Doped Sandwich Aggregates. NIR 797 isothiocyanate (Sigma-Aldrich Co.) was chosen. To link the dye molecules into the silica layer, we first linked the NIR 797 to an aminosilane precursor. The details are described in the Supporting Information. The fluorescence spectra proved NIR 797 was doped into the silica matrix. The dye-doped sandwich particles have been prepared similarly as the development of the sandwich nanoaggregates, but the APTS-dye complex (40 µL, 0.9 mg/mL dye) was added at the last step. (36) In Bioconjugate Techniques; Hermanson, G. T., Ed.; Academic Press: San Diego, 1996; p 597.
Xu et al. 2.7. Characterization. The size and morphology of the nanoparticles were characterized using a Hitachi 7500 transmission electron microscope (TEM), operating at 80 kV, and a Hitachi 4700 field scanning electron microscope (SEM). TEM samples were prepared by dripping one drop of particle ethanol solution onto the TEM copper grid (200 meshes). The UV-visible spectra were obtained from the Shimadzu UV 2501 PC spectrophotometer. A Jobin-Yvon-Horiba Fluorometer 3 Model FL 3-11 spectrofluorometer and an Olympus IX 71 fluorescence microscope were used to image nanoaggregates and measure fluorescence signals. A 100× oil immersed lens was employed in the fluorescence microscope. The microscope was equipped with a high resolution CCD camera (Hamamatsu C9100-12 back-thinned electron multiplier CCD). The filter cube (OMEGA Optical, purchased from Leeds Precision Instruments Inc.) is composed of an exciter (HQ775/50x bandpass filter), a dichroic filter (Q810LP), and an emitter (HQ845/55m, bandpass filter). SimplePCI software with the Automated Image Capture-Camera Devices (AIC-CD) component was used to instruct the CCD to capture and view images. The centrifugation was conducted using an Eppendorf 5804 centrifuge. A comparison of the fluorescence enhancement of different types of sandwich particles was carried out by measuring the fluorescence intensity of a number of individual sandwich nanoaggregates on the fluorescence microscope. For instance, a drop of diluted single-core nanoparticle solution was placed on a glass slide. The fluorescence intensity of an individual nanoparticle was directly measured under the microscope using the Simple PCI software. The final average intensity was obtained from a number of the same kind of particles. Then, the morphology of the single-core particles was confirmed by SEM image. The fluorescence intensities of double- and triple-core sandwich particles were obtained using the same method.
3. Results and Discussion 3.1. Fabrication of the Building Block. We first developed a building block of a SiO2-Au core-shell nanoparticle. The fabrication of the building blocks included three major steps (Scheme 1): (1) preparation of a prebuilding block, (2) formation of a building block, and (3) stabilizing the block. Au is an excellent plasmonic material possessing long-term stability and biocompatibility.37,38 The LSPR band wavelengths of Au nanoparticles are in the visible region. To enhance NIR dye signals, a plasmon band in the NIR region was needed. SiO2 nanoparticles can induce a red shift of Au plasmon bands.39 Thus, the prebuilding block was prepared by adsorbing Au nanoparticles on a SiO2 nanoparticle surface. The SiO2 nanoparticle was modified with amine groups to provide positive charges for electrostatic adsorption of Au nanoparticles.40 A TEM image of the prebuilding blocks is shown in Figure 1a. The SiO2-Au nanoparticles were not stable. They would aggregate after they were removed from the ultrasonic bath in 30 min. The formation of the SiO2-Au building block was completed through Au nanoparticle growth on the surface of the prebuilding block. A gold growth solution, chloroauric acid, was mixed with the SiO2-Au nanoparticles. With a reducing reagent (hydroxylamine hydrochloride) added, the Au nanoparticles grew to form a uniform Au shell on the SiO2 nanoparticles (Figure 1b). 3.2. Shift Wavelengths of the Plasmon Band to the NIR Region. The wavelength of the plasmon band depends on the thickness of the gold shell and the size of the silica core. During (37) Faulk, W. P.; Taylor, G. M. Immunocytochemisty 1971, 8, 1081–1083. (38) In Colloidal Gold: Principles, Methods, and Applications; Hayat, M. A., Ed.; Academic Press, San Diego, 1989; Vol. 1. (39) (a) Prodan, E.; Nordlander, P.; Halas, N. J. Nano Lett. 2003, 3(10), 1411– 1415. (b) Oldenburg, S. J.; Jackson, J. B.; Westcott, S. L.; Halas, N. J. Appl. Phys. Lett. 1999, 75(19), 2897–2899. (c) Oldenburg, S. J.; Averitt, R. D.; Westcott, S. L.; Halas, N. J. Chem. Phys. Lett. 1998, 288, 243–247. (40) Osterloh, F.; Hiramatsu, H.; Porter, R.; Guo, T. Langmuir 2004, 20(13), 5553–5558.
SiO2-Au-SiO2 Sandwich Nanoaggregates
Langmuir, Vol. 24, No. 14, 2008 7495
Figure 2. (a) UV-vis spectra of the Au shell growth process on Au-SiO2 nanoparticles. The Au shell became thicker from 0 nm (curve 1) to 14 nm (curve 5). (b) UV-vis spectra of SiO2-Au core-shell nanoparticles with different sized SiO2 cores. The diameter of SiO2 nanoparticles is (1) 86 ( 5 nm, (2) 98 ( 4 nm, and (3) 130 ( 6 nm.
Figure 1. TEM images of (a) SiO2-Au nanoparticles, (b) SiO2-Au core-shell nanoparticles, (c) PVP-stabilized SiO2-Au nanoparticles, and (d) SiO2-Au-SiO2 sandwich nanoaggregates.
the process of Au growth, the solution color changed from pink to purple, blue, and finally dark green, which was consistent with the literature.39 The color changes represented the red shifts of the plasmon peak as the Au shell became thicker. Finally, a 14 nm Au shell was formed on the SiO2 nanoparticle surface with a plasmon band at 626 nm (curve 5 in Figure 2a). To shift the plasmon peak further to the NIR region, larger sizes of SiO2 nanoparticles were needed. As the size of the SiO2 core was increased from 86 ( 5 nm to 130 ( 6 nm in diameter, their plasmon band was shifted from 626 to 794 nm (Figure 2b). Herein, we chose the SiO2-Au core-shell nanoparticle as a building block with a SiO2 core of 130 ( 6 nm in diameter. When the building block was doped into silica matrix to form the aggregates, the plasmon band shifted to a longer wavelength. 3.3. Stabilizing the Building Block and Orderly Formation of Nanoaggregates. The building blocks spontaneously formed
various irregular aggregates (Figure 1b). To fabricate orderly nanoaggregates, a short chain PVP molecule was employed to modify the surface of the building block. Different amounts of PVP were mixed with the building block aqueous solution and each was reacted for 12 h at a low stirring speed (400 rpm). Through adsorption of PVP hydrophilic side groups on the SiO2-Au nanoparticle, a PVP layer was formed. The dispersibility of the SiO2-Au core-shell nanoparticles was improved dramatically (Figure 1c). PVP is an amphiphilic and nonionic polymer that has been used as a stabilizer for preventing aggregation of nanomaterials, such as metal nanoparticles,41–45 metal oxides,46 and polystyrene.47 In our design, the function of PVP was not only as a stabilizer but most importantly as an adjuster for manipulating the extent of aggregation. By adjusting amount of PVP, as well as controlling the final silica coating process, the desired aggregates would be produced. The orderly formation of the nanoaggregates progressed by doping the SiO2-Au core-shell particles in a silica matrix. Silica was chosen because it has no absorption in the visible and NIR region. Meanwhile, the optical properties of building blocks were superiorly protected by the silica matrix. A modified Sto¨ber method was employed to dope the building blocks into a silica matrix.42 The SiO2-Au building blocks were first dispersed in (41) (a) Graf, C.; Vossen, D. L. J.; Imhof, A.; Van Blaaderen, A. Langmuir 2003, 19(17), 6693–6700. (b) Graf, C.; Dembski, S.; Hofmann, A.; Ruehl, E. Langmuir 2006, 22(13), 5604–5610. (42) (a) Pastoriza-Santos, I.; Gomez, D.; Perez-Juste, J.; Liz-Marzan, L. M.; Mulvaney, P. Phys. Chem. Chem. Phys. 2004, 6(21), 5056–5060. (b) Rodrı´guezFerna´ndez, J.; Pastoriza-Santos, I.; Pe´rez-Juste; Garcı´a de Abajo, J.; F. J.; LizMarza´n, L. M. J. Phys. Chem. C 2007, 111(36), 13361–13366. (43) Chen, M.; Xing, Y. Langmuir 2005, 21(20), 9334–9338. (44) Haas, I.; Shanmugam, S.; Gedanken, A. J. Phys. Chem. B. 2006, 110(34), 16947–16952. (45) Me´traus, G. S.; Mirkin, C. A. AdV. Mater. 2005, 17, 412–415. (46) Pattanaik, M.; Bhaumik, S. K. Mater. Lett. 2000, 44, 352–360. (47) Smith, J. N.; Meadows, J.; Williams, P. A. Langmuir 1996, 12, 3773– 3778.
7496 Langmuir, Vol. 24, No. 14, 2008
Xu et al.
Figure 3. (a) The relationship of PVP concentration with the percentages of different type of nanoaggregates. (b) TEM and SEM images of single-core sandwich nanoparticles (1), double-core aggregates (2), and triple-core and polycore aggregates (3). (c) Schematic diagram of the effect of PVP concentration on the formation of aggregates.
the Sto¨ber synthesis solution and then started a slow aggregation. The extent of aggregation was controlled by the amount of PVP on the building block and the time of aggregation. Both conditions were controllable. The aggregation was stopped when TEOS was added to the solution. To avoid formation of polycore aggregates, the aggregation was allowed for 10 min. The polymerization of TEOS produced a silica layer on the surface of SiO2-Au, forming a SiO2-Au-SiO2 sandwich nanostructure. With a suitable PVP layer on the building blocks and using the optimal aggregation time, the produced aggregates were largely limited to three types: single-, double-, and triple-core sandwich nanostructures (Figure 1d). The percentages of each type of nanostructures were tunable as the synthesis conditions changed. The conditions included: (1) the PVP concentration, (2) the amount of TEOS, and (3) the concentration of SiO2-Au building blocks. Among them, the PVP concentration was a critical factor. 3.4. Effect of PVP Concentration on the Formation of Nanoaggregates. The effect of PVP concentration on the formation of sandwich nanoaggregates is shown in Figure 3a. (PVP concentration should be calculated as the unit of number of PVP molecules per surface area of nanoparticles. To avoid errors, we prefer to directly present it in the unit of mg/mL.) When low concentrations of PVP were used, a large portion of the products was irregular polycore aggregates. For instance,
approximately 59% of the polycore aggregates was formed when the concentration of PVP e 0.20 mg/mL. Meanwhile, the percentages of the single-, double-, and triple-core nanoaggregates were about 14%, 19%, and 8%. As the PVP concentration was increased, the percentage of polycore aggregates was greatly reduced. Finally, as the PVP concentration was over 3.5 mg/mL, the irregular polycore nanoaggregate was limited to less than 5%. The yield of the single-core sandwich nanoparticles was adjustable in the range from 16% to 57%. The maximum percentage of the double-core aggregates was 33%. The triplecore aggregates were adjustable in the range from 7% to 16%, which was relatively small compared to the single- and doublecore nanostructures. On the basis of the above results, one can adjust the amount of PVP to favor the yield of the desired nanoaggregates. Although the product was a mixture, the simple compositions made the separation of the three aggregates feasible. Each type of the nanoaggregate could be purified based on their distinct weights and sizes. Several separation methods were effective, such as size-exclusive chromatography, gravitational field-flow fractionation, and centrifugation. Centrifugation was the most simple and economical way to separate the three types of nanoaggregates. Using different centrifuge speeds, the single, double, and triple building block aggregates were separated
SiO2-Au-SiO2 Sandwich Nanoaggregates
Langmuir, Vol. 24, No. 14, 2008 7497
Table 1. Effect of Centrifuge Speed on Separation of Nanoaggregates percentage centrifuge speed (rpm) single core double core triple core and polycore
