Template-Guided Self-Assembly of Discrete Optoplasmonic Molecules and Extended Optoplasmonic Arrays : nano Online

1 Introduction

The advancements of nanofabrication methods in the last 10 to 20 years have facilitated plasmonic nanoantennas [1–3] for a broad range of applications in the visible range of the electromagnetic spectrum. Due to the superb lightfocusing properties of metal nanoparticle clusters, plasmonic gap antennas allow the creation of electromagnetic hot-spots in which the incident E-field is enhanced by many orders of magnitude [4–8]. The hot-spots are of high interest for enhancing excitation rates or for boosting nonlinear optical effects [9–12]. At the same time, the density of states associated with the plasmon resonances in the nanostructures can result in a boost of the radiative rates of near-by quantum emitters [9, 13], and the combination of quantum emitter and plasmonic antenna is, thus, commonly referred to as a superemitter [14]. Due to Ohmic losses at optical frequencies in metals, plasmon resonances have short (femtoseconds to tens of femtosecond) lifetimes and relatively broad spectral features [15]. Since high dissipative losses and broad spectral features limit the performance of conventional metalbased plasmonic nanostructures in key applications, different strategies are currently pursued to mitigate these effects [16–19]. One promising approach seeks to integrate plasmonic nanostructures into a defined photonic environment where photonic and plasmonic modes can synergistically interact [20–25]. The coupling of localized surface plasmon resonances (LSPRs) with in-plane diffracted Rayleigh resonances to form surface lattice resonances (SLRs) in extended nanoparticle arrays is one example for this approach [26–29]. In an array of metal nanoparticles, the individual nanoparticles experience the inci-

Figure 1:

E-field intensity distributions in the discrete optoplasmonic dimer at the wavelength of the hybrid TE^14,1 (A, λ = 580.9 nm) and TM_13,1 (B, λ = 590 nm) mode. The dimer is composed of two 2 μm polystyrene microspheres (refractive index = 1.59), each encircled by four 148 nm diameter Au nanoparticles separated from the microsphere surface by 1 nm gaps. The insets show the direction and polarization of the incident plane wave. (C) Near-field intensity distribution at the TM_13,1 mode wavelength averaged over the in-plane angle of incidence. Reprinted with permission from Ahn et al. ACS Nano 6, 951 (2012). Copyright 2012 American Chemical Society.

dent E-field plus the E-field re-radiated by all of the other nanoparticles in the array. The effective polarizability, α_eff of the individual nanoparticles is then given as [18]:

α_{e f f} = \frac{1}{\frac{1}{α} - S}

where α refers to the single particle polarizability, and the retarded dipole sum, S, captures the effect of the array. It has been shown that a positive singularity in S gives rise to sharp spectral features [18, 29–32].

An alternative strategy for overcoming the limitations of conventional plasmonic nanocircuitry in a more compact footprint is based on the combination of plasmonic nanoantennas with dielectric microcavity resonators into structurally defined, discrete hybrid optoplasmonic structures [21, 22, 33–36]. Although whispering gallery mode (WGM) resonators have larger mode volumes and, therefore tend to show lower E-field enhancements than metallic nanoantennas, their eigenmodes have low losses and, therefore, create higher Qfactors than metallic nanostructures [37]. The design strategy of the discrete optoplasmonic structures is, therefore, to position plasmonic antennas at defined locations in the evanescent field of the dielectric resonators. The discrete optoplasmonic structures satisfying this fabrication requirement induce the formation of photonic-plasmonic hybrid-modes in which part of the mode volume is localized around metallic nanoparticles in electromagnetic hot-spots [38]. A representative example for this approach is given in Figure 1. The hybrid structure contains electromagnetic hot-spots located outside of the dielectric microsphere around the metal nanoparticles where the E-field is available for interactions with the ambient medium.

We note here in passing that discrete optoplasmonic structures do not only provide unique opportunities for enhancing E-fields but also for engineering LDOS in the vicinity of fluorescent emitters. In the case of the so-called superemitters, which represent nanostructures that combine quantum emitters and plasmonic antennas, the effect of the enhanced LDOS on the emission efficiency has been shown to differ depending on the superemitter’s emission efficiency in the absence of the microsphere resonator [39]. For emitters with emission wavelengths in resonance with the antenna plasmon, two cases can be distinguished. For emitters with low emission rates, the radiative rate can be improved by enhanced LDOS. Intriguingly, for bright emitters that operate close to the unitary limit the emission rate has been predicted to drop with increasing LDOS when radiation damping reduces the polarizability of the antenna [39, 40].

A second class of optoplasmonic materials, the extended optoplasmonic nanoparticle array, is inspired by the formation of SLRs in conventional metal nanoparticle arrays. In optoplasmonic arrays metal nanoparticles are replaced as a sole building block through a combination of high refractive index dielectric nanoparticles and clusters of strongly coupled metal nanoparticles [24]. Noble metal nanoparticle clusters are chosen as plasmonic component as they provide much higher E-field intensities than individual nanoparticles. Furthermore, plasmon coupling shifts the plasmon resonance wavelength into the red [41, 42], where the slope of the real part of the gold (Au) dielectric function as a function of wavelength is steeper [43, 44]. This behavior is beneficial, for example, for the design of sensitive colorimetric sensors [45].

Conventional plasmonic nanoparticle cluster arrays (NCAs) [42, 46, 47] have been investigated in detail and have been shown to provide new opportunities for engineering a multiscale E-field enhancement through electromagnetic coupling in the near-, intermediate-, and farfield region [46]. The availability of two building blocks in optoplasmonic arrays provides new opportunities for controlling the light field distribution in the plane of the array and for efficiently “squeezing” light into electromagnetic hot-spots. In optoplasmonic arrays low-loss dielectric nanoparticles can be used to form grating-like structures that generate in-plane diffracted resonances [24]. In arrays with appropriate morphology the photonic resonances can create enhanced E-field intensity at the inter-

Figure 2:

(A) – (C): Schematic outline of the individual steps (left) of the template guided self-assembly for optoplasmonic atoms and molecules, and SEM images (right) of the corresponding hybrid structures. Scale bars are 1 μm. Reprinted with permission from Ahn et al. ACS Nano 7, 4470 (2013). Copyright 2013 American Chemical Society.

stitial spaces between the dielectric nanoparticles. If clusters of electromagnetically strongly coupled noble metal nanoparticles are located at these locations, the in-plane diffracted light is further localized by these efficient plasmonic antennas to achieve a cascaded E-field intensity enhancement.

Discrete optoplasmonic structures and extended optoplasmonic nanoparticle arrays have in common that they contain plasmonic antennas in a complex photonic environment to enhance or modulate their electromagnetic response. For the former this environment is defined by the WGM resonator, for the latter it is generated by a twodimensional nanoparticle array. The fabrication of both optoplasmonic materials requires the integration of building blocks with different chemical composition and, especially in the case of discrete optoplasmonic structures, of different sizes in precise geometric arrangements. In this review we will introduce template-guided self-assembly approaches to address these challenges and characterize selected examples of optoplasmonic materials obtained with them.

2 Discrete optoplasmonic atoms and molecules

Discrete optoplasmonic structures that contain a single dielectric microsphere resonator and one or more plasmonic antennas are referred to as optoplasmonic atoms in the following [21]. We chose this nomenclature to indicate similarity with the more established concept of the “photonic atom” [48] as well as to imply that these structures represent the simplest possible unit of optoplasmonic materials of this kind. More complex optoplasmonic molecules can be obtained by integrating multiple optoplasmonic atoms into networks. Hybrid structures containing more than one OM as a WGM resonator are referred to as optoplasmonic molecules [21].

The particular challenge for the experimental realization of optoplasmonic atoms and molecules as the one outlined in Figure 1 is the requirement to position a plasmonic antenna in the close vicinity to a microsphere resonator. While this can be accomplished – in principle – by directly attaching noble metal nanoparticles to a dielectric resonator [49–53], this approach is very limited in scope as it does not provide control over the location and, in the case of anisotropic nanoparticles, orientation of the noble metal nanoparticles. Geometric control over the positioning of nanoparticles is crucial, in particular, for a successful integration of the optoplasmonic structures into an on-chip platform. Furthermore, it is difficult to create more complex antenna structures, such as gap antennas, on OMs in a controlled fashion by nanoparticle attachment. Simple attachment of nanoparticles provides also no degree of freedom for varying the separation between the antenna and the dielectric resonator. We conclude that the large difference in size between OMs (microns) and plasmonic antennas (tens of nanometers) creates significant challenges for the integration of the individual components into defined hybrid structures. We decided to address these challenges with directed self-assembly strategies that utilize Au nanostructure tipped posts as pattern to guide the assembly of OMs in discrete structures. Our rationale was that if the length of the posts is chosen as 1/2 of the OM diameter, the plasmonic nanostructures are automatically positioned in the evanescent field located in the equatorial plane of the WGM resonators.

2.1 Fabrication of discrete optoplasmonic structures comprising OMs and nanoantennas

The template-guided self-assembly approach for positioning plasmonic antennas in the equatorial plane of WGM resonators is outlined in Figure 2 with the scanning electron microscope images taken at each step of the fabrication [21, 22]. In the first step noble metal nanoparticle tipped pillars that define binding cavities for microspheres

Figure 3:

(A), (C), (E) Top and (B), (D), (F) side view (30°tilt angle) of discrete optoplasmonic structures with 1, 2 or 3 OMs (diameter: 2 μm). Side views of pillar scaffolds (G) before and (H) after integration of OMs. The Au nanoparticles are clearly discernible at the tip of the pillars. Scale bars are 1 μm. Reprinted with permission from Ahn et al. ACS Nano 6, 951 (2012). Copyright 2012 American Chemical Society.

are created. In a subsequent step, these binding cavities are filled with dielectric microspheres by dispersing a solution of dielectric microspheres across the surface (convective self-assembly). We have used polystyrene spheres with a diameter of 2 μm. The height (h) of the nanopillars (including the radius of the Au nanoparticles) is chosen to correspond to the radius of the microspheres of interest, to localize the metal nanoparticles in the equatorial plane of the WGM resonator. One additional advantage of the outlined approach is that it facilitates the generation of networks of optoplasmonic atoms that contain the plasmonic components in one defined plane where they can interact synergistically. The separation between the nanopillars can be conveniently varied by the geometry of the fabricated post-pattern, for instance, to change the average separation between adjacent microspheres and, thus, to modulate the coupling between the OMs [54]. In the following, we provide a detailed description of the fabrication of discrete optoplasmonic structures containing microspheres of 2 μm diameter. With some modifications, the same approach can also be applied to generate optoplasmonic structures containing microspheres with larger diameters.

First, regular arrays of nanoholes of 150 nm diameter are patterned in 200 nm thick PMMA layers spincoated on a quartz (or silicon) substrate through electron beam lithography (EBL) (Figure 2A). A Ti/Au/Cr layer with a thickness of 10/80/35 nm, respectively, is then deposited on the EBL-patterned surface by electron beam evaporation with deposition rates of 0.5/1.0/0.5 Å/s for each metal to create metal nanoparticle dimer arrays as shown in Figure 2A. The Ti layer improves adhesion of the Au structure to the substrate, and a Cr layer protects the Au structures from anisotropic dry etch gases. After the PMMA layer is lifted off by immersion in acetone for 1 min followed by 10, 20, and 30 s of sonication each in a fresh acetone bath, substrates are ready for reactive ion etching (RIE, Plasma-Therm. model 790). We commonly use a mixture of CHF₃/O₂ (50/5 sccm, standard cubic centimeters per minute) at a pressure of 200 mTorr and a power of 150 W for a total etching time of 18–20 min. The anisotropic etching process removes the substrate material only in exposed areas, which are not protected by Cr layers, resulting in the formation of nanopillar arrays with Cr layer tips (Figure 2B). The height of the nanopillars can be controlled through the RIE process parameters, including the etching time, power, pressure, and ratio between the two mixed gases. A RIE chamber is purged with O₂ for 15 min and prerun with CHF₃/O₂ for 30 min prior to each RIE process on samples to ensure reproducible etching results. The Cr layer is subsequently removed by swirling in Cr etchant solution at 40°C for 15 s, after which the substrate is rinsed with copious amounts of distilled water. The substrate can then be heated in a rapid thermal annealer (Molecular Process Technology Corp.) to 800°C (3 min) to create spherical Au nanoparticles on top of the nanopillars. The regular arrays of micropillar cavities with defined widths and depths are then ready for use as template for the assembly of optoplasmonic structures in the subsequent assembly step. To generate photonic control structures (no metal nanoparticles), all metal layers are removed from the substrate after nanopillar formation by immersion in etchant solutions.

The etched cavities are filled with polystyrene (PS) microspheres using a convective self-assembly strategy (Figure 2C). An aqueous suspension of PS microspheres (2 μm in diameter; 1%; 50 μL) is sandwiched between the patterned microcavity substrate and a blank quartz substrate separated by a 380 μm gap. Upon the evaporation of the water, the meniscus of the microsphere-containing solution is dragged across the substrate surface by capillary forces. These forces have a component pointing perpendicular to the patterned surface [55], which enables an ef-

Figure 4:

Characterization of the optoplasmonic dimer through scattering darkfield microscopy/spectroscopy. (A) Geometry of the optoplasmonic dimer, scattering image and spectrally resolved scattering image. (B) Experimental (top) and simulated (bottom) scattering spectra of the optoplasmonic dimer (red) and photonic control (blue) without Au nanoparticle tips. (C) Full width at half maximum (fwhm) of the individual modes shown in (B). Reprinted with permission from Ahn et al. ACS Nano 6, 951 (2012). Copyright 2012 American Chemical Society.

ficient trapping of microspheres in the cavities formed by the fabricated pillars. The insertion of microspheres into the Au nanoparticle functionalized pillars completes the assembly of the discrete optoplasmonic structures.

Figure 3 shows scanning electron microscope (SEM) images of discrete optoplasmonic structures that contain one (A, B), two (C, D), or three (E, F) 2 μm diameter OMs. The edge-to-edge distance (d_p) between neighboring pillars in the optoplasmonic molecules (C-F) was chosen as d_p = 330 nm to enable strong electromagnetic coupling between the individual constituent optoplasmonic atoms. At d_p = 330 nm the embedded microspheres are nearly touching, ensuring an efficient coupling of the WGM modes in the optoplasmonic molecules. The height of the pillars (h) in the structures was determined to be ∼870 nm from SEM images acquired at a tilt angle of θ = 80° using the formula h = h_m/sin(θ), where h_m is the measured height of pillars in the tilted SEM image. In the magnified view of a representative nanopillar scaffold at θ = 70° in Figure 3G the spherical Au nanoparticles on top of the created nanopillars are clearly visible. The diameter of the Au nanoparticles in these optoplasmonic structures is measured as 148 nm. The side view in Figure 3H confirms that the assembly process was successful and that spherical Au nanoparticles are located in the equatorial plane of the PS microspheres inserted into the binding cavities.

2.2 Optical responses of discrete optoplasmonic structures in the near- and far-field

Figure 4A shows the far-field scattering image as well as the spectrally resolved scattering image of an optoplasmonic dimer obtained with unpolarized white-light. The optoplasmonic dimer contains two OMs, Au nanoparticles, and nanopillars, as illustrated in the schematic of Figure 4A. The imaging plane lies in the equatorial plane of the optoplasmonic dimer and the scattering spectra of the optoplasmonic dimer were obtained at the center of the structure (dotted red boxes) where the coupling of the two microspheres is highest and the signal-to-noise low-

Figure 5:

Near-field characterization of an optoplasmonic atom comprising an Au nanoparticle separated from an OM by a 1 nm gap. The diameter of the PS microsphere is 2 μm and the diameters of the Au nanoparticles are (red) 148 or (blue) 100 nm, as indicated. (A) E-field intensity enhancement in the gap between the Au nanoparticle and the OM. The location of the numerical director for evaluating the E-field intensity is indicated by a red dot in the inset. (B) and (C) Maps of the spatial E-field intensity enhancement around the Au nanoparticle at λ = 589.7 nm and λ = 614.4 nm. Reprinted with permission from Ahn et al. ACS Nano 6, 951 (2012). Copyright 2012 American Chemical Society.

est. The vertical lines that appear in the spectrally resolved scattering image in Figure 4A (right) indicate distinct WGM peaks of an optoplasmonic dimer. The top panel in Figure 4B shows representative normalized scattering spectra of an optoplasmonic dimer (red) and of a control with an overall identical structure but without Au nanoparticles on the nanopillars (photonic dimer, blue). The bottom panel in Figure 4B contains the corresponding generalized multiple particle Mie theory (GMT)-simulated scattering spectra of the optoplasmonic (red) and the control photonic (blue) dimer in an ambient medium of refractive index n = 1. The experimental scattering spectra of the optoplasmonic dimer are broadened when compared with the spectra of the control photonic control (microsphere with pillar but no Au nanoparticles). This broadening becomes more obvious in Figure 4C, where the FWHM of the fitted peaks is plotted. The spectral broadening is more pronounced in the experiments than in the simulations, but the latter reproduces the general trend well. The TE modes are broader than the TM modes in the experimental spectra, although the simulations predict the opposite behavior. The detected TE modes have orbits that are pointing perpendicular to the substrate. Consequently, the TE modes experience more losses at the substrate than the detected TM modes whose orbits lie in a plane parallel to the substrate [56]. Substrate effects were not considered in the performed electromagnetic simulations.

The spectral broadening of the TE and TM modes observed in the optoplasmonic dimer in the experimental and, albeit to a lesser degree, also in the simulated spectra is a direct consequence of photonic-plasmonic mode coupling. One consequence of this coupling is the redistribution of electromagnetic energy from the OMs into the nanoparticles. Dissipative losses in the Au nanoparticles shorten the lifetime of the hybridized modes and account for the observed broadening of the spectral features.

In the optoplasmonic dimer every OM is located in the effective field of four Au nanoparticles. Although this significant association of the OMs with Au nanoparticles leads to a measurable peak broadening, the induced broadening is relatively small, which reaffirms the advantage of our experimental strategy to place Au nanoparticles into the evanescent field of OMs in a controlled fashion. Figure 4B also contains the scattering spectra of the Au nanoparticles used in this work as dashed gray lines. The Au nanoparticles show a broad resonance that peaks at approximately 555 nm. The scattering peaks of the optoplasmonic dimer are significantly sharper, which has tangible effects for a broad range of applications.

After investigating the far-field properties of the optoplasmonic dimers, we performed GMT simulations to characterize the E-field intensity enhancement around Au nanoparticles of two different diameters (red: d = 148 nm, blue: d = 100 nm) separated from the OM surface by a 1 nm air gap (Figure 5A). Our simulations show that the most efficient coupling between photonic and plasmonic components is achieved if the light is incident along the OM— nanoparticle axis. The k-vector and polarization direction for this case are included in the inset of Figure 5A. The E-field intensity on the nanoparticle surface under these conditions is enhanced by up to two orders of magnitude over that of a pure metal nanoparticle (Figure 5A). The simulations show that in the long wavelength tail of the LSPR spectrum the coupling of LSPR with the WGMs of the OM yields E-field intensity maxima that coincide with discrete WGM resonances. Close to the LSPR peak wavelength the strong electromagnetic coupling, especially for the larger Au NP, results in significant perturbations of the WGM peak shape. Figure 5B and 5C shows E-field intensity maps for two different modes. The maps show again an E-field intensity concentration around the nanoparticle and, thus, outside of the OM.

The OM in the vicinity of a plasmonic antenna does not only enhance the peak E-field intensity and the spatial E-field intensity distribution but is also anticipated to modulate the emission properties of superemitters. Inter-

Figure 6:

Scheme of template-guided self-assembly approach of the extended optoplasmonic arrays. (A) An array with two different binding sites of D1 and D2 is created on a PMMA film. (B) The separation between two identical binding sites is defined by the grating period, Λ. (C) Dispersing a suspension of TiO₂ nanoparticles that only bind to the larger binding site D1 creates (D), a dielectric nanoparticle array. (E) Smaller Au nanoparticles are then targeted to the vacant binding sites D2, generating the optoplasmonic array (F) after lift-off. Reprinted with permission from Hong et al. Adv. Funct. Mater. 24, 739 (2014). Copyright 2013 Wiley-VCH GmbH & Co. KgaA.

estingly, we found that if superemitters (Cy-3 functionalized nanoparticle dimer) are located in the vicinity of an OM, the emission is highly directional into the OM, which efficiently traps and recirculates the photons [22].

3 Extended optoplasmonic arrays

An alternative approach to create a photonic environment around a plasmonic antenna is the integration of the antenna into a regular array of dielectric nanoparticles in which the individual elements are diffractively coupled [27, 57]. Optoplasmonic arrays are unique in the sense that they combine high refractive index dielectric nanoparticles and noble metal nanoparticles (or their clusters) into arrays, which can use the dielectric component to focus the light onto nanoparticle clusters [24], where the incident light is localized into deeply subdiffraction limit hot-spots via excitation of cluster plasmons. The challenge of this approach is to position nanoparticles of different composition at predefined locations. A second complication derives from the need to limit the interparticle separations in the metal clusters to a few nanometers to maximize the near-field coupling. As we will show in the following, template-guided self-assembly is a versatile fabrication strategy capable of addressing both of these challenges.

3.1 Fabrication of extended optoplasmonic arrays through template guided self-assembly

In the first step of the assembly procedure nanoparticle assembly sites are created through a lithographically fabrication technique (e.g. EBL) for the subsequent assembly of colloidal NPs onto the lithographically defined assembly sites [24]. While EBL makes it possible to define the separation, Λ, between individual assembly sites on the tens to hundreds of nanometer scale, the separation within the self-assembled clusters is determined by the assembly conditions as well as the surface of the assembled nanoparticles, and is typically on the few nanometer length scale or below [58]. The unique flexibility for realizing nanoparticle separations over such a broad range makes the template guided self-assembly a useful tool for realizing multiscale nanoparticle arrays.

The diameter (D) [42, 47] and shape [58] of the assembly sites represent control parameters for guiding nanoparticles to specific assembly sites. In the case of the optoplasmonic nanoparticle arrays these control parameters are utilized to generate an extended regular structure that contains TiO₂ nanoparticles and clusters of 60 nm Au nanoparticles on separate lattice sites [24]. The assembly strategy of optoplasmonic arrays is schematically outlined in Figure 6. First, two different binding sites (D1 and D2) are templated in the electron beam resist (A, B). Then the larger TiO₂ nanoparticles are immobilized on the binding sites with larger diameter (C, D). Subsequently, smaller Au nanoparticles are bound to the vacant binding sites with smaller diameter (E) and, after photoresist lift-off, the extended optoplasmonic array is generated (F). For details regarding the assembly process, please refer to ref. [24]. Figure 7 shows SEM images of different arrays obtained from 60-nm Au nanoparticles and 250-nm-diameter TiO₂ nanoparticles with binding sites containing diameters of D2 = 140 nm and D1 = 270 nm, respectively [24]. The assembly yielded regular arrays containing TiO₂ NPs located on the vertices of regular squares around a central Au nanoparticle cluster. The average number of Au nanoparticles in the clusters was ∼6. We have shown previously that the average cluster resonance wavelength increases for two-dimensional clusters with growing size up to a cluster size of approximately four nanoparticles [42, 47]. The additional red-shift stagnates with larger cluster size. Due to this behavior we expect only moderate fluctuations in

Figure 7:

SEM images of the extended optoplasmonic arrays containing 250 nm TiO₂ nanoparticles and 60 nm Au nanoparticles. (A) – (C) Complete optoplasmonic hybrid arrays at different magnifications. (D) TiO₂ nanoparticle array. (E) Au nanoparticle cluster array. Scale bars are 500 nm in (A), (C) – (E), and 2 μm in (B). The grating period is set for all arrays, Λ = 1000 nm. Reprinted with permission from Hong et al. Adv. Funct. Mater. 24, 739 (2014). Copyright 2013 Wiley-VCH GmbH & Co. KgaA.

Figure 8:

Simulated E-field intensity enhancement spectra at the hottest spot in (A) an array of 20 × 20 TiO₂ nanoparticles (solid) and a single trimer of Au nanoparticles (dashed), (B) array of 20 x 20 Au nanoparticle trimers, (C) combined optoplasmonic array. The incident k-vector is normal to the array plane in all models. Reprinted with permission from Hong et al. Adv. Funct. Mater. 24, 739 (2014). Copyright 2013 Wiley-VCH GmbH & Co. KgaA.

the resonance wavelength of the individual clusters with sizes of approximately six nanoparticles.

3.2 Morphology-dependent E-field enhancement in extended optoplasmonic arrays

To quantify the synergistic gain that results from electromagnetic interactions between metallic and dielectric building blocks, we simulated the peak E-field intensity enhancement spectra for (i) a TiO₂ nanoparticle only array, (ii) an Au nanoparticle cluster only array, and (iii) the extended optoplasmonic hybrid array that contains both Au nanoparticle clusters and TiO₂ nanoparticles. In the hybrid array, Au nanoparticle clusters were positioned in the center of a square unit cell formed by four TiO₂ nanoparticles. For simplicity we modeled the Au nanoparticle cluster as a trimer. The results of our multiple sphere T-matrix simulations [59] are summarized in Figure 8. The E-field enhancement for the all-dielectric TiO₂ NP array in Figure 8A shows discrete photonic resonances, whose spectral positions red-shift with increasing grating period, Λ. As expected, the Au NCAs (Figure 8B) show much higher E-field enhancements than the all-dielectric nanoparticle array. The near-field spectra of the metal arrays are dominated by the cluster plasmon resonance with a simulated peak wavelength of ∼720 nm. The near-field intensity spectrum for an individual Au NP trimer is included for comparison in Figure 8A (black dashed line).

Optoplasmonic arrays (Figure 8C) achieve a further increase of the near-field intensity beyond the Au NCA level when the resonance wavelength of the grating period dependent photonic array mode overlaps with that of the cluster plasmon resonance. Under these conditions photonic–plasmonic mode coupling facilitates a cascaded

Figure 9:

SERS spectra of para-mercaptoaniline (pMA) obtained on the optoplasmonic arrays with Λ = 700 – 1100 nm. The inset shows filling fraction-corrected peak SERS signal intensities as a function of Λ. Reprinted with permission from Hong et al. Adv. Funct. Mater. 24, 739 (2014). Copyright 2013 Wiley-VCH GmbH & Co. KgaA.

enhancement of the E-field intensity provided by the individual clusters. According to our simulations optoplasmonic arrays with Λ ≈ 800 nm fulfill this requirement and, indeed, the overlap of array and nanoparticle cluster resonance yields an additional enhancement of the E-field intensity for this grating period.

To test the Λ-dependence of the E-field intensity enhancement experimentally, we set out to measure the surface enhanced Raman scattering (SERS) spectra of the small test molecule para-mercaptoaniline (pMA) chemisorbed onto the extended optoplasmonic arrays with Λ = 700 – 1100 nm. As the SERS signal intensity of a molecule scales as the product of the E-field intensities at the pump and Raman scattering wavelengths [60], the SERS signal intensity can provide valuable information of the E-field intensity at the probe and emission wavelengths. Figure 9 shows the background corrected SERS spectra in the spectral range 1000 – 1180 cm^–1 that contains the C-S stretch mode at 1077 cm^–1. The SERS signal intensity—even after correction of the different filling fractions (see inset)—shows a clear Λ-dependence and peaks at Λ = 900 nm. The Λ-dependence of the E-field intensity at the plasmon resonance wavelength is consistent with our simulations. The discrepancy in the precise value of Λ that maximizes the E-field intensity between simulations and experiment is justified by necessary simplifications in the computational model. Overall, the correspondence between simulation and experiment is good and both electromagnetic simulations and SERS spectroscopy confirm the synergistic interaction between dielectric and metallic components in the optoplasmonic nanoparticle array.

4 Conclusion

Template-guided self-assembly is a versatile fabrication strategy for the fabrication of novel electromagnetic materials from different building blocks. We have taken advantage of the method’s ability to combine OM and nanogap antennas into discrete optoplasmonic atoms and molecules with defined geometry. Furthermore, we utilized template-guided assembly strategies to integrate high refractive index dielectric nanoparticles and clusters of electromagnetically strongly coupled noble metal nanoparticles into hetero-nanoparticle arrays. The ability to position metallic and dielectric nano- and microstructures at predefined locations with high lateral resolution using template guided self-assembly strategies provides rational control over the interplay of photonic and plasmonic modes in these new optoplasmonic materials. We conclude that the combination of mutually synergistic properties of metallic and dielectric materials in optoplasmonic structures generates new functionalities for sensing, light harvesting, information processing and many other applications.

Acknowledgement

This work was supported by the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Science and Engineering under Award DOE DE-SC0010679.

References

[1]
Muhlschlegel P., Eisler H.J., Martin O.J.F., Hecht B., Pohl D.W., Resonant optical antennas, Science 2005, 308, 1607 – 09.Google Scholar
[2]
Novotny L., Effective wavelength scaling for optical antennas, Phys Rev Lett 2007, 98, 266802.Google Scholar
[3]
Novotny L., van Hulst N., Antennas for light, Nature Photon 2011, 5, 83 – 90.Google Scholar
[4]
Halas N.J., Lal S., Chang W.S., Link S., Nordlander P., Plasmons in strongly coupled metallic nanostructures, Chemical Reviews 2011, 111, 3913 – 61.Google Scholar
[5]
Bharadwaj P., Deutsch B., Novotny L., Optical antennas, Adv Opt Photon 2009, 1, 438 – 83.Google Scholar
[6]
Li J.J., Salandrino A., Engheta N., Shaping light beams in the nanometer scale: A Yagi-Uda nanoantenna in the optical domain, Phys Rev B 2007, 76, 245403.Google Scholar
[7]
Muskens O.L., Giannini V., Sanchez-Gil J.A., Rivas J.G., Optical scattering resonances of single and coupled dimer plasmonic nanoantennas, Optics Express 2007, 15, 17736 – 46.Google Scholar
[8]
Schuller J.A., Barnard E.S., Cai W., Jun Y.C., White J.S., Brongersma M.L., Plasmonics for extreme light concentration and manipulation, Nat Mater 2010, 9, 193 – 204.Google Scholar
[9]
Muskens O.L., Giannini V., Sanchez-Gil J.A., Gomez Rivas J., Strong enhancement of the radiative decay rate of emitters by single plasmonic nanoantennas, Nano Lett 2007, 7, 2871-75.Google Scholar
[10]
Zhang J., Fu Y., Chowdhury M.H., Lakowicz J.R., Metal-enhanced single-molecule fluorescence on silver particle monomer and dimer: Coupling effect between metal particles, Nano Lett 2007, 7, 2101 – 07.Google Scholar
[11]
Li K.R., Stockman M.I., Bergman D.J., Self-similar chain of metal nanospheres as an efficient nanolens, Phys Rev Lett 2003, 91, 227402.Google Scholar
[12]
Camden J.P., Dieringer J.A., Wang Y., et al., Probing the structure of single-molecule surface-enhanced Raman scattering hot spots, J Am Chem Soc 2008, 130, 12616 – 17.Google Scholar
[13]
Giannini V., Fernandez-Dominguez A.I., Heck S.C., Maier S.A., Plasmonic nanoantennas: Fundamentals and their use in controlling the radiative properties of nanoemitters, Chem Rev 2011, 111, 3888 – 912.Google Scholar
[14]
Farahani J.N., Pohl D.W., Eisler H.J. and Hecht B., Single quantum dot coupled to a scanning optical antenna: A tunable superemitter, Phys Rev Lett 2005, 95, 017402.Google Scholar
[15]
Link S., El-Sayed M.A., Spectral properties and relaxation dynamics of surface plasmon electronic oscillations in gold and silver nanodots and nanorods, J Phys Chem B 1999, 103, 8410 – 26.Google Scholar
[16]
Fan J.A., Wu C., Bao K., et al., Self-assembled plasmonic nanoparticle clusters, Science 2010, 328, 1135 – 38.Google Scholar
[17]
Noginov M.A., Zhu G., Belgrave A.M., et al., Demonstration of spaser-based nanolaser, Nature 2009, 460, 1110 – 12.Google Scholar
[18]
Zou S.L., Schatz G.C., Narrow plasmonic/photonic extinction and scattering line shapes for one and two dimensional silver nanoparticle arrays, J Chem Phys 2004, 121, 12606 – 12.Google Scholar
[19]
Brunazzo D., Descrovi E., Olivier J.M., Narrowband optical interactions in a plasmonic nanoparticle chain coupled to a metallic film, Optics Letters 2009, 34, 1405 – 07.Google Scholar
[20]
Benson O., Assembly of hybrid photonic architectures from nanophotonic constituents, Nature 2011, 480, 193 – 99.Google Scholar
[21]
Ahn W., Boriskina S.V., Hong Y., Reinhard B.M., Photonicplasmonic mode coupling in on-chip integrated optoplasmonic molecules, ACS Nano 2012, 6, 951 – 60.Google Scholar
[22]
Ahn W., Hong Y., Boriskina S.V., Reinhard B.M., Demonstration of efficient on-chip photon transfer in self-assembled optoplasmonic networks, ACS Nano 2013, 7, 4470 – 78.Google Scholar
[23]
Hong Y., Pourmand M., Boriskina S.V., Reinhard B.M., Enhanced light focusing in self-assembled optoplasmonic clusters with subwavelength dimensions, Adv Mater 2013, 25, 115 – 19.Google Scholar
[24]
Hong Y., Qiu Y., Chen T., Reinhard B.M., Rational assembly of optoplasmonic hetero-nanoparticle arrays with tunable photonicplasmonic resonances, Adv Funct Mater 2014, 24, 739 – 46.Google Scholar
[25]
Mukherjee I., Hajisalem G., Gordon R.E., One-step integration of metal nanoparticle in photonic crystal nanobeam cavity, Optics Express 2011, 19, 22462 – 69.Google Scholar
[26]
Vecchi G., Giannini V., Gomez Rivas J., Shaping the fluorescent emission by lattice resonances in plasmonic crystals of nanoantennas, Phys Rev Lett 2009, 102, 146807.Google Scholar
[27]
Vecchi G., Giannini V., Gomez Rivas J., Surface modes in plasmonic crystals induced by diffractive coupling of nanoantennas, Phys Rev B 2009, 80, 201401.Google Scholar
[28]
Vakevainen A.I., Moerland R.J., Rekola H.T., et al., Plasmonic surface lattice resonances at the strong coupling regime, Nano Lett 2013, 14, 1721 – 27.Google Scholar
[29]
Auguié B. and Barnes W.L., Collective resonances in gold nanoparticle arrays, Phys Rev Lett 2008, 101, 143902.Google Scholar
[30]
Markel V.A., Divergence of dipole sums and the nature of non-lorentzian exponentially narrow resonances in onedimensional periodic arrays of nanospheres, J Phys B: At Mol Opt Phys 2005, 38, 115 – 21.Google Scholar
[31]
Zou S., Schatz G.C., Response to “comment on ‘silver nanoparticle array structures that produce remarkable narrow plasmon line shapes”’[J. Chem. Phys. 120, 10871 (2004)], J Chem Phys 2005, 122, 097102.Google Scholar
[32]
Markel V.A., Comment on “silver nanoparticle array structures that produce remarkably narrow plasmon line shapes”[J. Chem. Phys. 120, 10871 (2004)], J Chem Phys 2005, 122, 097101.Google Scholar
[33]
Zou S., Schatz G.C., Combining micron-size glass spheres with silver nanoparticles to produce extraordinary field enhancements for surface-enhanced raman scattering applications, Isr J Chem 2006, 46, 293 – 97.Google Scholar
[34]
Hong Y., Ahn W., Boriskina S.V., Zhao X., Reinhard B.M., Directed assembly of optoplasmonic hybrid materials with tunable photonic-plasmonic properties, J Phys Chem Lett 2015, 6, 2056 – 64.Google Scholar
[35]
Barth M., Schietinger S., Fischer S., et al., Nanoassembled plasmonic-photonic hybrid cavity for tailored light-matter coupling, Nano Lett 2010, 10, 891 – 95.Google Scholar
[36]
Devilez A., Stout B., Bonod N., Compact metallo-dielectric optical antenna for ultra directional and enhanced radiative emission, ACS nano 2010, 4, 3390 – 96.Google Scholar
[37]
Vernooy D.W., Furusawa A., Georgiadis N.P., Ilchenko V.S., Kimble H.J., Cavity QED with high-Q whispering gallery modes, Phys Rev A 1998, 57, R2293.Google Scholar
[38]
Ahn W., Boriskina S.V., Hong Y., Reinhard B.M., Electromagnetic field enhancement and spectrum shaping in plasmonically integrated optical vortices, Nano Lett 2012, 12, 219 – 27.Google Scholar
[39]
Frimmer M., Koenderink A.F., Superemitters in hybrid photonic systems: A simple lumping rule for the local density of optical states and its breakdown at the unitary limit, Phys Rev B 2012, 86, 235428.Google Scholar
[40]
Frimmer M., Koenderink A.F., Spontaneous emission control in a tunable hybrid photonic system, Phys Rev Lett 2013, 110, 217405.Google Scholar
[41]
Yan B., Boriskina S.V., Reinhard B.M., Optimizing gold nanoparticle cluster configurations (n = 7) for array applications, J Phys Chem C 2011, 115, 4578 – 83.Google Scholar
[42]
Yan B., Thubagere A., Premasiri R., Ziegler L., Dal Negro L., Reinhard B.M., Engineered SERS substrates with multiscale signal enhancement: Nanoparticle cluster arrays, ACS Nano 2009, 3, 1190 – 202.Google Scholar
[43]
Li Z., Butun S., Aydin K., Touching gold nanoparticle chain based plasmonic antenna arrays and optical metamaterials, ACS Photonics 2014, 1, 228 – 34.Google Scholar
[44]
Johnson P.B., Christy R.W., Optical constants of the noble metals, Phys Rev B 1972, 6, 4370.Google Scholar
[45]
König M., Rahmani M., Zhang L., et al., Unveiling the correlation between nanometer-thick molecular monolayer sensitivity and near-field enhancement and localization in coupled plasmonic oligomers, ACS nano 2014, 8, 9188 – 98.Google Scholar
[46]
Yan B., Boriskina S.V., Reinhard B.M., Design and implementation of noble metal nanoparticle cluster arrays for plasmon enhanced biosensing, J Phys Chem C 2011, 115, 24437 – 53.Google Scholar
[47]
Yang L., Yan B., Premasiri R.W., Ziegler L.D., Dal Negro L., Reinhard B.M., Engineering nanoparticle cluster arrays for bacterial biosensing: The role of the building block in multiscale sers substrates, Adv Funct Mater 2010, 20, 2619 – 28.Google Scholar
[48]
Arnold S. Microspheres, Photonic atoms and the physics of nothing, American Scientist 2001, 89, 414.Google Scholar
[49]
Shopova S., Blackledge C., Rosenberger A., Enhanced evanescent coupling to whispering-gallery modes due to gold nanorods grown on the microresonator surface, Appl Phys B: Laser Opt 2008, 93, 183 – 87.Google Scholar
[50]
Melnikau D., Savateeva D., Chuvilin A., Hillenbrand R., Rakovich Y.P., Whispering gallery mode resonators with J-aggregates, Optics Express 2011, 19, 22280.Google Scholar
[51]
Santiago-Cordoba M.A., Boriskina S.V., Vollmer F., Demirel M.C., Nanoparticle-based protein detection by optical shift of a resonant microcavity, Appl Phys Lett 2011, 99, 073701.Google Scholar
[52]
Baaske M.D., Foreman M.R., Vollmer F., Single-molecule nucleic acid interactions monitored on a label-free microcavity biosensor platform, Nat Nanotechnol 2014, 9, 933 – 39.Google Scholar
[53]
Gartia M.R., Seo S., Kim J., et al., Injection-seeded optoplasmonic amplifier in the visible, Scientific reports 2014, 4, 6168.Google Scholar
[54]
Kanaev A.V., Astratov V.N., Cai W., Optical coupling at a distance between detuned spherical cavities, Appl Phys Lett 2006, 88, 111111.Google Scholar
[55]
Xia Y.N., Yin Y.D., Lu Y., McLellan J., Template-assisted selfassembly of spherical colloids into complex and controllable structures, Adv Funct Mater 2003, 13, 907 – 18.Google Scholar
[56]
Mitsui T., Wakayama Y., Onodera T., Takaya Y., Oikawa H., Observation of light propagation across 90° corner in chains of microspheres on a patterned substrate, Optics Letters 2008, 33, 1189 – 91.Google Scholar
[57]
de Abajo F.J.G., Colloquium: Light scattering by particle and hole arrays, Rev Mod Phys 2007, 79, 1267 – 90.Google Scholar
[58]
Chen T., Pourmand M., Feizpour A., Cushman B., Reinhard B.M., Tailoring plasmon coupling in self-assembled one-dimensional au nanoparticle chains through simultaneous control of size and gap separation, J Phys Chem Lett 2013, 4, 2147 – 52.Google Scholar
[59]
Mackowski D.W., Mishchenko M.I., Calculation of the T matrix and the scattering matrix for ensembles of spheres, J Opt Soc Am A 1996, 13, 2266 – 78.Google Scholar
[60]
Le Ru E.C., Etchegoin P.G., Rigorous justification of the |E|⁴ enhancement factor in surface enhanced raman spectroscopy, Chemical Physics Letters 2006, 423, 63 – 66.Google Scholar

nano Online

Physics, Chemistry and Materials Science at the Nanoscale

Search

Supplementary Materials

Open Access

Template-Guided Self-Assembly of Discrete Optoplasmonic Molecules and Extended Optoplasmonic Arrays

Abstract