pubs.acs.org/NanoLett
Alignment, Rotation, and Spinning of Single Plasmonic Nanoparticles and Nanowires Using Polarization Dependent Optical Forces Lianming Tong,* Vladimir D. Miljkovic´, and Mikael Ka¨ll* Department of Applied Physics, Chalmers University of Technology, 412 96 Go¨teborg, Sweden ABSTRACT We demonstrate optical alignment and rotation of individual plasmonic nanostructures with lengths from tens of nanometers to several micrometers using a single beam of linearly polarized near-infrared laser light. Silver nanorods and dimers of gold nanoparticles align parallel to the laser polarization because of the high long-axis dipole polarizability. Silver nanowires, in contrast, spontaneously turn perpendicular to the incident polarization and predominantly attach at the wire ends, in agreement with electrodynamics simulations. Wires, rods, and dimers all rotate if the incident polarization is turned. In the case of nanowires, we demonstrate spinning at an angular frequency of ∼1 Hz due to transfer of spin angular momentum from circularly polarized light. KEYWORDS Optical manipulation, nanoparticles, surface plasmons, laser tweezers
T
example, been used to construct specific nanoarchitectures of GaN nanowires1 and to drive micropumps in microfluidics devices,19 but experimental evidence for optical rotation of nanoplasmonic objects has remained scarce. On the basis of ensemble studies, Pelton et al.20 and Oddershede and coworkers8 argued that gold nanorods align in polarized optical tweezers, and the rotation of 2-3 µm large aggregates of gold nanorods in an optical trap was demonstrated very recently by Jones et al.21 However, there has been no firm evidence for the possibility to align and rotate individual nanoscale plasmonic structures by optical torques on demand. In this contribution, we utilize the spectroscopic properties of elongated plasmonic structures to experimentally verify that this is indeed possible. Moreover, we investigate silver nanowires and find that they spin when subject to rotating linear polarization or a circularly polarized laser beam. These observations introduce a new modality of optical control of nanoplasmonic structures and devices.
he past decade has seen an increasing interest in optical manipulation of micro- and nanoscale structures, including semiconductor nanowires, carbon nanotubes, and biological objects.1-3 Most researchers have utilized a single beam of strongly focused laser light, socalled optical tweezers, in which field gradients push the object toward the laser focus. If the gradient force overcomes the Brownian motion, the object may be moved around at will for a variety of applications, such as sorting of cells and investigations of the mechanical properties of biomacromolecules.4-6 Although optical tweezers are usually applied to dielectric objects, it is now known that it is possible to also trap metallic objects, given that they are of sufficiently small dimensions to reduce the radiation pressure.7,8 Noble metal nanostructures are of special importance in this respect, because of the strong light-matter interaction originating in surface plasmon resonances.9,10 It has, for example, been shown that plasmonic silver nanoparticles can be optically aggregated in a laser focus, resulting in large Ramanenhancementduetoplasmonicnear-fieldcoupling,11,12 and nanostructured plasmonic surfaces has been shown to enable optical trapping with precision beyond the diffraction limit.13,14 A special but important class of optical manipulation is to utilize optical torques to rotate and align a trapped object. This can be achieved in many ways,15 for example, by an elongated laser focus through the gradient force, by a circular but polarized laser focus acting on an anisotropic object,16,17 or by direct transfer of photon spin and/or orbital angular momentum to the trapped object.18 Optical torques has, for
METHODS Laser tweezers operating at 830 nm were constructed around a Nikon TE2000E inverted microscope equipped with a Ti:sapphire laser (Spectra-Physics, 3900S), as shown schematically in Figure 1. The NIR laser beam was focused by a 60× objective (Nikon Plan Fluo, NA ) 0.7) to a diffraction limited spot on the sample, where the power used ranged between 50 and 100 mW. This is corresponding to a power density in the focal spot of the order 15-30 MW/ cm2. The laser polarization was controlled by half-wave and quarter-wave plates, while a polarizer positioned above the dark-field (DF) condenser (Nikon, NA ) 0.80-0.95) was used to create linearly polarized white light illumination for spectroscopic interrogation of the optically trapped objects. The forward scattered white light was collected by the microscope objective and split into two paths, 80% was fiber
* To whom correspondence should be addressed. E-mail: (L.T.) lianming.tong@ chalmers.se; (M.K.)
[email protected]. Received for review: 10/15/2009 Published on Web: 12/23/2009 © 2010 American Chemical Society
268
DOI: 10.1021/nl9034434 | Nano Lett. 2010, 10, 268-273
FIGURE 1. Scheme of the experimental setup. DBS, dichroic beam splitter; BS, beam splitter.
coupled to a spectrometer (Andor, SR-303I-B) via the front port of the microscope, where the 830 nm laser line was filtered out by a notch filter (NF830), while 20% was reflected to a digital camera (Nikon Coolpix 4500) for DF imaging. Elongated silver nanoparticles, hereafter referred to as nanorods, were selected by visual inspection from a colloid synthesized using the Lee and Meisel method.22 This protocol yields nanocrystalline particles with an average diameter of ∼40 nm. Although most particles are isotropic, DF microscopy revealed a significant fraction of particles with clear polarization anisotropy. Although we cannot determine the particular shape of each selected particle, previous electron microscopy investigations of similar colloids have found a range of nanorods with varying aspect ratios (see ref 23 for examples). We also utilized gold nanoparticles bought from BBInternational. These particles are nearly spherical and have an average diameter of 80 nm. Finally, we investigated single crystalline Ag nanowires synthesized according to the protocol reported in ref 24. The wires were ∼100 nm in diameter and typically several micrometers in length. A typical sample for trapping studies consists of a droplet (∼3 µL) of metal colloid placed between two glass slides separated by a 100 µm spacer. The colloids were diluted to low enough concentrations to avoid accidental trapping of multiple objects. In all the trapping experiments, the laser focus was positioned such that the trapped object is pressed against the upper glass slide by the dissipative optical force. Hence, the movement in the z-direction is hindered and trapping only occurs in two dimensions, similar to our earlier investigations of trapped plasmonic nanoparticles.10-12
FIGURE 2. Dark-field (DF) images and scattering spectra demonstrating alignment of elongated nanostructures in an optical trap. Note that the spectra are red-shifted when the white light polarization (white arrows) is turned parallel to the laser tweezers polarization (red arrows). (a) Alignment of two different silver nanorods. (b) Alignment of a dimer composed of two ∼80 nm gold nanoparticles. The upper graphs are spectra for the immobilized particle (left) and the optically trapped particle (right) before dimerization. The lower spectra correspond to the case when the two particles have been brought into near-field coupling range.
surface plasmon resonance (LSPR) mode of the nanorods. When the white light polarization is turned parallel to the laser polarization, the spots change color to green and the scattering peaks at ∼500 nm. This corresponds to excitation of a localized plasmon polarized parallel to the long axis of the rods. These results then imply that the trapped nanorods align parallel to the laser polarization. Figure 2b shows another example of alignment of an anisotropic nanoplasmonic structure, in this case a dimer composed of two isotropic gold nanoparticles with diameters of the order 80 nm. The dimerization is performed in a similar way as in our earlier investigations of optical aggregation of plasmonic nanoparticles for surface-enhanced Raman scattering (SERS),11,25 that is, one first identifies a single particle that has immobilized on the cover glass then uses the optical tweezers to capture a second diffusing particle, which is brought into the near-field zone of the immobilized particle. As can be seen from the upper two spectra in Figure 2b, the isolated particles do not exhibit any marked polarization dependence, that is, the LSPR occurs at ∼573 nm for the two orthogonal white light polarizations. However, a pronounced polarization anisotropy appears after the two particles have dimerized. This result is completely analogous to the nanorod case, that is, one measures a strongly red-shifted spectrum if the white light polarization is turned parallel to the laser tweezers polarization. However, the “long axis” mode at ∼588 nm now corresponds to the in-phase coupled dipole mode of the dimer polarized parallel
RESULTS AND DISCUSSION Figure 2a shows two examples of trapping and alignment of silver nanorods. The red and white arrows in the figure represent the laser polarization and the white light polarization, respectively. With the white light polarization perpendicular to the laser polarization, a bluish spot appears in both cases. The corresponding DF scattering spectra exhibit peaks at ∼460 nm, which we interpret as the transverse localized © 2010 American Chemical Society
269
DOI: 10.1021/nl9034434 | Nano Lett. 2010, 10, 268-273
FIGURE 3. Real and imaginary parts of the polarizability of an elongated silver nanoparticle, modeled as a prolate spheroid with diameter 60 nm and length 80 nm, in water. The red and blue curves correspond to the polarizability components probed by light polarized parallel and perpendicular to the long axis of the spheroid, respectively. The calculation utilizes a size-corrected quasi-static model, the so-called modified long-wavelength approximation (MLWA),27 and the dielectric function of Ag from ref 28. FIGURE 4. Rotation of elongated plasmonic nanostructures. (a) Rotation of a silver nanorod through the gradual rotation of the NIR laser tweezers polarization (red arrow) with respect to the fixed white light polarization (white arrow). (b) Rotation of a dimer composed of two isotropic gold nanoparticles. The spectral changes imply that the dimer axis orients parallel to the laser tweezers polarization irrespective of the polarization of the white light.
to the dimer axis. The corresponding mode polarized perpendicular to the dimer axis is almost degenerate with the single particle LSPR, in agreement with, for example, results for nanofabricated particle dimers.26 The alignment of an elongated nanoparticle in the laser focus is easy to understand from a point dipole picture. The incident electric field E induces a dipole moment P ) b RE in the particle, where b R is the complex particle polarizability tensor, which we can assume to be diagonal. Analogous to the case of a static dipole in an applied field, the induced dipole will acquire a potential energy due to the incident optical wave according to U ) -〈P·E〉 ) -1/2∑ Re(Rii)E2i , i ) x,y, z. A rod or dimer is characterized by two polarizability components, R| and R⊥, parallel and perpendicular, respectively, to the long axis of the structure. The calculation displayed in Figure 3 illustrates this for the case of an elongated silver nanoparticle approximated as a prolate spheroid with LSPR positions similar to the nanorods probed in the experiments. It is apparent that the real part of the polarizability at 830 nm, that is, at the trapping wavelength, is significantly larger for polarization parallel than perpendicular to the long axis. Hence, in a plane wave, the potential energy well experienced by the spheroid will be a factor Re[R|]/Re[R⊥] deeper if the particle is oriented with its long axis parallel rather than perpendicular to the polarization of the incident field. The particle thus tends to align along the laser polarization to minimize the optical potential energy. The experimental results above suggest that an elongated plasmonic structure will rotate if the incident polarization is rotated. That this is actually the case is shown in Figure 4 for the case of a silver nanorod and a gold particle dimer. In the former case, we gradually vary the angle θ between the laser polarization and the white light polarization, which is fixed. When the angle is decreased, by rotating the polarization of the trapping beam, the scattering peak at ∼520 nm gradually increases in intensity while a green spot appears in the DF images. This implies that the long axis of the nanorod gradually becomes colinear with the white light polarization. Similarly, the Au particle dimer in Figure 4b can © 2010 American Chemical Society
be rotated at will so that its axis becomes either parallel or perpendicular to the white light polarization. These results can be understood from the potential energy arguments above or, equivalently, from the torque experienced by an induced dipole in an applied field. For an elongated nanostructure oriented with its long axis at an angle β with respect to the laser polarization, one have τ ) |〈P × E〉| ) -1/4(R| - R⊥)sin 2β)E2.29 Hence, the torque will drive the particle to align with its long axis parallel to the laser polarization, that is, rotate the particle if the polarization is changed (see Supporting Information Movie 1). The estimated maximum amplitude of the optical torque |τ| ) 1/4|(R| - R⊥)|·E2 for the nanorod schemed in Figure 3 is of the order ∼0.4 nN·nm for 50 mW irradiation at a wavelength of 830 nm. This is in accordance with the theoretical calculations reported in ref 17 and significantly higher than the drag coefficient of the same nanorod in water. We may also note that the adhesion between the nanorod and the glass substrate is expected to be low since the glass is piranhacleaned and thus hydrophilic. In the previous experiments, nanostructures much smaller than the beam waist were confined to the focal spot by the gradient force. However, we found that silver nanowires of lengths much larger than the beam waist could also be trapped and rotated. The DF images in Figure 5a show the trapping and rotation of a silver nanowire with a length of the order ∼5 µm. Surprisingly, we found that the nanowires align perpendicular to the laser polarization instead of parallel, as the nanorods and dimers do. What’s more interesting is that the nanowires could only be trapped at their ends (see Supporting Information Movie 2) or, with a lower probability, at their midpoints (see Supporting Information Movie 3). Thus, when the laser focus is moved close 270
DOI: 10.1021/nl9034434 | Nano Lett. 2010, 10, 268-273
FIGURE 5. (a) Alignment and rotation of a silver nanowire. The DF images show a wire that has been grabbed by one of its ends by the laser tweezers and then rotated by turning the laser polarization (red arrows). (b) Optical near-fields around nanorods and nanowires. The figures show DDA simulations of the intensity enhancement factor around a nanorod (D ) 40 nm, L ) 80 nm) and two nanowires (D ) 100 nm, L1 ) 2 µm, L2 ) 5 µm) under linearly polarized plane wave irradiation (830 nm) and for two orthogonal laser polarizations (red arrows).
to any other position of a nanowire, it is trapped but then immediately slides along the long axis so that the trapping site moves to the closest end or to the midpoint, while it simultaneously aligns perpendicular to the laser polarization. We may also note that trapped wires sometimes oriented themselves along the optical axis, that is, perpendicular to the glass surface (see Supporting Information Movie 4). This orientation is typical for elongated objects trapped in three dimensions because it minimizes the scattering force acting on the structure. To understand the different trapping characteristics, we performed electrodynamics simulations of the optical nearfields generated by nanoplasmonic structures. We used the discrete dipole approximation (DDA) according to refs 30 and 31 and regular mesh sizes between 1 and 5 nm. Figure 5b shows the spatial variation of the intensity enhancement factor (E/E0)2 around a silver nanorod (D ) 40 nm, L ) 80 nm) and two silver nanowires (D ) 100 nm, L1 ) 2 µm, L2 ) 5 µm) under linearly polarized plane wave irradiation and for the same wavelength as the trapping laser (830 nm). For the nanorod, we see that an incident polarization parallel to the long axis generates a much stronger near-field than for perpendicular polarization. This is as expected from the higher parallel than perpendicular polarizability seen in © 2010 American Chemical Society
Figure 3 and is consistent with an optical alignment of the nanorod parallel to the polarization of the trapping beam. The two nanowires, on the other hand, exhibit field patterns that are very different from the nanorod case. In particular, the highest local fields are generated for polarization perpendicular to the wire axis and at points very close to the nanowires ends. This is thus fully consistent with the majority of experimental observations. However, the fact that some wires are trapped at their midpoints is difficult to understand. Multipolar/standing-wave plasmon modes, which can be seen as periodic intensity modulations in Figure 5b but are more efficiently excited by fields incident with a large angle toward the nanowire axis,32-34 could play a role, but it is not obvious how such modes could generate a stable midpoint equilibrium. Clearly, this issue requires further study, in particular simulations that goes beyond the plane wave approximation and takes the finite width and divergence of the trapping beam into account. Finally, we investigated whether it was possible to actually spin the nanowires by rapidly rotating the incident linear polarization or by applying a direct photon induced torque through circularly polarized light. Indeed, both methods resulted in spinning at frequencies of the order of one up to a few hertz, as summarized in Figure 6. In order to track 271
DOI: 10.1021/nl9034434 | Nano Lett. 2010, 10, 268-273
In the example above, we induced spinning through the mechanical rotation of an external device, the half-wave plate. However, Figure 6b shows that the same effect can be achieved by simply utilizing the torque induced by circularly polarized light. The origin of this effect, that is, the transfer of angular momentum to an absorbing body,35 is thus conceptually different from the “polarization” forces utilized in the previous examples of alignment, rotation, and spinning of elongated plasmonic structures. Specifically, the torque results from the spin of the individual photons (each absorbed photon transfers an angular momentum of p), but the beam itself does not carry angular momentum (in contrast to beams with helical phase fronts15,36). We also observed that nanowires were trapped and rotated either at one of the ends or at the midpoint (see Supporting Information Movie 5 for an example of midpoint trapping). As shown in the lower inset of Figure 6b, the spinning frequency shows the expected linear scaling with incident laser power. However, the maximum frequency that we reach for a ∼5 µm long nanowire with the present laser power (∼100 mW) is somewhat lower than when the trap polarization is rotated. We should also point out that it was straightforward to vary the spinning frequency and the sense of rotation by varying the ellipticity of the beam and by changing from right- to left-circularly polarized light, respectively (also see Supporting Information Movie 5).
FIGURE 6. Spinning of trapped nanowires using rotating linear polarization and circularly polarized light. (a) Spinning by rotating linear polarization. Upper figures illustrate the rotation detection scheme, in which the scattering detection spot (blue circle) is displaced 1-2 µm downward from the laser tweezers focus (red arrow signifying polarization). When the nanowire passes the detection spot, a large scattering signal is observed (red/yellow regions in DF spectral readouts vs time). The trapping laser gives rise to a polarization dependent background (e.g., peak at 830 nm) that allows one to estimate the angle between the nanowire and the laser polarization. (b) Spectral readout of nanowire trapped at the end and rotated by circularly polarized light. Note that the background from the trapping laser is constant in this case. The inset to the right shows the variation in spinning frequency with applied laser power.
SUMMARY AND OUTLOOK Using spectroscopic characterization based on elastic scattering measurements, we have experimentally demonstrated that it is possible to align and rotate elongated plasmonic Ag nanoparticles, dimers of spherical Au particles, as well as micrometer-sized Ag nanowires in two dimensions using linearly polarized NIR laser beam. We have also shown that it is possible to directly spin plasmonic nanowires by transfer of photon spin angular momentum. The orientations of the trapped objects relative to the incident polarization exhibit interesting differences that we can link to differences in plasmonic spectral properties, but we anticipate that a plethora of other wavelength dependent trapping and rotation effects await exploration. The possibility to align and rotate nanoplasmonic objects may also have practical implications. For example, the ability to grab silver nanowires by their ends and orient them by polarization may be utilized for construction of various nanophotonic architectures, in particular if it turns out to be possible to optically weld wires together, as has been shown for semiconducting wires.1 New functionalities might also be realized by combining the wellknown biosensing and spectroscopic finger-printing abilities of plasmonic nanoparticles37,38 with the possibility to probe hydrodynamic properties and particle attachment to biological surfaces via optical rotation.
the rotation of a nanowire, we spatially separated the position of the tweezers focus and the point from which we collect DF spectra by a few micrometers (see upper inset in Figure 6a). Thus, for each turn of the nanowire, the scattered signal from the wire peaks only once if the wire is attached at one of its ends and twice if it is attached at the midpoint. A polarization dependent background from the trapping laser (830 nm) allows us to simultaneously monitor the polarization direction of the trapping beam. Figure 6a shows a series of scattering spectra (50 ms exposure time, 126 ms cycle time) from a nanowire that has been trapped at one end and is being rotated at ∼0.8 Hz angular frequency. The wire escapes after ∼9 s, but before this one can estimate the angle between the wire and the incident polarization to approximately 30 degrees. For a given torque, this angle, or phase shift, is expected to sensitively depend on the viscosity of the medium, the friction against the glass surface and the morphology of the nanowire.16,29 We observe phase slipping (i.e., the wire “misses” a turn) and eventual chaotic motion when the rotation frequency of the polarizer is gradually increased (see Supporting Information Movie 2 for an example). © 2010 American Chemical Society
Acknowledgment. This work was financially supported by the Swedish Research Council and the Swedish Founda272
DOI: 10.1021/nl9034434 | Nano Lett. 2010, 10, 268-273
tion for Strategic Research. The authors thank Yingzhou Huang and Prof. Hongxing Xu for the silver nanowire samples and Professors Peter Johansson, Peter Apell, and Jari Kinaret for stimulating discussions.
(16) Bonin, K. D.; Kourmanov, B.; Walker, T. G. Opt. Express 2002, 10, 984–989. (17) de Abajo, F. J. G. J. Quant. Spectrosc. Radiat. Transf. 2004, 89, 3– 9. (18) Friese, M. E. J.; Nieminen, T. A.; Heckenberg, N. R.; RubinszteinDunlop, H. Nature 1998, 394, 348–350. (19) Maruo, S.; Inoue, H. Appl. Phys. Lett. 2006, 89, 144101. (20) Pelton, M.; Liu, M. Z.; Kim, H. Y.; Smith, G.; Guyot-Sionnest, P.; Scherer, N. E. Opt. Lett. 2006, 31, 2075–2077. (21) Jones, P. H.; Palmisano, F.; Bonaccorso, F.; Gucciardi, P. G.; Calogero, G.; Ferrari, A. C.; Marag, O. M. Acs Nano 2009, 3, 3077– 3084. (22) Lee, P. C.; Meisel, D. J. Phys. Chem. 1982, 86, 3391–3395. (23) Xu, H. X.; Aizpurua, J.; Kall, M.; Apell, P. Phys. Rev. E 2000, 62, 4318–4324. (24) Sun, Y. G.; Xia, Y. N. Adv. Mater. 2002, 14, 833–837. (25) Li, Z. P.; Kall, M.; Xu, H. Phys. Rev. B 2008, 77, No. 085412. (26) Gunnarsson, L.; Rindzevicius, T.; Prikulis, J.; Kasemo, B.; Kall, M.; Zou, S. L.; Schatz, G. C. J. Phys. Chem. B 2005, 109, 1079– 1087. (27) Bohren, C. F.; Huffman, D. R. Absorption and Scattering of Light by Small Particles; John Wiley & Sons: New York, 1998. (28) Johnson, P. B.; Christy, R. W. Phys. Rev. B 1972, 6, 4370–4379. (29) Shelton, W. A.; Bonin, K. D.; Walker, T. G. Phys. Rev. E 2005, 71, No. 036204. (30) Draine, B. T.; Flatau, P. J. J. Opt. Soc. Am. A 1994, 11, 1491–1499. (31) Draine, B. T.; Flatau, P. J. User Guide for the Discrete Dipole Approximation Code DDSCAT 7.0.; http://arxiv.org/abs/0809.0337v4 (accessed Oct. 2008). (32) Knight, M. W.; Grady, N. K.; Bardhan, R.; Hao, F.; Nordlander, P.; Halas, N. J. Nano Lett. 2007, 7, 2346–2350. (33) Ditlbacher, H.; Hohenau, A.; Wagner, D.; Kreibig, U.; Rogers, M.; Hofer, F.; Aussenegg, F. R.; Krenn, J. R. Phys. Rev. Lett. 2005, 95, 257403. (34) Li, Z.; Hao, F.; Huang, Y.; Fang, Y.; Nordlander, P.; Xu, H. Nano Lett. 2009, 9 (12), 4383–4386. (35) Friese, M. E. J.; Nieminen, T. A.; Heckenberg, N. R.; RubinszteinDunlop, H. Opt. Lett. 1998, 23, 1–3. (36) He, H.; Friese, M. E. J.; Heckenberg, N. R.; Rubinszteindunlop, H. Phys. Rev. Lett. 1995, 75, 826–829. (37) Lal, S.; Link, S.; Halas, N. J. Nat. Photonics 2007, 1, 641–648. (38) Anker, J. N.; Hall, W. P.; Lyandres, O.; Shah, N. C.; Zhao, J.; Duyne, R. P. V. Nat. Mater. 2008, 7, 442–453.
Supporting Information Available. Movie clips showing the optical alignment, rotation, and spinning of elongated silver nanorods and nanowires. This material is available free of charge via the Internet at http://pubs.acs.org. REFERENCES AND NOTES (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15)
Pauzauskie, P. J.; Radenovic, A.; Trepagnier, E.; Shroff, H.; Yang, P. D.; Liphardt, J. Nat. Mater. 2006, 5, 97–101. Tan, S. D.; Lopez, H. A.; Cai, C. W.; Zhang, Y. G. Nano Lett. 2004, 4, 1415–1419. Nishimura, Y.; Misumi, O.; Matsunaga, S.; Higashiyama, T.; Yokota, A.; Kuroiwa, T. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 12577–12582. Ashkin, A.; Dziedzic, J. M. Science 1987, 235, 1517–1520. Grover, S. C.; Skirtach, A. G.; Gauthier, R. C.; Grover, C. P. J. Biomed. Opt. 2001, 6, 14–22. Bennink, M. L.; Scharer, O. D.; Kanaar, R.; Sakata-Sogawa, K.; Schins, J. M.; Kanger, J. S.; de Grooth, B. G.; Greve, J. Cytometry 1999, 36, 200–208. Hansen, P. M.; Bhatia, V. K.; Harrit, N.; Oddershede, L. Nano Lett. 2005, 5, 1937–1942. Selhuber-Unkel, C.; Zins, I.; Schubert, O.; Sonnichsen, C.; Oddershede, L. B. Nano Lett. 2008, 8, 2998–3003. Svoboda, K.; Block, S. M. Opt. Lett. 1994, 19, 930–932. Prikulis, J.; Svedberg, F.; Kall, M.; Enger, J.; Ramser, K.; Goksor, M.; Hanstorp, D. Nano Lett. 2004, 4, 115–118. Svedberg, F.; Li, Z. P.; Xu, H. X.; Kall, M. Nano Lett. 2006, 6, 2639– 2641. Tong, L. M.; Righini, M.; Gonzalez, M. U.; Quidant, R.; Kall, M. Lab Chip 2009, 9, 193–195. Righini, M.; Zelenina, A. S.; Girard, C.; Quidant, R. Nat. Phys. 2007, 3, 477–480. Grigorenko, A. N.; Roberts, N. W.; Dickinson, M. R.; Zhang, Y. Nat. Photonics 2008, 2, 365–370. Padgett, M.; Leach, J. In Structured Light and Its Applications; Andrews, D. L., Ed.; Elsevier Inc.: New York, 2008; pp 237-248.
© 2010 American Chemical Society
273
DOI: 10.1021/nl9034434 | Nano Lett. 2010, 10, 268-273
