Nanoscale subsurface-and material-specific identification of single nanoparticles

We report on high resolution subsurface and material specific differentiation of silica, Au and silica-capped Au nanoparticles using scattering-type scanning near-field optical microscopy (s-SNOM) in the visible (λ=633 nm) and mid-infrared (λ=10.7 μm) frequencies. Strong optical contrast is observed in the visible wavelength, mainly because of the dipolar plasmon resonance of the embedded Au nanoparticles which is absent in the infrared. We show that the use of small tapping amplitude improves the apparent image contrast in nanoparticles by causing increased tip-particle and reduced tip-substrate interactions. Experimental results are in excellent agreement with extended dipole model calculations modified to include the capping layer characterized by its refractive index. ©2011 Optical Society of America OCIS codes: (180.4243) Near-field microscopy; (160.4236) Nanomaterials; (240.6680) Surface plasmons References and links 1. S. K. Basiruddin, A. Saha, N. Pradhan, and N. R. Jana, “Advances in coating chemistry in deriving soluble functional nanoparticle,” J. Phys. Chem. C 114(25), 11009–11017 (2010). 2. I. Pastoriza-Santos, J. Perez-Juste, and L. M. Liz-Marzan, “Silica-coating and hydrophobation of CTABstabilized gold nanorods,” Chem. Mater. 18(10), 2465–2467 (2006). 3. A. Bao, H. Lai, Y. M. Yang, Z. L. Liu, C. Y. Tao, and H. Yang, “Luminescent properties of YVO4:Eu/SiO2 core–shell composite particles,” J. Nanopart. Res. 12(2), 635–643 (2010). 4. J. P. Zimmer, S. W. Kim, S. Ohnishi, E. Tanaka, J. V. Frangioni, and M. G. Bawendi, “Size series of small indium arsenide-zinc selenide core-shell nanocrystals and their application to in vivo imaging,” J. Am. Chem. Soc. 128(8), 2526–2527 (2006). 5. F. Teng, Z. J. Tian, G. X. Xiong, and Z. S. Xu, “Preparation of CdS–SiO2 core–shell particles and hollow SiO2 spheres ranging from nanometers to microns in the nonionic reverse microemulsions,” Catal. Today 93–95, 651– 657 (2004). 6. M. Yu, J. Lin, and J. Fang, “Silica Spheres Coated with YVO4:Eu 3+ Layers via sol−gel process: a simple method to obtain spherical core−shell phosphors,” Chem. Mater. 17(7), 1783–1791 (2005). 7. G. A. Lawrie, B. J. Battersby, and M. Trau, “Synthesis of optically complex core–shell colloidal suspensions: pathways to multiplexed biological screening,” Adv. Funct. Mater. 13(11), 887–896 (2003). 8. A. P. Alivisatos, “Semiconductor clusters, nanocrystals, and quantum dots,” Science 271(5251), 933–937 (1996). 9. M. De, P. S. Ghosh, and V. M. Rotello, “Applications of nanoparticles in biology,” Adv. Mater. (Deerfield Beach Fla.) 20(22), 4225–4241 (2008). 10. H. C. Dong, M. Z. Zhu, J. A. Yoon, H. F. Gao, R. C. Jin, and K. Matyjaszewski, “One-pot synthesis of robust core/shell gold nanoparticles,” J. Am. Chem. Soc. 130(39), 12852–12853 (2008). 11. F. Keilmann and R. Hillenbrand, “Near-field microscopy by elastic light scattering from a tip,” Phil. Trans. R. Soc. Lond. A 362(1817), 787–805 (2004). 12. R. Hillenbrand and F. Keilmann, “Complex optical constants on a subwavelength scale,” Phys. Rev. Lett. 85(14), 3029–3032 (2000). 13. L. Novotny, “Effective wavelength scaling for optical antennas,” Phys. Rev. Lett. 98(26), 266802 (2007). 14. Z. H. Kim and S. R. Leone, “High-resolution apertureless near-field optical imaging using gold nanosphere probes,” J. Phys. Chem. B 110(40), 19804–19809 (2006). 15. A. Cvitkovic, N. Ocelic, and R. Hillenbrand, “Material-specific infrared recognition of single sub-10 nm particles by substrate-enhanced scattering-type near-field microscopy,” Nano Lett. 7(10), 3177–3181 (2007). 16. A. Cvitkovic, N. Ocelic, J. Aizpurua, R. Guckenberger, and R. Hillenbrand, “Infrared imaging of single nanoparticles via strong field enhancement in a scanning nanogap,” Phys. Rev. Lett. 97(6), 060801 (2006). 17. Z. H. Kim, S. H. Ahn, B. Liu, and S. R. Leone, “Nanometer-scale dielectric imaging of semiconductor nanoparticles: size-dependent dipolar coupling and contrast reversal,” Nano Lett. 7(8), 2258–2262 (2007). #149682 $15.00 USD Received 22 Jun 2011; revised 21 Aug 2011; accepted 28 Aug 2011; published 5 Oct 2011 (C) 2011 OSA 10 October 2011 / Vol. 19, No. 21 / OPTICS EXPRESS 20865 18. A. Hartschuh, E. J. Sánchez, X. S. Xie, and L. Novotny, “High-resolution near-field Raman microscopy of single-walled carbon nanotubes,” Phys. Rev. Lett. 90(9), 095503 (2003). 19. M. Brehm, T. Taubner, R. Hillenbrand, and F. Keilmann, “Infrared spectroscopic mapping of single nanoparticles and viruses at nanoscale resolution,” Nano Lett. 6(7), 1307–1310 (2006). 20. J.-S. Samson, R. Meißner, E. Bründermann, M. Böke, J. Winter, and M. Havenith, “Characterization of single diamondlike and polymerlike nanoparticles by midinfrared nanospectroscopy,” J. Appl. Phys. 105(6), 064908 (2009). 21. T. Taubner, F. Keilmann, and R. Hillenbrand, “Nanoscale-resolved subsurface imaging by scattering-type nearfield optical microscopy,” Opt. Express 13(22), 8893–8899 (2005). 22. Y. Abate, A. Schwartzberg, D. Strasser, and S. R. Leone, “Chem. “Nanometer-scale size dependent imaging of cetyl trimethyl ammonium bromide (CTAB) capped and uncapped gold nanoparticles by apertureless near-field optical microscopy,” Phys. Lett. 474, 146–152 (2009). 23. T. Nann and P. Mulvaney, “Single quantum dots in spherical silica particles,” Angew. Chem. Int. Ed. Engl. 43(40), 5393–5396 (2004). 24. S. O. Obare, N. R. Jana, and C. J. Murphy, “Preparation of polystyreneand silica-coated gold nanorods and their use as templates for the synthesis of hollow nanotubes,” Nano Lett. 1(11), 601–603 (2001). 25. S. Eustis and M. A. el-Sayed, “Why gold nanoparticles are more precious than pretty gold: noble metal surface plasmon resonance and its enhancement of the radiative and nonradiative properties of nanocrystals of different shapes,” Chem. Soc. Rev. 35(3), 209–217 (2006). 26. J. Aizpurua, T. Taubner, F. J. García de Abajo, M. Brehm, and R. Hillenbrand, “Substrate-enhanced infrared near-field spectroscopy,” Opt. Express 16(3), 1529–1545 (2008). 27. V. V. Gozhenko, L. G. Grechko, and K. W. Whites, “Electrodynamics of spatial clusters of spheres: Substrate effects,” Phys. Rev. B 68(12), 125422 (2003). 28. N. Ocelic, A. Huber, and R. Hillenbrand, “Pseudoheterodyne detection for background-free near-field spectroscopy,” Appl. Phys. Lett. 89(10), 101124 (2006). 29. R. Hillenbrand and F. Keilmann, “Material-specific mapping of metal/semiconductor/dielectric nanosystems at 10 nm resolution by backscattering near-field optical microscopy,” Appl. Phys. Lett. 80(1), 25–27 (2002). 30. G. P. Wiederrecht, G. A. Wurtz, and J. Hranisavljevic, “Coherent coupling of molecular excitons to electronic polarizations of noble metal nanoparticles,” Nano Lett. 4(11), 2121–2125 (2004). 31. N. T. Fofang, T. H. Park, O. Neumann, N. A. Mirin, P. Nordlander, and N. J. Halas, “Plexcitonic nanoparticles: plasmon-exciton coupling in nanoshell-J-aggregate complexes,” Nano Lett. 8(10), 3481–3487 (2008). 32. E. W. Palik, Handbook of Optical Constants of Solids (Academic Press, Berlin, 1985). 33. R. Hillenbrand, T. Taubner, and F. Keilmann, “Phonon-enhanced light matter interaction at the nanometre scale,” Nature 418(6894), 159–162 (2002). 34. S. Amarie, T. Ganz, and F. Keilmann, “Mid-infrared near-field spectroscopy,” Opt. Express 17(24), 21794– 21801 (2009). 35. S. Amarie and F. Keilmann, “Broadband-infrared assessment of phonon resonance in scattering-type near-field microscopy,” Phys. Rev. B 83(4), 045404 (2011).


Introduction
Over the past decade, synthesis of coated nanoparticles and core-shell nanostructures has become an active field of research [1].This is because coatings can serve to convert a hydrophobic nanoparticle into a water-soluble nanoparticle, to prevent oxidation of plasmonic photocatalysts, or to introduce chemical functionality on the nanoparticle surface to link with various chemicals and biochemicals of interest [2,3].To date, applications of coated nanoparticles have been realized in the areas of electronics [2], photonics [4], catalysis [5], nanotechnology [6], and biomedical research [7].As the ability to synthesize coated nanoparticles and complex core-shell nanostructures improves, an increased need to detect and map these materials on the nanoscale with material and spectroscopic sensitivity is recognized [8][9][10].Optical methods, in addition to being nondestructive and chemically specific, allow investigation of structure-property relationships, effects of size and geometry on light induced resonant properties of nanostructures (such as absorption and scattering), and plasmon-plasmon interactions.Classical light microscopy allows the resolution of sample features down to about half the illumination wavelength (~250 nm in the visible).This limited resolution hinders single particle characterization and spectroscopic investigation.Scatteringtype scanning near-field optical microscopy (s-SNOM) offers ultrahigh optical resolution that depends on the radius of curvature of the probe and not on the wavelength of light used [11][12][13][14].S-SNOM has been used to detect optical material contrast of single nanoparticles, spectroscopic mapping of single nanobeads and viruses, polydisperse nanoparticles, single carbon nanotubes, diamond and polymerlike nanoparticles [15][16][17][18][19][20].This technique also permits subsurface imaging of nanostructures through transparent extended surface layers [21].In spite of various optical material contrast studies of nanoparticles using s-SNOM, near-field single particle investigation of coated nanoparticles and core-shell nanostructures still remain largely unexplored.Recently, the near-field signal amplitude contrast from Au nanoparticles (AuNPs) capped with the cationic surfactant cetyl trimethyl ammonium bromide (CTAB) and uncapped AuNPs were studied using s-SNOM in the visible wavelength [22].However lack of uniformity of the CTAB thickness layer on AuNP surfaces prevented extensive investigation of near-field contrast formation in capped nanoparticles.
Herein, we employ s-SNOM to directly investigate the near-field optical contrast mechanism in coated nanoparticles by imaging three types of individual particles (silica, Au and silica-capped AuNPs) and mixed particles adsorbed on Si substrate.The widely used coating layer, namely, silica [23,24], was chosen since it protects nanoparticles from aggregation and helps to maintain a greater layer uniformity [1].Identification of capped and uncapped nanoparticles with nanoscale spatial resolution is demonstrated.An increased optical contrast between silica and silica-capped AuNPs is observed at 633 nm compared to that at 10.7 µm.The difference is mainly from the on-and off-plasmon resonances in visible and mid-infrared frequencies: AuNPs and silica-capped AuNPs have dipolar plasmon resonance near 633 nm whereas they are off-resonant in 10.7 um [25].In the visible wavelength the strong plasmon-polariton-resonant tip-substrate coupling allow to distinguish silica-capped AuNPs from bare silica particles [26].The plasmon-enhanced near-field coupling between the tip's near-field and the Au core through the silica cover layer enhances the weak near-field interaction of the silica capping layer [21].This additional signal strength, offered by the coupling to the Au core, results in a brighter contrast (larger amplitude) of the capped particle at lower tapping amplitude of the probing tip compared to a bare silica nanoparticle.As the tapping amplitude surpasses the particle diameter, there is an increased interaction with the silicon substrate on which the particles are adsorbed.Hence, the difference in near-field amplitude contrast between the capped particles and the uncapped particles is reduced.To support our interpretation, we also performed an extended dipole model calculation that includes the refractive index of the capping layer [27].The unique combined high resolution material specific imaging 15 and subsurface imaging [21] capabilities of s-SNOM provide a nondestructive, highly detailed method for mapping and chemical characterization of coated nanoparticles and complex core-shell nanostructures.

Experimental Method
Near-field scattering measurements in the visible (He-Ne laser at 633 nm) and infrared (CO 2 laser at 10.7 µm) frequencies are performed using a commercial s-SNOM setup (NeaSNOM, neaspec.com)shown in Fig. 1.Development and details of s-SNOM are well reviewed in various references [11,12,28,29].Briefly, s-SNOM is based on a tapping mode atomic force microscope (AFM) operated in the intermittent contact mode.S-SNOM is capable of simultaneous topography, amplitude and phase contrast imaging by recording scattered laser light from commercial PtIr-coated cantilevered Si tips with a vertical oscillation frequency of 240 kHz and amplitude of ~20 nm.Near-field scattering data is acquired using a combination of demodulation of the detector signal at higher harmonics of the resonance frequency, nΩ (demodulation order n>1) and a pseudoheterodyne interferometric signal detection scheme for greater background suppression [28].Nanoparticle samples were prepared on a Si substrate that was first cleaned by sonication in methanol and dried with nitrogen gas.The substrate was then dipped in a 1% APTMS (3aminopropyltrimethoxysilane) solution in ethanol.Nanoparticle solutions (nanoComposix Inc., nanocomposix.com)were drop-casted on the wafer, allowed to react for 12.5 minutes.The substrate was then rinsed with ethanol and dried with nitrogen gas.The samples containing mixed particles of three kinds (silica, silica-capped Au and pure Au nanoparticles) were prepared by diluting each solution with aqueous isopropyl alcohol (IPA) solution (3:1 IPA: H 2 O) and drop-casting the solutions in the order Au, silica-capped gold, and then silica particles allowing each solution to react for 10 minutes.

Results and discussion
Figure 3 shows topography and near-field optical images of a mixture of three types of particles (silica, Au and silica-capped AuNPs) on a Si substrate at a laser wavelength of, λ=633 nm.In the topography image [Fig.3(a)] all particles look identical while the optical amplitude image [Fig.3(b)] shows that the nanoparticles exhibit three different image contrasts compared to the substrate.Particles appear either bright, dark or medium compared to the Si substrate.In Fig. 3(d) a zoomed-in image of the three contrast types captured in a 500 nm x 500 nm area is shown.Line plots of AFM heights [Fig.3(e)] and optical amplitude signals [Fig.3(f)] of two selected particles in Fig. 3(c) show that the heights of the two particles are the same ~50 nm [Fig.3(e)].However, the amplitude signals differ [Fig.3(f)], which suggests that the two particles are not the same type in the mixture composed of silica, Au and silica-capped AuNPs [15].Furthermore, in Fig. 3(d) the brightest central particle which has a normalized amplitude signal of s (s=s Au /s Si ) ~3:1, is easily identified as Au as it is well known [11,12].Near-field interaction of the probing tip is expected to be stronger with the silica-capped AuNP than with a bare silica particle due to a strong plasmon enhanced coupling of the tip near-field with the buried AuNP below the coating [21].These observations could suggest that the red line represents a silica-capped AuNP and the blue line a bare silica particle.To validate this hypothesis we performed a controlled experiment on two samples that contained only either silica nanoparticles or silica-capped AuNPs adsorbed on Si substrate.The images used for analysis on these samples were acquired using the same scanning parameters, the same scanning tip, and the same tip vibrational amplitude.We performed pixel-by-pixel analysis of the data on each sample by correlating topography and near-field signal [15].This is done by plotting the optical signal amplitude s 3 (x,y) measured at a pixel (x,y) versus the height h(x,y) extracted from the corresponding topography image on each nanoparticle.The result of the analysis is plotted in Fig. 4 which shows the bare silica containing sample in blue and the capped sample in red.For capped particles (red points), amplitude signal contrast values recorded at pixels near the center of the particles, which correspond to larger height values, are brighter than those farther from the center.These results support our hypothesis that the signal contrast on capped particles are brighter than those on bare silica particles, thus allowing identification of particles based on their material contrast.A further confirmation of our hypothesis comes from a theoretical calculation using an extended dipole-dipole coupling model.The model is based on the solution of the electrostatic boundary-value problem (Laplace's equation) for a system of many interacting dipoles in the presence of a substrate [27].In this model, the extended structure of the probing tip is approximated by a point dipole located at the extreme end of the tip.The orientation of this dipole is along the direction of the external excitation laser which is assumed to be along the axis of the tip (perpendicular to the sample surface, z-axis in Fig. 1).The nanoparticles are also considered as point dipoles.The dipole orientation of the nanoparticles is also taken to be along the z-axis, similar to the tip dipole orientation [27].Both the tip dipole and the nanoparticle dipole interact with their own image dipoles generated by the sample surface.The effective polarizability, described by the sum of the polarizability of the tip dipoles and the nanoparticle dipoles, allows determination of both the amplitude s and the phase φ of the near-field interaction [12].The model is further improved by taking into account signal harmonic demodulation in which the 3rd harmonic signal is numerically calculated by Fourier transforming the modulated effective polarizability [14].
The dielectric function (ε c ) of the hybrid core-shell nanoparticle (silica-coated Au) is included in the simulation as [30,31] ( ) ( ) where ε 1 and r 1 are the dielectric constant and radius of the metal core respectively, ε 2 and r 2 are the dielectric constant and radius of the silica shell, respectively, and ε m is the dielectric constant of the surrounding medium taken to be air.For visible wavelength experiments (at 633 nm) the dielectric constants for AuNPs, silica particles/shell and the Si substrate are taken to be ε 1 =−11.8+ 1.25i and ε 2 =2 and ε Si = 15.1 + 0.16i, respectively [32].To simulate experimental data, we assumed a constant thickness of the capping layer and varied the diameter of the core AuNP.Therefore each point in the calculation represents a spherical AuNP covered uniformly by constant thickness of silica shell.
Figure 5 shows the result of the calculation for the bare silica (solid line) and the silicacapped AuNP (broken line).Superimposed on the theoretical graphs are experimental data points acquired on the two samples: bare silica (blue data points) and capped (red data points).Each experimental data point plotted in Fig. 5 represents an average signal value of pixels taken near the center of a particle.The theoretical calculation agrees excellently with the experimental average data points.In the calculation the only adjustable free parameter is the thickness of the capping layer.For the fit shown in Fig. 5, the thickness is taken to be 13 nm, in very good agreement to the average thickness value of ~10 nm estimated from the synthesis of the nanoparticles.As noted above the calculations assume constant thickness of the capping layer.If, on the other hand, a constant core size and a varying capping thickness are assumed in the simulation, a fit to experimental data cannot be achieved.Capped and uncapped particles are identical in the size range when the height of the capped particle is less than the total thickness of capping layer (13 nm), and result in an overlapping amplitude signal until up to 26 nm particle height.As the core AuNP increases in size, the amplitude signal of a capped silica particle becomes stronger than a bare silica nanoparticle of the same size.The broad plasmon resonance of AuNPs in the visible [25] (expressed by the negative real part of the dielectric constant of Au) [33] results in resonant tip-substrate interaction that enhances the near field contrast of the capped particles [26].The increasing signal level with size is due to the increasing plasmon-enhanced near-field coupling between tip's near-field and the AuNP core below the silica capping layer.The weak near-field interaction of the silica capping layer is therefore augmented by the strong near-field interaction of the embedded AuNP.This additional signal strength offered by the plasmonic Au core results in a brighter contrast (larger amplitude) of the capped particle compared to a bare silica nanoparticle of the same size.To study the near-field amplitude contrasts in silica-capped metal particles and bare silica particles in a spectral region where the AuNPs do not possess plasmon resonance, we perform s-SNOM imaging in the infrared (λ=10.7 µm).The same two samples used in the visible study, silica particles and silica-capped AuNPs adsorbed on silicon, were utilized.The results of analysis and calculation in the infrared (λ=10.7 µm), similar to what was carried out for visible scans is shown in Figs.6(a) and 6(b).The amplitude signal contrast of the silicacapped AuNPs (red dots) overlaps with amplitude signal contrast of bare silica particles (blue dots) with a small branching near the center (larger pixel height) of the particles.At 10.7 µm, AuNPs of size considered in this work are not expected to show plasmon resonance and cannot lead to signal enhancement based on a resonant near-field interaction.However, the embedded highly reflecting AuNP below the coating layer can enhance the near-field signal of a thin sample layer based mainly on the effect of reflection of the substrate as shown by the narrower and more overlapping amplitude contrast branching at 10.7 µm.In addition, discrimination between capped and uncapped particles in the infrared is made more difficult because of amplitude signal fluctuation as documented before for materials showing weak dielectric contrast [15].An improved image analysis will certainly minimize this fluctuation allowing better material specific discrimination in the infrared.The strong optical contrast in the visible (633 nm) region demonstrated experimentally and theoretically in Figs. 4 and 5 above, can be used to identify each nanoparticle with material specificity in a mixed sample [see Figs.7(a) and 7(b)] composed of silica, Au and silica-capped AuNPs adsorbed on Si substrate.To achieve this goal we plotted the average (using ~35 pixels per particle) near-field amplitude signal contrast at the center of a particle as a function of the height of the nanoparticles.A clear separation of the average amplitude data points between capped particles (red data points) and uncapped particles (blue points) is shown in Fig. 7(c) which is qualitatively similar to Fig. 5.These particles are marked by a colored circled (red (capped) or blue (uncapped)) on the topography image [Fig.7(d)] depending on whether they belong to the red or blue points on Fig. 7(c) thereby allowing us to identify particles as capped or uncapped.To explore the role of the tapping amplitude of the probing tip on the near-field contrast of nanoparticles we carried out experimental study and showed the result in Fig. 8.We used the third harmonics signal normalized to the signal value on Si substrate at a laser wavelength, λ=633 nm.The normalized amplitude signal increases monotonically, until approximately the tapping amplitude is comparable to the particle diameter (~75 nm), and then approaches saturation.A similar saturation has been reported on extended samples but explanations differ [34].The saturation observed at higher tapping on extended samples is because the near-field extent is a fixed length, equal to about the tip diameter.Therefore, much larger tapping amplitude will not increase the signal further and leads to saturation [34,35].However for small particles, as the tapping amplitude surpasses particle diameter, the near-field interaction is more and more with the silicon substrate on which the particles are adsorbed rather than with the particles.This leads to a signal saturation which approaches a value unity (s 3 (particle)/s 3 (Si) ~1) as can be seen on Fig. 8.We note that the saturation behavior is universal in nature for nanoparticles.It is not affected by the change in dielectric constants and therefore material compositions of the nanoparticles as shown by the similar trend for both the capped and uncapped particles.Interestingly, for capped particles higher tapping doesn't improve contrast, i.e. the contribution of the core AuNP diminishes as tapping increases [Fig.8].These results contrast with what was documented for layered thin films [21].In case of thin films, there is a 'seethrough effect' of the near-field, which means that the substrate under the thin coating layer contributes significantly to the signal [21].At increasing tapping amplitude, the interaction probes a deeper overall depth, therefore the influence of the material under the coated film increases.Thus, in a thin film coated Au sample, Au should have increasing influence; any low near-field contrast should increase as the tapping amplitude increases.But for small particles, the thin-film approximation is not valid.At larger tapping amplitudes, the contrast becomes more and more similar between the capped and uncapped particles at larger tapping amplitudes.These results suggest that in small particle near-field imaging, the use of lower tapping amplitude allows enhanced material specific identification of particles.
In conclusion, we successfully imaged and identified silica, Au and silica-capped AuNPs adsorbed on Si substrate using s-SNOM in the visible and infrared frequencies.Subsurface near-field interaction between tip and plasmon resonant or even a reflecting core nanoparticle combined with material specific optical contrast imaging of nanoparticles makes s-SNOM very attractive for many applications.Examples of these possible applications include quality control in pharmaceutical or nanochemistry, photothermal therapeutic applications and detecting dopants and defect sites in single semiconductor nanoparticles.

Fig. 1 .
Fig. 1.Schematics of the experimental setup showing optical beam paths of the s-SNOM for visible (left side of tip) and infrared frequencies (right side of tip).M = mirror, OI = optical isolator, BE = beam expander, and BS = beam splitter.

Figures 2 (
a) and 2(c) are transmission electron microscopy (TEM) images of monodispersed silica-capped AuNPs and bare silica nanoparticles prepared by drop-casting nanoparticle solutions (nanoComposix, nanocomposix.com)on a Si wafer.TEM images shown in Fig.2(a) indicate that the shapes of the composite silica-coated AuNPs and the thickness of the capping silica layer (average size of ~10 nm) are highly uniform.Nevertheless, slight non-uniformities of shape and size of the AuNP core and the capping layer should be considered in modeling near-field interactions of these particles and extracting quantitative information.Size statistics performed on samples of over 153 capped particles [Fig.2(b)] and 275 bare silica particles [Fig.2(d)]demonstrate particle size distributions of 10-20%, typical values for many colloidal samples.

Fig. 2 .
Fig. 2. Transmission electron microscope (TEM) images and size statistics for monodisperse nanoparticles.(a-b) Silica-capped AuNPs, showing uniformity of shape and thickness of the silica capping layer and (c-d) silica nanoparticles.N is the number of particles measured.

Fig. 3 .
Fig. 3. Near-field images of a mixture of 3 types of nanoparticles: silica, Au, and silica-capped AuNPs absorbed on a Si substrate.(a) AFM topographic image of the mixture and (b) third harmonic s3 near field amplitude image.(c-d) High resolution zoom in scan results of the 3 particle types shown in green circle in (a) in a 500 nm x 500 nm area, (c) topography, and (d) third harmonic s3 near-field amplitude.(e) Line profiles of the topography of nanoparticles (red and blue dashed lines) showing similar height of the two particles.(f) Different optical signal line profiles for the two particles shown in red and blue dashed lines.Amplitude signal line profiles shown in (f) are normalized to the signal of a Si substrate.

Fig. 4 .
Fig.4.Pixel-by-pixel correlation of measured near-field third harmonic signal contrast as a function of the topographic pixel height at 633 nm wavelength recorded for two samples, pure silica (blue data points) and silica-capped AuNPs (red data points) adsorbed on a Si substrate.Signal values on Si substrate are used to normalize all optical data.

3 Fig. 5 .
Fig. 5. Calculated results of the normalized optical signals for silica-capped AuNPs (dashed line) and silica nanoparticles (solid line) as a function of particle height.Superimposed on the theoretical curves are experimental signal average values of pixels near the center of particles.

Fig. 6 .
Fig. 6.(a) Pixel-by-pixel correlation of measured near-field second harmonic signal s2 as a function of the topographic pixel height at λ=10.7 µm wavelength recorded for two samples, silica (blue data points) and silica-capped AuNPs (red data points) adsorbed on a Si substrate.(b) Calculated results of the normalized optical signals for silica-capped AuNPs (dashed line) and silica nanoparticles (solid line) as a function of particle height.Superimposed on the theoretical curves are shown experimental signal average values of pixels near the center of particles.

Fig. 7 .
Fig. 7. Identification of particles based on their near-field amplitude contrast.(a) Topography of a mixed sample (b) Near-field amplitude contrast (c) Experimental signal average values of pixels near the center of particles red (capped), blue (uncapped) (d) Particles labeled according to their material contrast as capped (red) and uncapped (blue).

Fig. 8 .
Fig. 8. Third harmonic near-field amplitude signals versus tapping amplitude measured at λ=633 nm.Red dots represent data points on silica-capped Au and blue dots on pure silica nanoparticles.Signal values on Si substrate are used to normalize all optical data.