Aspect-Ratio Dependence on Formation Process ... - ACS Publications

Feb 12, 2010 - Department of Chemistry, Faculty of Education, Aichi University of Education, Hirosawa 1, Igaya, Kariya, Aichi 448-8542, Japan, and Div...
1 downloads 4 Views 440KB Size
Download PDF
3804

J. Phys. Chem. C 2010, 114, 3804–3810

Aspect-Ratio Dependence on Formation Process of Gold Nanorods Studied by Time-Resolved Distance Distribution Functions Takeshi Morita,*,† Eiichi Tanaka,† Yukihiro Inagaki,† Hiroyasu Hotta,† Rie Shingai,† Yoshikiyo Hatakeyama,‡ Keiko Nishikawa,‡ Hiromi Murai,† Hirofumi Nakano,† and Kazuyuki Hino† Department of Chemistry, Faculty of Education, Aichi UniVersity of Education, Hirosawa 1, Igaya, Kariya, Aichi 448-8542, Japan, and DiVision of Nanoscience, Graduate School of AdVanced Integration Sciences, Chiba UniVersity, Yayoi, Inage-ku, Chiba 263-8522, Japan ReceiVed: October 18, 2009; ReVised Manuscript ReceiVed: January 10, 2010

Gold nanorods with aspect ratios of 2, 4, and 6 were generated by a seed-mediated growth method developed by Jana et al. and improved by Nikoobakht and El-Sayed. Nanorod formation during synthesis was investigated by time-resolved measurements of small-angle X-ray scattering (SAXS) and in situ UV-vis-NIR absorption. A sample holder made entirely of titanium was newly constructed for the simultaneous measurements. Distance distribution functions were calculated from the SAXS profiles, and the growth in nanorod length as a function of reaction time was determined without having to approximate or simplify the actual nanorod shape. Structural anisotropy, which corresponds to aspect ratio of nanorods, was also evaluated from the functions. For formed nanorods with aspect ratios in the range of 2-4, the time dependence of anisotropy shows maxima. In contrast, for formed aspect-ratio 6 nanorods, a maximum is not observed, and anisotropy continuously increases until an early reaction-equilibrium is reached. Growth kinetics for the variation in the aspect ratios of the formed nanorods is evaluated based on the length growth rate and the width growth rate, each considered individually. Finally, structural changes of nanorod end-caps during synthesis are discussed. Introduction Nanorods, rodlike shaped nanoparticles, constitute an important class of nanomaterials. Surface plasmon resonance, individually localized along the major and transversal axes, and strengthened electric fields at both ends of the rod result from their nonspherical structure.1–4 Beyond size control of spherical nanocrystals, shape control of nanorods leads to greater tunability of their optical properties.5 In addition, the unique endcap portion of a nanorod plays an important role in determining its optical properties6 and its inter-rod linear orientationsthat is, head-to-tail interactionsused in a biorecognition system.7 An understanding of the growth kinetics of nanomaterials during chemical synthesis is essential for development of structural controls and synthesis methods. The time dependence of growth has been analyzed by UV-vis spectroscopy for gold nanorods generated by a seed-mediated growth method.8–10 In addition to the optical properties, in situ structural investigations have been performed using various experimental methodologies such as wide-angle X-ray scattering (WAXS), small-angle X-ray scattering (SAXS), X-ray absorption fine structure (XAFS), energy-dispersive XAFS (DXAFS), transmission electron microscope (TEM), and combinations of these and/or other techniques. For spherical gold particles, time dependences of growth and size distribution have been elucidated in detail by Rao et al.11,12 In situ observations by SAXS/WAXS and UV-vis spectroscopy have also been reported and the progress of particle generation, especially in chemical reactions with high rate constants, has been clarified by means of a novel measurement * To whom correspondence should be addressed. E-mail: tmorita@ auecc.aichi-edu.ac.jp. † Aichi University of Education. ‡ Chiba University.

system.13 The mechanisms of nanoparticle formation by photoreduction of Rh(III) and Pd(II) in a polymer solution have been investigated by in situ DXAFS measurement.14 Growth in the length of ZnO nanorods has been investigated by WAXS using the Scherrer equation.15 The growth kinetics of ZnO nanorods, investigated by in situ SAXS and TEM observations, has been reported by Biswas et al.16,17 These papers address the time dependence of the length and its distribution of ZnO nanorods generated in alkaline solutions of tetrakis(hydroxymethyl) phosphonium chloride (THPC). Recently, the growth kinetics of Au and AuCu nanorods generated by a seedmediated growth method was investigated by in situ SAXS and UV-vis-NIR absorption measurements.18 Time dependences of growth in length and width were evaluated independently for formed nanorods with aspect ratio (AR) of 3-3.5 in the final product, and the aspect ratio was shown to take a maximum at 8-10 min during synthesis. SAXS techniques have been widely applied to investigate the growth kinetics of nanorods as well as their basic characteristics. SAXS structural investigations, including time-resolved measurements, generally assume that nanorods have a cylindrical shape with two flat ends. However, TEM observations show that the structure is actually an end-capped cylinder. We have used SAXS techniques to determine the nanorod structure using the end-capped cylinder model developed by Kaya and de Souza.19,20 Our investigation shows that the end caps of gold nanorods with AR of 4 constitute about 20% of the overall length.21 Furthermore, TEM observations show that nanorod samples usually contain other different-shapedsthat is, cubic or sphericalslarge-volume particles. Our structural investigations, which take into consideration the effect of a small population of other different-shaped particles with large volume on the scattering profile, indicate that such particles, even though

10.1021/jp909965t  2010 American Chemical Society Published on Web 02/12/2010

Formation Process of Gold Nanorods

Figure 1. Distance distribution function obtained from the experimental SAXS profile (solid line), separated component of the nanorods (dashed line), and separated component of other different-shaped particles with large volume (dot-dashed line). The inset illustrates the obtained structure parameters of the nanorods.21 While the overall profile of P(r) is seriously affected by other differently shaped particles, the maximum lengthsthat is r value at P(r) ) 0sis estimated to be the same. Therefore, nanorod lengths are correctly evaluated from P(r).

present in only very small proportions, significantly affect the structural characteristics of nanorods.21 The present study considers the formation process of gold nanorods, studied by time-resolved distance distribution functions obtained from SAXS profiles. A distance distribution function P(r) gives the most reliable information on the structure of a scattering center (that is, a scatterer) in real space, because no approximations or assumptions are used in the function. The value of r at which P(r) ) 0 is exactly equal to the maximum length of the scatterer. For nanorods, the maximum length reliably corresponds to the major-axis length, even though other different-shaped particles, such as cubic and spherical nanoparticles, exist in the system. Figure 1 shows the contribution of other different-shaped particles to P(r). While the overall profile of P(r) is seriously affected by such particles, the maximum lengthsthat is the r value at P(r) ) 0sis estimated to be the same. Therefore, time-resolved analysis of the maximum length of a scatterer, evaluated from P(r), enables us to determine the progress of nanorod formation during synthesis without simplification of actual nanorod shape. In the present study, gold nanorods with three ARs (2, 4, and 6) were prepared for investigation of aspect-ratio dependence. UV-vis-NIR spectra were measured simultaneously by a newly designed and constructed system involving a sample holder made of highly corrosion-resistant titanium. Aspect ratios were calculated from the optical spectra as reported in the literature.8,22 On the basis of their evaluated length and aspect ratio, the progress of nanorod formation and the differences among the three aspect ratios are described assuming that a nanorod is an end-capped cylinder. Finally, structural changes at both ends of the nanorod during synthesis are discussed. Experimental Section To confirmation of reproducibility, we performed experiments at three different beam-times in the synchrotron facility. Gold Nanorod Synthesis. We prepared gold nanorods using a seed-mediated growth method developed by Jana et al.23–25

J. Phys. Chem. C, Vol. 114, No. 9, 2010 3805 and improved by Nikoobakht and El-Sayed.8 The method enables us to control the length of a nanorod by changing the silver ion content in the growth solution. To generate nanorods with ARs of 2 and 4, we prepared a growth solution containing 5.0 mL of HAuCl4 (1.0 mM) and 5.0 mL of cetyltrimethylammonium bromide (CTAB, 0.20 M). To this solution, we added 50 and 200 µL of AgNO3 (4.0 mM) for ARs 2 and 4, respectively. And then, we mixed 110 µL of ascorbic acid (78.8 mM) and 100 µL of seed (0.236 mM as gold content) into the respective solutions, which is somewhat different from the literature.8 To generate nanorods with AR 6, we prepared a growth solution of 5.0 mL of benzyldimethylhexadecylammonium chloride (BDAC) plus CTAB as surfactant (in a BDAC/CTAB molar ratio of 2.5), to which we added 5.0 mL of HAuCl4 (1.0 mM) and 200 µL of AgNO3 (4.0 mM). Into the solution, we mixed 80 µL of ascorbic acid (78.8 mM) and 100 µL of seed (0.236 mM as gold content), which is also different from the literature.8 During synthesis, the temperature of the sample was controlled by circulating water at 27 ( 0.1 °C. Figure 2 shows typical TEM images of the gold nanorods synthesized by these procedures. Titanium Sample Holder. We constructed a sample holder for simultaneous measurements of SAXS and UV-vis-NIR absorption, shown in Figure 3. The holder was made of pure titanium (JIS, grade 2) because, due to the high corrosion activity of HAuCl4, other materials such as stainless steel and titanium alloy with lower corrosion resistance are not suitable for this purpose. For the X-ray window, we initially used a single-crystal diamond (De Beers Industrial Diamond Division Co., type Ιb, 4.5 mm in diameter and 0.3 mm in thickness). We exchanged this for SUPERIO-UT plastic film (Mitsubishi Plastics, Inc., type-F, 25 µm in thickness) to improve the small-angle resolution and detectable scattering intensity. We strongly recommend that SUPERIO-UT should be used for window materials in SAXS experiments under atmospheric pressure due to its extremely small parasitic scattering in the small-angle region, high chemical stability, and high heat resistance. The window materials were sealed to the titanium parts by means of a highly corrosion-resistant compound (Aremco Products, Inc., Aremoco bond-315). Polycrystalline quartz glass (5.0 mm in diameter and 1.0 mm in thickness) was used for the windows in the UV-vis-NIR absorption experiments. View ports were installed in the top and bottom of the holder to allow observation of the reaction solution. The holder was kept small in size to attain gentle mixing of seed and growth solutions8,23–25 when the holder is turned upside down. SAXS Measurements. SAXS experiments were carried out using the apparatus at the BL-15A station,26 photon factory (PF) at the High Energy Accelerator Research Organization (KEK), Tsukuba. An imaging plate (IP; FUJIFILM Co., BAS-MS2040; 200 × 400 mm2 in size, positional resolution of 100 µm) was used for a detector. The choice of detector is based on the following three requirements: (1) wide dynamic range, because the SAXS nanorod intensities increase significantly in the smallangle region; (2) long detectable area, because measurement accuracy in the smaller s-region requires a longer camera length for small-angle resolution (on the other hand, scattering data over a wide-angle range is essential for Fourier transform calculations of distance distribution functions); (3) twodimensional area detector, because the experiments involve timeresolved measurements examined over a short exposure time, which is admitted by the integration of the two-dimensional

3806

J. Phys. Chem. C, Vol. 114, No. 9, 2010

Morita et al.

Figure 3. Cross sections of the sample holder for simultaneous measurements of SAXS and UV-vis-NIR absorption: (a) horizontal plane; (b) vertical plane. The holder is made of pure titanium. Air space is necessary for gentle mixing of seed and growth solutions in the holder and for holding bubbles generated by the surfactant during mixing.

Figure 2. Typical TEM images of gold nanorods in the final product for various aspect ratios: (a) AR 2, (b) AR 4, and (c) AR 6.

data. Furthermore, although it is important to observe the early stage of a reaction in order to clarify the progress of rod formation, SAXS intensities of the particles are relatively weak during that period. We constructed a frame-moving apparatus for the timeresolved measurements using an IP. The apparatus takes up to 7 frames per IP; the time resolution of the X-ray exposure interval is 5 s. X-ray signals recorded in the IP were read out by an IP reader (FUJIFILM Co., BAS-2500) with careful

attention to fading characteristics of the IP. The camera length was 2400 mm. The observable s-region was 0.007-0.3 Å-1, with scattering parameter s ) 4π sin θ/λ (2θ: scattering angle). Accumulation times for each exposure were 50, 60, and 120 s for ARs 2, 4, and 6 nanorods, respectively. During synthesis, the X-ray absorption factor of the solution continuously changes with the progress of the chemical reaction. We had constructed an in situ beam monitor apparatus27 for the SAXS beamline, and the absorption factor of the sample, which seriously affects the precision of the background and absorption corrections, was measured simultaneously. X-ray profiles from the sample holder filled only with water were used to determine and subtract out background intensity and absorption. The path length of the sample was 2.0 mm. Simultaneous Measurements of UV-vis-NIR Absorption. UV-vis-NIR light emitted by the light source (Hamamatsu Photonics, L10671; output range 200-1600 nm) was guided to the sample holder by optical fiber. Parallel lenses were set at both the incident and output sides of the sample holder. The output signal was guided from the holder to a charge-coupled device (CCD) array spectrometer (BW TEK, inc., BRC 112; resolution of 1.3 at 546 nm), and continuously recorded at intervals of 20 s. The path length of the sample was 6.0 mm. We ensured that the probe light did not affect the nanorod synthesis, and that absorption spectra obtained for the final product corresponded to our TEM observations. Results and Discussion Nanorod Length and Anisotropy Obtained from SAXS. Figure 4 shows time-resolved SAXS intensity I(s) as a function of reaction time. In Figure 4a, the extreme increases in intensity

Formation Process of Gold Nanorods

Figure 4. (a) Time-resolved SAXS intensity I(s) as a function of reaction time for AR 2 at typical measured points of 105, 165, 225, 285, 345, 405, and 525 s. (b) Time dependence of I(s) at s ) 0.02 Å-1 for ARs of 2 (red triangles), 4 (blue squares), and 6 (black circles).

in a small s-region suggest that nanorod formation and growth are occurring in the solution. In Figure 4b, for ARs of 2 and 4, the intensities increase sharply with reaction time immediately after the beginning of the reaction; for ARs of 2 and 4, intensities increase up to 525 and 1370 s, respectively, and then remain almost constant. In contrast, for AR 6, intensity increases gradually up to 4000 s and then continues to increase slightly. The intensity change is much slower for AR 6 than for ARs of 2 and 4. The distance distribution function P(r) is calculated by Fourier transform of a SAXS profile.28 In the present calculation, the exponential term exp(-Bs2) (B, damping factor) was added to remove the termination effect of the transform.29,30 Other procedures of the calculation are the same as reported in our previous papers.21,30,31 Figure 5 shows calculated P(r) as a function of reaction time. As mentioned above, the value of r at which P(r) ) 0 corresponds to nanorod length. The anisotropy of the particlessthat is, the degree to which the shape is rodlikesis derived from the ratio of the r value at the peak top to that at P(r) ) 0. For a spherical particle, the theoretical ratio is 0.525.28 The inset in Figure 5a shows the enlarged profile of P(r) at the beginning of the reaction for AR 6. The ratio is estimated to be 0.524 and the size to be 31 Å, which corresponds to the structure of the spherical seed particle and confirms the validity of the SAXS measurements and P(r) calculations. The progress of growth from seed particle to nanorod is precisely observed from the change in P(r), especially for AR 6.

J. Phys. Chem. C, Vol. 114, No. 9, 2010 3807

Figure 5. Time-resolved distance distribution function P(r) obtained from measured SAXS profiles for various aspect ratios: (a) AR 6, (b) AR 4, and (c) AR 2. The inset is the enlarged profile corresponding to the seed spherical particles. The particle diameter is estimated to be 31 Å. Measure-point times are the same as those in Figure 4b.

Figure 6. Nanorod lengths (solid symbols) and the ratio of r at which P(r) is at a maximum to that at which P(r) ) 0 (open symbols) for (a) AR 4 and (b) AR 2 as a function of reaction time. The ratios are ∼0.5 at the beginning of the reaction and decrease to their respective minima, and then increase gradually.

Figure 6 shows the change in length and anisotropy of growing nanorods for ARs of 2 and 4 as a function of reaction time. The time dependence of the change in length shows that growth rapidly proceeds immediately after the beginning of the reaction, up to 320 Å at 400 s for AR 2 and 420 Å at 1400 s

3808

J. Phys. Chem. C, Vol. 114, No. 9, 2010

Figure 7. (a) Time dependence of the major and transversal axes plasmon resonance for ARs of 2 (red triangles) and 4 (blue squares). (b) Aspect ratios as a function of reaction time. 9 and 2 depict calculated values using the equation reported by Brioude et al.;22 O represents calculated values using relationships based on experimental data by Nikoobakht and El-Sayed;8 ) represents calculated values using relationship based on experimental data by Brioude et al.22

for AR 4. Anisotropic, rodlike shaped particles grow from the spherical seed particle; the ratio reaches minima around at 200 s for AR 2 and 410 s for AR 4. The same trend in the change of growing particle with reaction time is observed in three runs for ARs in the range 2-4 in our other experiments. This is the first observation of such phenomena for gold nanorods based on structural investigation. For AuCu nanorods, observation of these phenomena by SAXS investigation, though using other analytical procedures, has been reported.18 Aspect Ratio Obtained from Optical Properties. We recently performed structural investigations taking into account the effect of a small population of other different-shaped particles with large volume. Analysis without consideration of variations in shape and size yields AR 2.2; analysis with consideration of such variations yields AR 3.8 (length 305 Å/width 81 Å), as shown in the inset of Figure 1. The longitudinal plasmon maximum of the analyzed sample has been determined experimentally to be 774 nm.32 This value is consistent with AR 3.8, for which the calculated wavelength is 779.5 nm.22 Therefore, longitudinal plasmon maxima are not affected by other different-shaped particles with large volume in the solution, and the aspect ratio of nanorods separated from such particles is correctly determined by the wavelength. Figure 7 shows the change in plasmon resonance for major and transversal axes and the calculated aspect ratio of the nanorods as a function of reaction time. Maxima vary according to the aspect ratio of the final product and are at 190 s (3.2 min) for AR 2 and 405 s (6.8 min) for AR 4. A similar trend has been reported for ARs of 3-3.5 by Henkel et al.18 Investigations by optical absorption and SAXS analysis show a crossover of growth modes from length extension to alldirections extension at 8-12 min for AR ∼3.18

Morita et al.

Figure 8. (a) Lengths and widths for ARs of 2 (red triangles) and 4 (blue squares), (b) growth rate for AR 4, and (c) growth rate for AR 2 as a function of reaction time. Solid symbols and open ones refer to changes in length and width, respectively.

Progress of Nanorod Formation and AR Dependence. Because we take into consideration the effect of other differentshaped particles in the reaction solution, we are confident of the nanorod length obtained from P(r) and the aspect ratio evaluated from longitudinal plasmon maxima, as supported by our recent paper.21 On the basis of the combined information of nanorod length and aspect ratio, we can discuss growth in length and width for nanorods without any approximation or simplification for the actual nanorod shape. Hereafter, we use aspect ratios calculated from the equation reported by Brioude et al.22 for structural analyses. Calculated values are in good agreement with those obtained from our TEM observations. From the obtained time dependence of length and width, we can calculate the growth rate in Å/s from the derivative by the following equation

{

∂Li Li+1 - Li 1 ) (t - ti-1) + ∂ti (ti+1 - ti-1) ti+1 - ti i Li - Li-1 (t - ti) ti - ti-1 i+1

}

(1)

where t is reaction time and L is length. Equation 1 gives the rate not at the average measure-point time but rather at the exact measure-point time. Figure 8 shows changes in length and width growth rates as a function of reaction time. At the beginning of the reaction, the length growth rate for AR 2, 1.8 Å/s, is larger (by a factor of 2.3) than that for AR 4, 0.8 Å/s. The width growth rate for AR 2, 0.4 Å/s, is much larger (by a factor of 4.0) than that for

Formation Process of Gold Nanorods

J. Phys. Chem. C, Vol. 114, No. 9, 2010 3809

Figure 9. Nanorod length (open circles) and longitudinal plasmon maximum (solid circles) for AR 6 as a function of reaction time.

AR 4, 0.1 Å/s. In contrast, the ratio of length growth rate to width growth rate shows the opposite trend: the ratio for AR 2 is smaller than that for AR 4. For AR 2, both length and width growth rates are high. For AR 4, the length growth rate is high and the width growth rate is low, and so the longer rod is formed. Figure 9 shows nanorod length and longitudinal plasmon maxima for AR 6 as a function of reaction time. The maxima continuously shift to longer wavelength until, after 5000 s, they remain constant. This suggests that the growth crossover, observed for ARs of 2 and 4 in the present study and for AR ∼3 in the literature,18 does not occur for AR 6 and growth in length predominates over growth in width. For AR 2 and 4, growth process shown in Figure 8a does not show good relation to growth model of diffusion-limited Ostwald ripening.33,34 This suggests that diffusion-limited reaction model is not proper to the present reaction and combined diffusion and surface reaction model should be applied to the growth process, as discussed in detail in the literature.16,17 On the other hand, for AR 6, sigmoidal reaction model35 shows good relationship, especially in early stage, compared to diffusion and surface reaction model. It is well-known that the sufficient correlation for sigmoidal growth model corresponds to existence of induction period, at which nuclei are formed.36 Therefore, the reason why SAXS intensities and nanorod lengths do not increase up to 1000 s is that this period is in the induction stage, at which the formation of small nuclei gradually occurs in the AR 6 reaction solution. End-Cap Structure during Rod Formation. In the present study, we investigate rod formation without having to approximate or simplify the actual nanorod shape. Therefore, the obtained information reflects structural changes at both ends of the nanorods during synthesis. Nanorod lengths determined in the present study remain almost constant in the final stage of the reaction, as shown in Figure 6. As mentioned above, the length corresponds exactly to the maximum length of the scatterer. For a flat-end cylinder with rod height H and rod diameter R, the maximum length estimated from P(r) is (H2 + R2)1/2. For a globular end-capped cylinder, which is the actual shape of a nanorod, the maximum length is H. In our investigation, the radius of the end-cap is approximately equal to that of the cylinder part of the nanorod and the end-cap is slightly drawn into the cylinder.21 If structural changes at the both ends occur in the final stage, the maximum length obtained from P(r) also changes similarly. The change

Figure 10. Formation processes for gold nanorods with ARs of 2, 4, and 6.

for AR 2 is estimated to be 23 Å, 6.8% of the nanorod length; that for AR 4 is estimated to be 10 Å, 3.5% of the nanorod length. However, this difference in estimated value for these two final stages was not observed. Furthermore, the length for AR 2 changes within 10 Å in the final stage. Therefore, we consider that the structural change in the end-cap does not occur and does not contribute significantly to the gradual blue shift of the optical spectrum in the final stage, shown in Figure 8. Rather, we suggest that the blue shift in the final stage is caused by a slight growth in width and the associated change in aspect ratio. From the present study, it is not clear whether structural changes in both ends occur in the early and middle periods of the reaction. In situ TEM study is necessary to clarify this question. In other synthesis methods, in situ TEM observations have shown that a globular end-cap is already formed early in the reaction.17,37 Conclusions Formation and growth of gold nanorods with ARs of 2, 4, and 6 generated by seed-mediated growth method were clarified based on time-resolved distance distribution functions obtained from SAXS profiles without having to approximate or simplify the actual nanorod shape. The time dependence of anisotropy shows maxima for nanorods with final-product ARs in the range 2-4. A maximum is not observed for nanorods with AR 6. The mechanism for the variation in aspect ratios of the formed nanorods is evaluated on the basis of the length growth rate and the width growth rate, considered individually. Figure 10 shows the mechanisms of nanorod formation for ARs of 2, 4, and 6. The following tendencies are evident. (1) For AR 2, both length growth rate and width growth rate are large. (2) For AR 4, the length growth rate is greater than the width growth rate, and so a longer rod is formed. (3) Reaction rates descend in the following order of aspect-ratio: 2 > 4 . 6. For both ends of nanorods with ARs of 2 and 4, structural changes of end-cap do not seem to contribute significantly to the gradual blue shift of the optical spectrum in final stage of the reaction, and the blue shift is thought to be caused by slight growth in width and the associated change in aspect ratio.

3810

J. Phys. Chem. C, Vol. 114, No. 9, 2010

Acknowledgment. We thank Professor H. Okamoto and Dr. K. Imura (IMS) for valuable comments on sample preparation and Professor N. Nishi and Mr. S. Numao (IMS) for TEM observation of the sample. We are grateful to Professor Y. Amemiya and Professor Y. Shinohara for helpful suggestions regarding construction of the X-ray detector. We are grateful to PF at KEK for providing the opportunity to perform the SAXS experiments. (No. 2008G552) This work was supported by grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan (Special Funds for Education and Research). References and Notes (1) Torigoe, K.; Esumi, K. Langmuir 1992, 8, 59. (2) Daniel, M.-C.; Astruc, D. Chem. ReV. 2004, 104, 293. (3) Murphy, C. J.; Sau, T. K.; Gole, A. M.; Orendorff, C. J.; Gao, J.; Gou, L.; Hunyadi, S. E.; Li, T. J. Phys. Chem. B 2005, 109, 13857. (4) Alekseeva, A. V.; Bogatyrev, V. A.; Khlebtsov, B. N.; Mel’nikov, A. G.; Dykman, L. A.; Khlebtsov, N. G. Colloid J. 2006, 68, 661. (5) Chang, S.-S.; Shih, C.-W.; Chen, C.-D.; Lai, W.-C.; Wang, C. R. C. Langmuir 1999, 15, 701. (6) Prescott, S. W.; Mulvaney, P. J. Appl. Phys. 2006, 99, 123504. (7) Chang, J.-Y.; Wu, H.; Chen, H.; Ling, Y.-C.; Tan, W. Chem. Commun. 2005, 1092. (8) Nikoobakht, B.; El-Sayed, M. A. Chem. Mater. 2003, 15, 1957. (9) Sau, T. K.; Murphy, C. J. Langmuir 2004, 20, 6414. (10) Pe´rez-Juste, J.; Liz-Marza´n, L. M.; Carnie, S.; Chan, D. Y. C.; Mulvaney, P. AdV. Funct. Mater. 2004, 14, 571. (11) Seshadri, R.; Subbanna, G. N.; Vijayakrishnan, V.; Kulkarni, G. U.; Ananthakrishna, G.; Rao, C. N. R. J. Phys. Chem. 1995, 99, 5639. (12) Biswas, K.; Varghese, N.; Rao, C. N. R. Small 2008, 4, 649. (13) Abe´cassis, B.; Testard, F.; Spalla, O.; Barboux, P. Nano Lett. 2007, 7, 1723. (14) Harada, M.; Inada, Y. Langmuir 2009, 25, 6049. (15) Zhu, Z.; Andelman, T.; Yin, M.; Chen, T.-L.; Ehrlich, S. N.; O’Brien, S. P.; Osgood, Jr., R. M. J. Mater. Res. 2005, 20, 1033.

Morita et al. (16) Biswas, K.; Das, B.; Rao, C. N. R. J. Phys. Chem. C 2008, 112, 2404. (17) Biswas, K.; Varghese, N.; Rao, C. N. R. J. Mater. Sci. Technol. 2008, 24, 615. (18) Henkel, A.; Schubert, O.; Plech, A.; So¨nnichsen, C. J. Phys. Chem. C 2009, 113, 10390. (19) Kaya, H. J. Appl. Crystallogr. 2004, 37, 223. (20) Kaya, H.; de Souza, N.-R. J. Appl. Crystallogr. 2004, 37, 508. (21) Morita, T.; Hatakeyama, Y.; Nishikawa, K.; Tanaka, E.; Shingai, R.; Murai, H.; Nakano, H.; Hino, K. Chem. Phys. 2009, 364, 14. (22) Brioude, A.; Jiang, X. C.; Pileni, M. P. J. Phys. Chem. B 2005, 109, 13138. (23) Jana, N. R.; Gearheart, L.; Murphy, C. J. AdV. Mater. 2001, 13, 1389. (24) Jana, N. R.; Gearheart, L.; Murphy, C. J. Chem. Commun. 2001, 617. (25) Jana, N. R.; Gearheart, L.; Murphy, C. J. J. Phys. Chem. B 2001, 105, 4065. (26) Amemiya, Y.; Wakabayashi, K.; Hamanaka, T.; Wakabayashi, T.; Matsushita, T.; Hashizume, H. Nucl. Instrum. Methods 1983, 208, 471. (27) Morita, T.; Tanaka, Y.; Ito, K.; Takahashi, Y.; Nishikawa, K. J. Appl. Crystallogr. 2007, 40, 791. (28) Glatter, O.; Kratky, O. Small Angle X-ray Scattering; Academic Press: London, 1982. (29) Waser, J.; Schomaker, V. ReV. Mod. Phys. 1953, 25, 671. (30) Fukuyama, K.; Kasahara, Y.; Kasahara, N.; Oya, A.; Nishikawa, K. Carbon 2001, 39, 287. (31) Fukuyama, K.; Nishizawa, T.; Nishikawa, K. Carbon 2001, 39, 1863. (32) Hino, K., et al. Preparation of manuscript is in progress. (33) Lifshitz, I. M.; Slyozov, V. V. J. Phys. Chem. Solids 1961, 19, 35. (34) Wagner, C. Z. Elektrochem. 1961, 65, 581. (35) Seshadri, R.; Subbanna, G. N.; Vijayakrishnan, V.; Kulkarni, G. U.; Ananthakrishna, G.; Rao, C. N. R. J. Phys. Chem. 1995, 99, 5639. (36) Peng, Z. A.; Peng, X. J. Am. Chem. Soc. 2002, 124, 3343. (37) Wu, J. S.; Fu, J.-X.; Zhao, Y.-P. Microsc. Microanal. 2007, Suppl 2, 13.

JP909965T