11796
J. Phys. Chem. B 2006, 110, 11796-11799
Bending Contours in Silver Nanoprisms Benito Rodrı´guez-Gonza´ lez,* Isabel Pastoriza-Santos, and Luis M. Liz-Marza´ n* Departamento de Quı´mica Fı´sica, UniVersidade de Vigo, 36310, Vigo, Spain ReceiVed: February 24, 2006; In Final Form: May 1, 2006
A detailed electron microscopy analysis is reported on the structure of silver nanoprisms synthesized by controlled photoinduced aggregation of Ag nanoparticles. Bending of the nanoprism (111) face is directly revealed in the presence of fringes in both bright and dark field TEM images, which are unequivocally assigned to bending contours. Such contours can be individually imaged through shifting of the objective aperture and selectively allowing transmission of the corresponding diffraction spot. The assignment of a bent structure is supported by images of the lateral sides of the prisms, and high-resolution images where apparent changes in the lattice constant are observed.
Introduction The properties of anisotropic nanoparticles have raised a tremendous interest, not only because deviation from the spherical geometry leads to much larger variations in physical and chemical properties than those derived from changes in particle size (in particular for metal nanoparticles),1 but also because of the high degree of sophistication that such geometries offer from the point of view of the creation of nanostructured materials which can be useful for a large number of applications.2 Among various geometries, metal (specifically Au and Ag) nanoprisms are being intensely studied since a few years ago, when the first systematic synthetic and optical studies were published.3-16 For the actual application of such systems, it is not sufficient to understand and to be able to manipulate their optical response, but we also need to have a clear idea about their structure and mechanics. In the case of metal nanoprisms, both electron microscopy and electron diffraction studies have been carried out, which demonstrate the presence of [111] zone axis perpendicular to atomically flat surfaces of single crystals. Another typical feature of transmission electron microscopy images of such gold and silver nanoprisms is the appearance of fringes with different contrast, i.e., dark and bright lines and bands, which can even result in star-shaped features. Such fringes might in principle stem from various effects. Takayanagi17 showed the presence of dislocation networks in triangular-shaped gold islands, which develop on the (111) face because the surface layer presents lattice parameters slightly different to those of the bulk material. The restructuring of the crystalline lattice to accommodate the difference between the lattice parameters of the surface layer and those of the inner crystalline structure leads to formation of such dislocations and visualization of the stress field in the lattice which develops toward the interior. During TEM observation, these dislocation networks would be seen as band-shaped contrast variations. An alternative source of contrast variation in nanoparticles are the so-called “bending contours”, which have been reported for several types of nanoparticles and nanowires,18 as well as in silver halide microcrystals.19 On the other hand, authors reporting the electron diffraction analysis of gold and silver nanoprisms8,10,20,21 have consistently found 1/3{422} spots, forbidden for a pure fcc structure (see * Corresponding authors. E-mail:
[email protected];
[email protected].
Supporting Information, Figure S1). The presence of such forbidden spots has been assigned either to the presence of twin planes,8,20,21 to the presence of stacking faults lying parallel to the (111) surfaces and extending across the entire nanoprisms,22 or to a hexagonal-like monolayer on the nanoprism faces, which would slightly distort the (111) planes of the pure fcc cubic structure.10 In this paper we show a detailed study focusing on the fringes consistently observed in TEM images of silver nanoprisms, and we demonstrate that they result from bending contours. This observation is additionally supported by the eventual direct observation of bent nanoprisms in both SEM and TEM images of nanoprism stacks, as well as in HRTEM analysis of the crystalline structure. We propose that stacking faults or surface oxidation of the nanoprisms can play an important role in the formation of such bending. Experimental Section Ag nanoprisms were synthesized through a modification of the photoinduced method developed by Callegari et al.,11 as has been recently described.16 In short, a dispersion of poly(vinylpyrrolidone) (PVP, MW 10 000) protected, 10 nm Ag nanoparticles was illuminated with low intensity light emitting diodes (LEDs, average emission wavelength 641 nm) for 15 days, resulting in slow formation of nanoprisms with average edge length 242 nm and average thickness 10.5 nm. Transmission electron microscopy (TEM and HRTEM) experiments were carried out using a JEOL JEM 2010F fieldemission gun microscope operating at an acceleration voltage of 200 kV. Samples for TEM were prepared by depositing a drop of the nanoprism dispersion on a copper grid coated with Formvar and a carbon film. The nanoprisms deposit preferentially with their triangular faces parallel to the support film of the grid, so that the electron beam can be focused perpendicular to the {111} face, which allows to obtain easily the electron diffraction pattern in the [111] zone axis. Eventually, stacks of nanoprisms perpendicular to the grid can be found when holey grids are used. Scanning electron microscopy (SEM) images were obtained with a JEOL JSM-6700F FEG-SEM operating at an acceleration voltage of 20 kV in backscattering electron image (YAG type detector).
10.1021/jp061195l CCC: $33.50 © 2006 American Chemical Society Published on Web 05/27/2006
Bending Contours in Silver Nanoprisms
J. Phys. Chem. B, Vol. 110, No. 24, 2006 11797
Figure 1. Example of contrast bands in a silver nanoprism, slightly defocused to enhance contrast. The arrow indicates the absence of continuation of the contrast difference toward the supporting film.
Results and Discussion Figure 1 shows a typical image of a silver nanoprism, slightly defocused to enhance contrast, in which the presence of bands and lines can be readily observed, as well as several well-defined intersection points. The bands are in turn composed of dark and bright areas, already indicating the crystallographic origin of the fringes (see below). Bright “areas” can be seen at the edges in bright field imaging mode due to electron diffraction. One such band located next to a nanoprism edge is shown at higher magnification in Figure 2b, in which one can clearly distinguish a bright, central line surrounded by two lateral, dark bands, which is a rather general feature. Possible origins of these bands with different contrast are discussed next. A first possibility for the origin of contrast fringes would be the so-called mass-thickness contrast mechanism. Differences in composition can be readily ruled out since the nanoprisms are formed by pure metallic silver. Although there might in principle exist local differences in the thickness of the nanoprisms, they are not expected to generate such a high contrast. Additionally, images obtained for perpendicular nanoprism stacks16 clearly indicate that the thickness is very uniform in the whole prism. In this mass-thickness contrast mechanism one could consider the presence of the stabilizing polymer (PVP) surrounding the particle, but in Figure 2b we can clearly observe that the end of the fringes is located at the border between the metal and PVP. Another possible mechanism could be related to changes in crystalline orientation or even in crystalline structure within the nanoprisms, but electron diffraction clearly points toward single crystals, with orientation within the [111] zone axis (Figure S1). Finally, one could imagine the possibility of some sort of defect within the nanoprisms. However, if we carefully compare the type of contrast observed in our silver nanoprisms with that reported by Takayanagi,17 noticeable differences can be observed, which again seem to rule this mechanism out. Bending contours23 are often observed in TEM images of solid surfaces and offer a more plausible justification for the contrast fringes observed in the TEM images of Ag nanoprisms. Bending contours can be observed both in bright field and in dark field TEM. They stem from slight variations in the angle
Figure 2. (a) Scheme showing the origin of bending contours in bright field TEM. At point A, the incident bean is exactly parallel to the planes, while at point C the Bragg condition is fulfilled, with corresponding bright bands appearing at points B and dark bands at points D. (b) TEM image of a bending contour located at the edge of a silver nanoprism. The labels in (b) correspond to the labels of contrast areas in the diagram above. The lower area at the edge is the stabilizing PVP layer.
formed between atomic planes of the same particular plane (hkl). When atomic planes of the same family form a perfect crystalline structure, they are parallel to each other, but when the crystalline structure is bent, a small angle can arise between atomic planes of the same family at the bent area of the crystal, as schematically depicted in Figure 2a. When the sample is illuminated with an electron beam, the crystalline planes form a certain angle with respect to the direction of the incident beam, but when bending arises in the crystal, a range of angles exists between the beam and the local atomic plane within the bent area. In Figure 2a, if the beam is parallel to the plane at point A, the diffraction condition is not relevant, and in a bright field TEM image a bright contrast will be seen at point B (dark contrast in dark field TEM). If we assume that at point C the incident electron beam forms an angle with the local atomic plane such that Bragg condition is fulfilled, diffraction of the incident beam would occur and the corresponding bright field TEM image (point D) would show a dark contrast (and correspondingly bright in dark field). The comparison with the experimental image shown in Figure 2b shows a very reasonable similarity, suggesting that the fringes observed in Ag nanoprisms indeed arise from bending contours. Assuming that the previous reasoning is correct and the fringes stem from bending contours, a detailed analysis can be carried out using dark field TEM. Dark field images of the bending contours show a number of brighter areas, which correspond to those areas where Bragg’s law is exactly fulfilled.
11798 J. Phys. Chem. B, Vol. 110, No. 24, 2006
Rodrı´guez-Gonza´lez et al.
Figure 4. Scanning (a) and transmission (b) electron micrographs of nanoprism stacks. Bending is clearly observed, in particular in the prisms indicated by the arrows.
Figure 3. Dark field analysis of bending contours. (A) Bright field image of the selected nanoprism. (B) Corresponding [111] zone axis SAED diffraction pattern. (C-H) Dark field images obtained by shifting the objective aperture. The diffraction vectors of the corresponding spots in B are indicated.
Thus, when a single bending contour is present, dark field would show a single bright line. However, when all possible bendings are considered, the situation becomes more complicated. For a three-dimensional bending, where various directions and both concave and convex surfaces are involved, star-like fringes are observed in bright field TEM, with well-defined crossing points, as can be seen in the example shown in Figure 1. The central point of such star-like fringes corresponds to a zone axis of the crystal. Dark field observation allows us to obtain several different images,23 one for each of the diffracted beams, Ghkl, so that the dark field image would show the bright fringes of the bending contours arising from the atomic planes (hkl) that generate that particular diffraction spot. We need to take into account that the bending contours obtained in the dark field image from a particular diffraction spot, Ghkl, do not necessarily appear parallel to such (hkl) planes, since the curvature will be in general nonuniform and thus will not follow a direction perpendicular to the diffraction vector ghkl. In practice, we first acquire a bright field image of the nanoprism (see for example
Figure 3A), then a selected area electron diffraction (SAED) pattern (Figure 3B), which confirms that the nanoprism is oriented in the [111] zone axis, or close to it. Once the specimen is properly oriented, a dark field image is acquired from each of the diffracted beams, Ghkl (Figure 3C-H), by simply shifting the objective aperture and selectively permitting that the corresponding spot is transmitted through the objective aperture. As compared to the alternative approach based on tilting the electron beam, dark field imaging has the advantage that the image itself is not shifted on the screen. A representative example of such an analysis is shown in Figure 3. In this particular case the contours are thin (see Figure S2 for broader contours) and form a star-like pattern with a crossing point that could be assigned to a zone axis, thus indicating that this particular nanoprism possesses a net curvature (either concave or convex) around that point. The splitting of the individual bending contours is very clean and well defined, as a result of the sharp bending that makes Bragg’s law to be fulfilled in a thin area of the crystal. In other cases (see another example in the Supporting Information, Figure S2), the dark field images obtained using one or more diffraction vectors yield areas with thicker fringes, arising from Bragg’s law fulfillment in larger areas and thus pointing toward less sharp bending. We should however indicate that the curvature is never likely to be very pronounced, since the nanoprisms are always found to be within the zone axis. Further evidence in favor of a bent nanoprism structure is provided in Figures 4 and 5. In Figure 4, scanning and transmission electron micrographs of nanoprism stacks are shown, where bending to various extent can be clearly observed (see arrows). On the other hand, Figure 5 shows a HRTEM image obtained from an area close to the edge of a nanoprism in the [111] zone axis, together with the corresponding Fourier transform and the lattice spacing calculated from each diffraction spot. The first observation in this image is the presence of interfringe distances with values equal or very close to 0.25 nm, usually assigned to forbidden 1/3{422} reflections, and justified as 3×{422} lattice spacing of the fcc silver crystal.22 The analysis of the angles formed between six crystalline orientations, compared with those expected for a pure fcc lattice in the [111] zone axis, reveals deviations that can be as high as 4.3°, clearly indicating the structure is deformed, as corresponds to the bent morphology discussed above. However, again we should stress that the local bending near the edge is not expected to be drastic, and for this reason the atomic columns can still be seen with high contrast. Although the origin of the bending is not completely clear at this time, and it might be simply due to the presence of stacking faults in the structure, it seems that it may be due to slow oxidation occurring after the nanoprisms have been deposited
Bending Contours in Silver Nanoprisms
J. Phys. Chem. B, Vol. 110, No. 24, 2006 11799 are visualized as rather sharp changes of contrast in both bright and dark field TEM images, indicating slight deviations from the planar geometry. Although this remains to be proven, the origin of the bending may be due to stress in the crystalline lattice of the nanoprisms which can be linked either to the presence of a hexagonal monolayer on the (111) surfaces or to the presence of stacking faults in the crystal structure. A detailed study of the mechanical stability and ultimate degradation of the nanoprisms as a consequence of such stresses is underway and will be reported elsewhere. Acknowledgment. This work has been supported by the Spanish Ministerio de Educacio´n y Ciencia/ FEDER (Project MAT2004-02991). The authors are grateful to Prof. C. J. Kiely, Prof. M. Giersig, Prof. J. M. Gonza´lez-Calbet, and Dr. V. F. Puntes for useful comments. Supporting Information Available: Figure S1, SAED pattern from a silver nanoprism, and Figure S2, dark field analysis of bending contours. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes
Figure 5. (a) High-resolution transmission electron micrograph showing a selected area of a nanoprism and indicating the angles between six equivalent crystalline directions within the silver (111) face. (b) Fourier transform of the same image and calculated d spacing, attributed to 1/3 {422} forbidden reflections.
on the TEM grids and the solvent has been evaporated. Although the particles are stable in colloidal form for extended periods of time (as confirmed by measuring the UV-vis spectra over time), particles deposited on TEM grids have been observed to break into smaller pieces after several weeks. A similar process has been recently reported by Yacaman and co-workers for Ag nanowires.24 Conclusions Silver nanoprisms are single crystals with (111) crystalline planes forming the wide triangular face on which they contact with the supporting film of the grid. Such nanoprisms often show two- and three-dimensional bending contours, with a degree of bending that varies from prism to prism and between different areas within single nanoprisms. The bending contours
(1) Liz-Marza´n, L. M. Langmuir 2006, 22, 23. (2) Haes, A. J.; Van Duyne, R. P. J. Am. Chem. Soc. 2002, 124, 10596. (b) Maier, S. A.; Friedman, M. D.; Barclay, P. E.; Painter, O. Appl. Phys. Lett. 2005, 86, 071103/1. (c) Nam, J.-M.; Thaxton, C. S.; Mirkin, C. A. Science 2003, 301, 1884. (3) Malikova, N.; Pastoriza-Santos, I.; Schierhorn, M.; Kotov, N. A.; Liz-Marza´n, L. M. Langmuir 2002, 18, 3694. (4) Millstone, J. E.; Park, S.; Shuford, K. L.; Qin, L.; Schatz, G. C.; Mirkin, C. A. J. Am. Chem. Soc. 2005, 127, 5312. (5) Kim, F.; Connor, S.; Song, H.; Kuykendall, T.; Yang, P. Angew. Chem., Int. Ed. 2004, 43, 3673. (6) Shankar, S. S.; Rai, A.; Ankamwar, B.; Singh, A.; Ahmad, A.; Sastry, M. Nat. Mater. 2004, 3, 482. (7) Pastoriza-Santos, I.; Liz-Marza´n, L. M. Nano Lett. 2002, 2, 903. (8) Chen, S.; Carroll, D. L. J. Phys. Chem. B 2004, 108, 5500. (9) Me´traux, G. S.; Mirkin, C. A. AdV. Mater. 2005, 17, 412. (10) Jin, R.; Cao, Y.; Mirkin, C. A.; Kelly, K. L.; Schatz, G. C.; Zheng, J. G. Science 2001, 294, 1901. (11) Callegari, A.; Tonti, D.; Chergui M. Nano Lett. 2003, 3, 1565. (12) Jin, R.; Cao, C. Y.; Hao, E.; Me´traux, G.; Schatz, G. C.; Mirkin, C. A. Nature 2003, 425, 487. (13) Maillard, M.; Huang, P.; Brus, L. E. Nano Lett. 2003, 3, 1611. (14) Sun, Y.; Mayers, B.; Xia, Y. Nano Lett. 2003, 3, 675. (15) Sun, Y.; Xia, Y. AdV. Mater. 2003, 15, 695. (16) Bastys, V.; Pastoriza-Santos, I.; Rodrı´guez-Gonza´lez, B.; Vaisnoras, R.; Liz-Marza´n, L. M., AdV. Funct. Mater. 2006, 16, 766. (17) Takayanagi, K.; Tanishiro, Y.; Yagi, K.; Kobayashi, K.; Honjo, G. Surf. Sci. 1988, 205, 637. (18) Ding, Y.; Wang, Z. L. J. Phys. Chem. B 2004, 108, 12280. (19) Oleshko, V.; Gijbels, R.; Van Daele, A.; Jacob, W.; Xu, Y., Wang, S.; Park, I.; Kang, T. Microscop. Res. Tech. 1998, 42, 108. (20) Kirkland, A. I.; Edwards, P. P.; Jefferson, D. A.; Duff, D. G. Annu. Rep. Prog. Chem. Sect. C 1990, 87, 247. (21) Lofton, C.; Sigmund, W. AdV. Funct. Mater. 2005, 15, 1197. (22) Germain, V.; Li, J.; Ingert, D.; Wang, Z. L.; Pileni, M. P. J. Phys. Chem. B 2003, 107, 8717. (23) Williams D. B.; Carter, C. B. Transmission Electron Microscopy; Plenum Press: New York, 1996; p 374. (24) Elechiguerra, J. L.; Larios-Lopez, L.; Liu, C.; Garcia-Gutierrez, D.; Camacho-Bragado, A.; Yacaman, M. J. Chem. Mater. 2005, 17, 6042.