Subscriber access provided by Otterbein University
Article
Near-Field Plasmonics of an Individual Dielectric Nanoparticle above a Metallic Substrate Tanya Hutter, Fu Min Huang, Stephen Elliot, and Sumeet Mahajan J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/jp400963f • Publication Date (Web): 27 Mar 2013 Downloaded from http://pubs.acs.org on March 31, 2013
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Physical Chemistry C is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 24
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
Near-Field Plasmonics of an Individual Dielectric Nanoparticle above a Metallic Substrate
Tanya Hutter1, Fu Min Huang2, Stephen R. Elliott1 and Sumeet Mahajan,2,3*†
1
2
Department of Chemistry, University of Cambridge, Cambridge CB2 1EW, UK
Department of Physics, Cavendish Laboratory, University of Cambridge CB3 0HE, UK 3
*
Institute of Life Sciences, University of Southampton, SO17 1BJ, UK
To whom correspondence should be addressed. E-mail:
[email protected] +44-
2380593591
ACS Paragon Plus Environment
1
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 24
KEYWORDS: nanoparticles, plasmons, surface-enhanced Raman spectroscopy (SERS), nearfield enhancement.
ABSTRACT
In this work, we simulate and discuss the local electric-field enhancement in a system of a dielectric nanoparticle placed very near to a metallic substrate. We use finite-element numerical simulations in order to understand the field-enhancement mechanism in this dielectric NP-onmirror system. Under appropriate excitation conditions, the gap between the particle and the substrate becomes a ‘hot-spot’, i.e. a region of intense electromagnetic field. In this work, we also show how the optical properties of the dielectric NP placed on a metallic substrate affect the plasmonic field enhancement in the nano-gap and characterize the confinement in the gap. Our study helps to understand and design systems with dielectric NPs on metallic substrates which can be equally as effective for SERS, fluorescence and non-linear phenomena as conventional all-metal plasmonic structures.
ACS Paragon Plus Environment
2
Page 3 of 24
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
INTRODUCTION Collective oscillations of free electrons under optical excitation give rise to surface plasmon resonances (SPR) in nanoscale metallic structures. Normally, SPR modes cannot be excited on flat metallic surfaces due to the momentum mismatch between a free-space photon and a surface propagating wave, unless coupled through an evanescent field with a high refractive-index prism in the commonly used Otto or Kretschmann configuration 1 . Another way of coupling light into flat films is to use metallic nanoparticles near or on top of them in the nanoparticle-on-mirror (NPOM) configuration 2-4. In this configuration, very strong fields are excited in the gap between the metal nanoparticle and the metallic surface. Metallic NPs placed above a dielectric substrate have also been studied
5-8
. In this latter system, strong confined fields are generated in the gap
between the NP and the substrate due to interaction of plasmons excited on the NP with the surface. It is found that the higher the permittivity of the substrate, the stronger is the field enhancement in the gap. However, for the inverse system of a dielectric NP on a metallic substrate, the field enhancement under the particle is caused by the localized potential well that traps the surface plasmons of the substrate, although the surface plasmons are excited via the NP and cannot be excited in its absence. In recent years, interest has developed in employing various dielectric particles, such as metaloxide NPs 9, due to their diverse properties, including catalytic activity, easy deposition techniques and low fabrication costs. These are also important for surface-science investigations of semiconductor/adsorbate systems, the chemisorption process and reaction mechanisms. There are a growing number of papers that study, theoretically and experimentally, dielectric particles for surface-enhanced Raman scattering (SERS) 10, metamaterials 11, near-field nanopatterning 12 and antennas
13
. Non-metallic materials are increasingly being employed in fundamental studies
ACS Paragon Plus Environment
3
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 24
and practical applications to overcome limitations of metallic nanostructures. Noble-metal nanostructured substrates are expensive to fabricate and have poorer biocompatibility than some metal-oxide materials, such as silica and titania. Nevertheless, high field enhancements for SERS have been observed from various non-metallic materials
14
. Moreover, it has been reported that
bigger dielectric particles with a certain combination of size and refractive index can induce a larger near-field enhancement factor than metallic NPs under certain circumstances
12
. In
addition, enhancement of Raman scattering has been proposed from molecules adsorbed on arrays of dielectric spheres 15. A large field enhancement in the gap between two silicon particles was theoretically investigated and has been predicted to be comparable to that between metallic nanoparticles
16
. Enhanced Raman scattering from arrays of vertical silicon nanowires has been
recently reported as well 17. Akimov et al. studied solar-cell performance and demonstrated that the use of dielectric nanoparticles can lead to similar, or even higher, enhancements compared to metal nanoparticles 18. Placing such dielectric nanoparticles near metallic surfaces yields further interesting plasmonic properties of the system. Recently, Letnes et. al. showed that dielectric nanoparticle dimers on silver substrates exhibit dipolar activity in the wavelength range near the SPR of silver 19
. The proximity of the dielectric NP to the substrate converts the incoming electromagnetic
plane wave to evanescent waves that excite surface plasmons at the silver substrate surface and confines them to the gap region. This localization of fields results in their substantial enhancement for frequencies near that of the surface plasmon resonance of silver. However, in general there is very little literature that describes the enhancement factors that can be achieved with dielectric NPs on metallic substrates. We have recently shown that the local field can be large enough to generate adequate field enhancements with metal-oxide nanoparticles on metallic substrates sufficient to allow enhanced Raman scattering from molecules positioned in the gap 9.
ACS Paragon Plus Environment
4
Page 5 of 24
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
Experiments and simulations showed that the enhancement of the electromagnetic field is caused by a modulation of the optical properties of the underlying plasmonic metallic surface by the high-refractive-index metal-oxide NPs, which act as scattering antennae and couple light into the gap region between them and the underlying surface. Metal-oxide NPs acted as photocatalytic mediators and could simultaneously be used to monitor the chemical reaction at their interface, demonstrating their advantage over analogous all-metal systems. In this study, we aim to understand the electric-field enhancement in a system consisting of a dielectric NP and a metallic substrate, separated by a nano-sized gap. This is different from the case of plasmon coupling between two metallic nanoparticles, or a metallic NP and the underlying surface, to generate ‘hot-spots’. More specifically, the goal is to understand how the real and imaginary parts of the refractive index of a dielectric material affect the field enhancement at the nano-junction. In contrast with studies that focus on the extinction (the farfield), we focus instead on the near-field enhancement and field confinement in the gap. This heterogeneous system of plasmonic and non-plasmonic materials can be used strongly to enhance optical phenomena which depend on the near-field, such as surface-enhanced Raman scattering, metal-enhanced fluorescence, nano-patterning, thin-film solar cells and non-linear optical processes.
NUMERICAL SIMULATIONS A three-dimensional model using COMSOL Multiphysics v4.3, a commercial finite-element mode solver, was constructed to enable parametric studies. The simulations were performed in two steps: the first step computes the electric field for the substrate only, when illuminated with a p-polarized plane-wave excitation at the upper boundary. The second step was solved for the
ACS Paragon Plus Environment
5
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 24
field due to the presence of the NP on the substrate using the output from the first step. Perfectly matched layers (PMLs) were used to absorb the scattered radiation in all directions (in order to eliminate reflections back to the model); their thickness was 50 nm. In order to reduce the computational time, symmetry planes were used and only one fourth of the model was solved for. A schematic illustration of the two steps of the model and additional simulation details can be found in Figure S1 in the supporting information. The model was solved with wavelength interval sizes of 2.5 nm. The optical constants of the materials were taken from ref.20. A sufficiently small spatial mesh (0.2 nm) has been used at the gap and at the surfaces of the particle compared to the bulk of the model, where the maximum mesh size for the air domain was set to 20 nm and 12 nm for the substrate. The radius of the NP was R = 10 nm. The field was calculated as a ratio of its value with the NP at a given point to the incident field, |E|/|E0|. The normalized maximum electric field was taken in the middle of the gap between the NP and the substrate. The results for the near-field discussed in this work are spectral plots of the maximum values of |E|/|E0| found along a horizontal line in the middle of the gap.
RESULTS AND DISCUSSION Local fields near the nanoparticle are important for SERS and near-field microscopy, and this information cannot be directly inferred from the scattering cross-section. Hence, we study the near-field enhancements for a dielectric nanoparticle on a metallic surface. Such a particle can excite and localize surface plasmons (SPs) of the metal and thus enhance the electric field in the
ACS Paragon Plus Environment
6
Page 7 of 24
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
gap. We investigate the effect of the substrate and the optical properties of the particle on this interaction and describe our results below. 3.1. Substrate influence In this section, we present our simulations for the near-field generated by a dielectric NP placed above different metallic substrates. In the simulations, we use a non-dispersive and nonabsorbing dielectric particle (n = 2, k = 0) with a radius (R) of 10 nm (diameter D of 20 nm), and with a gap of 1 nm between the substrate and the nanoparticle. We use normal-incidence illumination, with the polarization in the x-direction, as schematically illustrated in the inset of Figure 2a. Polarization in x-direction is investigated primarily since it is closest to experimental conditions under normal incidence conditions; however, the change of polarization is considered while we study the angle dependence in Section 3.3. Here, we examine the effect of the following metallic substrates: aluminium (Al), silver (Ag), gold (Au) and copper (Cu). In Figures 1a and 1b, the maximum electric field in the gap for various metallic substrates is shown. It can be seen that the peak for each metal is at a different wavelength and follows the same trend as the respective wavelengths for their surface plasmons. The peaks are at λ = 343 nm, 520 nm and 580 nm for Ag, Au and Cu respectively, corresponding to their SPR wavelengths. The peak for Al, as expected, is below the studied wavelength range (200-500 nm). Hence, this clearly demonstrates that it is the surface plasmon polaritons (SPPs) of the metal which primarily determines the plasmonics of the system, and the dielectric NP helps provides the appropriate momentum (kvector) to couple light into the metal.
ACS Paragon Plus Environment
7
The Journal of Physical Chemistry
8
Ex
7
Ag
|E|/|E0|
6
y
n=2 k=0
5
x
4
metallic
3 2
Al
1 0 200
250
300
350
400
450
500
700
750
Wavelength, nm
(a) 1.4
Au
1.2 1
|E|/|E0|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 24
Cu
0.8 0.6 0.4 0.2 0 450
(b)
500
550
600
650
Wavelength, nm
Figure 1. Maximum electric field |E|/|E0| in the gap for a dielectric NP above different metallic substrates: (a) Al and Ag, (b) Au and Cu.
The effect of the substrate supports the assertion that the high enhancement in the gap is due to the excitation of SPPs of the underlying metal by the dielectric NP on top. Hence, for the sake of brevity, subsequently in this paper we discuss the case of a dielectric nanoparticle on a silver substrate only. The spatial field distribution |E|/|E0| at the peak (λmax = 343 nm) for a silver substrate is shown in Figure 2a. The |E|/|E0| along a 30 nm long horizontal line in the middle of the gap at 343 nm is shown in Figure 2b. To understand the field contribution of different polarizations, the spatial distributions in the middle of the gap (plan view; looking from top through the NP) of |E|/|E0|, Ex/|E0|, Ey/|E0| and Ez/|E0| are shown in Figure 2c-f at the same
ACS Paragon Plus Environment
8
Page 9 of 24
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
wavelength. It can be seen that the x-polarized field is very weak and has a maximum just in the centre below the NP due to dipolar resonances in the dielectric particle, inducing an image dipole in the metal in the gap. For y- and z-polarizations, there is a local minimum in the electric field directly in the middle of the gap below the NP due to the shifting of the electrons in the metallic substrate under the excitation field.
Figure 2. (a) Spatial field distribution |E|/|E0| for a dielectric NP (n = 2, k = 0, R = 10 nm) placed 1 nm above a silver substrate, and (b) the corresponding electric field |E|/|E0| profile along a horizontal line in the middle of the gap. The spatial variation in the middle of the gap of the electric field of (c) |E|/|E0|, (d) Ex/|E0|, (e) Ey/|E0| and (f) Ez/|E0| at 343 nm for a dielectric NP above a silver substrate.
ACS Paragon Plus Environment
9
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 24
3.2. Effect of gap distance The gap distance between surfaces can play a significant role in near-field interactions and this was studied for our case of a dielectric NP on a smooth silver substrate. Such a gap can be created by a molecular spacer layer
2
or a transparent dielectric layer 4, and thus can be
manipulated. The gap height was varied from 0.6 to 10 nm, and the corresponding maximum electric-field spectra in the gap are shown in Figure 3 for a NP of radius 10 nm. In the inset, the ratio |E|/|E0| is plotted as a function of the ratio of the gap to the NP radius. As expected, the electric field in the gap increases as the gap decreases due to a stronger interaction. At a gap of 10 nm, there is hardly any field enhancement in the gap due to the weak interaction with the metallic substrate. As the gap decreases, the interaction becomes stronger and an additional shoulder appears. For a 0.8 nm gap, two peaks are observed as the shoulder becomes bigger. At the smallest gap of 0.6 nm, the additional peak becomes dominant. This peak is at a higher wavelength, and thus a smaller energy. This additional peak is similarly observed where the real and imaginary parts of the refractive index of the dielectric NP are increased (section 3.4). We postulate that the first peak is merely due to the SP excitation of the metallic substrate. The second peak is a hybrid mode due to the interaction of the substrate’s SP with the NP. For the inverse system of a metallic NP on a dielectric substrate, the spectral shifts resulting from the presence of an adjacent dielectric are viewed as an interaction of the nanoparticle with its image with a strength determined by the nanoparticle-substrate separation, substrate permittivity, and polarization of incident light 6. Thus an alternative view considers the coupling between the ‘image’ generated in the dielectric NP as a result of the confined SP on the metallic substrate surface, which the particle helps excite, and this gives rise to this peak. In our case, the second peak at a higher wavelength becomes dominant (for a small gap and high n, k values of the NP) due to the fact that less energy is required to excite this hybrid mode
ACS Paragon Plus Environment
10
when the interaction between the substrate and NP is strong. This explanation is also consistent with the observation that this mode, which appears as a shoulder at a higher wavelength, disappears for a higher refractive index of the surrounding medium because the interaction of the NP and the substrate is screened, as described further in section 3.2 (Fig. 4). It is important to note that, in contrast with similar systems of a metallic NP on metallic substrates, where the peak maximum is significantly red-shifted as the gap decreases 21, here this effect is very small (only a 7 nm red-shift is seen as the gap decreases from 0.6 to 10 nm). This supports the fact that the interaction is determined by the metal, and the role of the dielectric particle is primarily to help couple light into the metal for generation of the SPPs. 11
10
Gap = 0.6 nm
|E|/|E0|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
10 9
|E|/|E0|
Page 11 of 24
8
0.8
7
1
5
0 0.05
0.25
6 1.5
5
0.45
Gap/R
2
4
3
3
5
2
10
1 330
340
350
360
370
380
390
400
Wavelength, nm
Figure 3. Plot of the maximum electric-field |E|/|E0| spectra in the gap for different gaps between a dielectric NP and a silver substrate. The real and imaginary parts of the refractive index of the NP are 2 and 0, respectively. The radius of the NP was fixed at 10 nm. The wavelength step size was 1 nm between 335 and 350 nm, and 2.5 nm below 335 nm and above 350 nm. Inset: electric field in the gap at λmax versus the ratio of the gap size to the NP radius, R.
We compare this system of dielectric NP above metallic substrate to the system of metallic NP above dielectric substrate as reported in Ref. 5. In both cases, as the gap between the NP and the substrate increases the near-field in the gap dramatically decreases. For metallic NP above
ACS Paragon Plus Environment
11
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 12 of 24
dielectric substrate only one peak is observed in the near-field spectrum corresponding to the SPR of the metallic NP. However, for a dielectric NP above metallic substrate, at smaller gaps, two peaks are observed in the near-field spectrum; one due to the SPP of the metallic substrate and the other is a hybrid mode as a result of the interaction between the dielectric NP and the metallic substrate. 3.2. Effect of the refractive index of the surrounding medium In the previous section, we assumed that the system is in vacuum. Here, we study the effect of the surrounding medium on the near-field of a dielectric NP (n = 2, k = 0) above a smooth silver substrate. The same normal-incidence illumination, with its polarization in the x-direction was used. The refractive index of the surrounding medium, nm, was varied from 1 to 2. The plot of the maximum electric field in the gap is shown in Figure 4 for a NP of radius 10 nm and with a 1 nm gap. In the inset, the value |E|/|E0| is plotted as a function of nm. In Figure 4b, the values of λmax and the corresponding electric field in the gap are plotted versus the refractive index of the surrounding medium. The value of λmax increases as nm increases, whereas the electric field in the gap reaches a maximum value of 10.3 for nm = 1.2, and then decreases. When nm = 2, the same as the refractive index of the dielectric NP, there is no peak in the spectrum, and beyond this value of nm the system can no longer be considered as a particle on a surface but becomes the inverse problem of a resonance cavity in a medium. Nevertheless, our finding on the effect of surrounding medium refractive index on the plasmonics of the dielectric on a metal system is of relevance to studies which are often carried out in aqueous conditions (n =1.33), which can lead to the observation of higher enhancements compared to air, and at a different wavelength. We hypothesize that the observed optimal nm is the result of a balance between the coupling of light into the particle and the substrate via the particle. As nm increases, light can be more efficiently coupled into the substrate; however, when the difference between the refractive index of the
ACS Paragon Plus Environment
12
medium and the particle becomes smaller, hardly any light is trapped by the NP itself, leading to less coupling into the substrate for the generation of its SPP. The red-shift of the SPP is a result of its scaling due to an increase in nm. 9 1.4
8
1.2 1.0
7
|E|/|E0|
1.6
6 5 4
1.8
3 2 2.0
1 0 330
340
350
360
370
380
390
400
Wavelength, nm
(a) 390
10
380
8
370 6 360
|E|/|E0|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
λmax, nm
Page 13 of 24
4
350 340
2 1
(b)
1.2
1.4
1.6
1.8
nm
Figure 4. (a) Plot of the maximum electric field |E|/|E0| spectra in the gap between a dielectric NP and the substrate for different values of the refractive index of the surrounding medium, nm. The real and imaginary parts of the refractive index of the NP are taken to be 2 and 0, respectively. The radius of the NP was fixed at 10 nm and the gap was 1 nm. (b) Plot of λmax and the corresponding electric field in the gap versus nm. 3.3 Dependence on the excitation angle The spectral plots of the maximum electric field in the gap are shown in Figure 5a for a dielectric particle (n = 2, k = 0) with a radius of 10 nm placed above Ag substrate with a gap of 1 nm between the substrate and the nanoparticle. The NP is illuminated with a p-polarized plane-wave
ACS Paragon Plus Environment
13
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 14 of 24
excitation at the angles of 0, 25, 50 and 75º with respect to the normal to the surface, thus electric field is polarized in x- and z- directions. At normal incidence (0º), the polarization is only in x-direction, and the maximum value of |E|/|E0| is ~6.5. As the angle increases, there is additional polarization component in z-direction. At the angle of 25º, the values of the electric field in the gap increase indicating better coupling efficiency and better ability to generate SPPs of the Ag substrate. Two peaks are present at this angle, one at ~343 nm (|E|/|E0| ~ 7.7) and the other at ~350 nm (|E|/|E0| ~ 4.5). The spatial field distributions at the two peaks are shown in Figure 5b and 5c. Also the field profiles along a horizontal line in the middle of the gap at the peaks for the studied angles are shown in Figure 5d. It can be seen that for an angle of 25º, at the first peak there are two lobes in the gap similar to the 0º excitation as shown in Figure 2. At the second peak, the profile is different showing only one lobe indicating stronger interaction between the NP and the substrate in the z-direction. As the angle is increased further to 50º, there is one peak dominated by the interaction in zdirection, and the electric field enhancement in the gap increases to 14.4. For the angle of 75º, this decreases to a value of 8.4. The increase and then the decrease of the field enhancement in the gap indicates that there is an optimal angle. This observation is similar to that observed in a system of metallic NP on metallic substrate, for which the electric field is highest for an angle of 60º 4. This arises due to a superposition of the EM field between the light ray directly illuminating a point and that reflected from the surface. For p-polarized light the phase difference between the field components of the incident and reflected rays changes by π-radians at ~60° (for gold) producing this maximum. For two metallic NPs however, the highest enhancement is obtained for a polarization parallel to the dimer axis as a result of a strong coupling between the collective surface plasmon modes of the
ACS Paragon Plus Environment
14
Page 15 of 24
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
two metallic particles and the incident field and lowest enhancement is obtained when it is perpendicular to the dimer axis 22.
Figure 5. (a) Plot of the maximum electric-field |E|/|E0| spectra in the gap between a dielectric NP and a silver substrate for different excitation angles. Spatial field distribution |E|/|E0| for an angle of 25º at wavelengths of (b) 342.5 nm and (c) 349.5 nm. (d) Electric field profiles along a horizontal line in the middle of the gap at the maxima for the different angles.
3.4 Effect of the real and imaginary parts of the refractive index of the dielectric NP
ACS Paragon Plus Environment
15
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 16 of 24
The interaction between the nanoparticle and the substrate, although dependent on the gap distance and surrounding medium, is likely to be affected by the value of the permittivity of the NP. Hence, we investigated this in detail by varying the real and imaginary parts of the refractive index of the NP independently in simulations.
3.4.1 Effect of n for non-absorbing materials In effect, spherical dielectric NPs act as Mie scatterers, coupling light into surface plasmon modes of the metallic surface which get localized in the gap. However, the strength of these hybrid modes critically depends on the ‘effective’ size (∝ n, R) and absorption by the nanoparticles, determined by the real and imaginary parts of the refractive index; thus, as a result of the dependence on these parameters, the gap modes can be modulated. Below, we describe the various characteristics of a dielectric NP on a metallic surface and their dependence on the optical constants, in order further to understand the mechanism of electric-field enhancement. We only consider the near-field, since our prime interest is in the enhancement of near-field phenomena, such as SERS. A single dielectric NP (R = 10 nm) by itself in air, with different n values of 2 to 5, does not exhibit any distinct resonances in the studied wavelength range. For a dielectric NP placed 1 nm above a metallic substrate, the plots of the maximum intensity along a horizontal line in the middle of the gap versus wavelength for various values of n are shown in Figure 6a. The peak in the spectra is attributed to the substrate SPP-mode. As discussed previously, this substrate SPPmode originates from the excitation of surface plasmon polaritons on the substrate surface. As n increases, the electric field in the gap increases and the peak is red-shifted. A refractive-index value higher than that of the surrounding medium ( > 1) gives a larger wave-vector of light inside the NP than outside; thus, it works as a potential well for the electromagnetic field
ACS Paragon Plus Environment
15
, which
16
Page 17 of 24
becomes stronger for higher n. In Figure 6b, the values of λmax and the corresponding electric field in the gap are plotted versus the real part of the refractive index of the nanoparticle. Both increase as n of the particle increases. In order to determine the field confinement in the gap, we have plotted the FWHM at λmax of the value |E|/|E0| in the gap in Figure 6c. The field distribution shows two lobes, and the FWHM was calculated from the edge of one lobe to the edge of the other at half-maximum intensity. As n increases, the field becomes increasingly more confined, reaching a value of about 14 nm at n = 5. Ex
25
5
|E|/|E0|
20
4
15
y
3
n k=0 x silver
2.5
10
2 5
1.5
0 330
340
350
360
370
380
390
400
Wavelength, nm
(a) 25
16
15 14
10 12
5 0
10
1.5
(b)
FWHM, nm
18
FWHM
20
|E|/|E0|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
2
2.5
3
3.5
4
4.5
5
n of particle
Figure 6. (a) Maximum electric field |E|/|E0| spectra in the gap for different values of the real part of the refractive index of the NP. The radius of the NP was fixed at 10 nm and the gap was 1 nm. (b) Electric field in the gap |E|/|E0| and the corresponding FWHM of the electric field as a function of n.
ACS Paragon Plus Environment
17
The Journal of Physical Chemistry
3.4.2 Effect of k We use the same system, with the same dimensions and field excitation as before, to study the effect of the variation of k of the NP placed over a silver substrate. Plots of the electric field along a horizontal line in the gap for various n - k combinations are shown in Figure 7. When n = 2 (|E|/|E0| ~ 6.8), increasing k from 0 to 0.5 causes a reduction in the electric field in the gap due to absorption of the light in the absorbing dielectric NP. Further increasing the value of k to 2 causes an increase in the electric field to |E|/|E0| ~ 12. Increasing k further to 5, causes a dramatic increase in the electric field in the gap, reaching a value |E|/|E0| = 27. The strong field enhancement in this case is probably due to the more metallic nature of the NP (n < k). When both n and k are high (n = k =5), the electric field is also high; this can be characteristic of semiconductors, such as silicon, that have high values of both n and k in this wavelength range.
30
n=2,k=5 25
|E|/|E0|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 18 of 24
n=5,k=5
20 15
n=2,k=2 10
n=2,k=0 5
n=2,k=0.5
0 330
340
350
360
370
380
390
400
Wavelength, nm
Figure 7. (a) Plot of the electric field |E|/|E0| spectra in the gap for different values of the real and imaginary parts of the refractive index of the NP. The radius of the NP was 10 nm with a 1 nm gap above silver substrate.
ACS Paragon Plus Environment
18
Page 19 of 24
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
Comparing the current system with that reported in Ref. 5 the electric-field in the gap significantly increases as n increases in both cases. The effect of the imaginary part of the refractive index on the near field is similar, for small k values there is initially a decrease in the electric field due to light absorption, and then increase as k becomes bigger due to the higher reflectivity of the material.
3.5 Real examples of dielectric NPs In this section, we show the results of simulations for NPs (R = 10 nm) made from real materials placed 1 nm above a silver substrate. It should be noted that we have used only those materials that have a Bohr exciton radius smaller than 10 nm so that none of our particles (R = 10 nm) behave as quantum dots. Table I summarizes the values of λmax and the corresponding electric field in the gap, |E|/|E0|, together with the real and imaginary parts of the refractive index at λmax. The dependency of the electric field in the gap on the real part of the refractive index for these materials shows the same trend as in Figure 7. The spectral plots of the electric field in the gap of real materials can be found in Figure S2 in the supporting information. Silica, having the smallest n value, shows the smallest SERS enhancement of (|E|/|E0|)4 = 135 at 340 nm, and silicon, with n = 5.47 at λmax, shows the highest enhancement value ~ 2*105 at 350 nm Although this is smaller compared to that of an all-metallic system of an Ag NP (R = 10 nm) placed 1 nm above an Ag substrate, which yields (|E|/|E0|)4 = 2*109 at a wavelength of 388 nm (Figure S3 in the supporting information), the enhancement value obtained with silicon is still significantly high. We plot the electric field in the gap generated with NPs made of real dielectric materials with dielectric NPs made of non-absorbing materials (k = 0) in Fig. 8. For small n values, when k = 0, the two curves nearly aligned on one line. As some real materials have non-zero k values, the
ACS Paragon Plus Environment
19
The Journal of Physical Chemistry
electric fields in the gap are lower compared to those of non-absorbing dielectric materials, illustrating the effect of the imaginary part of the refractive index.
Table 1. λmax, |E|/|E0|in the gap, real and imaginary parts of the refractive index 20 at λmax. Material
λmax, nm
|E|/|E0|
n @ λmax
k @ λmax
SiO2
340
3.41
1.48
0
Y2O3
342.5
7.01
2.02
0
Si3N4
342.5
7.54
2.11
0
ZnS
347.5
11.45
2.82
0.15
SiC
347.5
11.77
2.85
0.13
TiO2
350
17.94
4.44
0.64
GaP
350
19.93
5.19
1.02
Si
350
21.23
5.47
2.99
25
Non-absorbing NPs
20
|E|/|E0|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 20 of 24
15
Real NPs
10 5 0 1.5
2.5
3.5
4.5
5.5
n of the particle
Figure 8. (a) Plot of the electric field |E|/|E0| spectra in the gap for different material types as a function of the real part of the refractive index of the NP. The upper curve (squares) corresponds to NPs made from non-absorbing materials (k = 0) and the lower curve (diamonds) corresponds to real materials, as listed in Table I. The radius of the NP was fixed at 10 nm with a 1 nm gap.
ACS Paragon Plus Environment
20
Page 21 of 24
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
CONCLUSIONS We have simulated the electric fields produced by dielectric NPs above a metallic substrate. The field enhancement in the gap between a dielectric NP and the metallic substrate originates from surface plasmon polaritons of the metal substrate. We have systematically studied the dependence of the field enhancement in the nano-gap on metal substrates, gap distance, excitation angle, surrounding medium and the optical properties of the NP. These results provide an insight into the near-field interactions and are useful for understanding the underlying physics of a nano-junction between a dielectric particle and a metallic substrate. This can be extremely valuable for selecting and designing nanoparticles to be most appropriate for obtaining near-field enhancements in SERS, fluorescence and other near-field phenomena in different applications, such as catalysis, photovoltaics and sensing.
ASSOCIATED CONTENT Supporting Information Available. Model description used in the numerical simulation. Spectral plots of the electric field in the gap for different NP material. Spectral plot of electric field in the gap for a silver NP above a silver substrate. This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Author * E-mail:
[email protected] ACS Paragon Plus Environment
21
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 22 of 24
Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.
Funding Sources We acknowledge EPSRC (EP/H028757/1) for funding support. T. Hutter is grateful for a Geoffrey Moorhouse Gibson Studentship in Chemistry from Trinity College, University of Cambridge.
Notes The authors declare no competing financial interest. REFERENCES 1.
Sambles, J. R.; Bradbery, G. W.; Yang, F. Z., Optical-Excitation of Surface-Plasmons - an Introduction.
Contemporary Physics 1991, 32. 2.
Hill, R. T.; Mock, J. J.; Urzhumov, Y.; Sebba, D. S.; Oldenburg, S. J.; Chen, S.-Y.; Lazarides, A. A.;
Chilkoti, A.; Smith, D. R., Leveraging Nanoscale Plasmonic Modes to Achieve Reproducible Enhancement of Light. Nano Letters 2010, 10, 4150-4154. 3.
Nordlander, P.; Prodan, E., Plasmon hybridization in nanoparticles near metallic surfaces. Nano Letters
2004, 4, 2209-2213. 4.
Mubeen, S.; Zhang, S.; Kim, N.; Lee, S.; Kraemer, S.; Xu, H.; Moskovits, M., Plasmonic Properties of
Gold Nanoparticles Separated from a Gold Mirror by an Ultrathin Oxide. Nano Letters 2012, 12, 2088-2094. 5.
Hutter, T.; Elliott, S. R.; Mahajan, S., Interaction of metallic nanoparticles with dielectric substrates: effect
of optical constants. Nanotechnology 2013, 24, 035201. 6.
Knight, M. W.; Wu, Y.; Lassiter, J. B.; Nordlander, P.; Halas, N. J., Substrates Matter: Influence of an
Adjacent Dielectric on an Individual Plasmonic Nanoparticle. Nano Letters 2009, 9, 2188-2192. 7.
Chen, H.; Shao, L.; Ming, T.; Woo, K. C.; Man, Y. C.; Wang, J.; Lin, H.-Q., Observation of the Fano
Resonance in Gold Nanorods Supported on High-Dielectric-Constant Substrates. Acs Nano 2011, 5, 6754-6763. 8.
Spinelli, P.; van Lare, C.; Verhagen, E.; Polman, A., Controlling Fano lineshapes in plasmon-mediated light
coupling into a substrate. Optics Express 2011, 19, A303-A311.
ACS Paragon Plus Environment
22
Page 23 of 24
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
9.
Li, L.; Hutter, T.; Finnemore, A. S.; Huang, F. M.; Baumberg, J. J.; Elliott, S. R.; Steiner, U.; Mahajan, S.,
Metal Oxide Nanoparticle Mediated Enhanced Raman Scattering and Its Use in Direct Monitoring of Interfacial Chemical Reactions. Nano Letters 2012. 10.
Terakawa, M.; Tanaka, Y.; Obara, G.; Sakano, T.; Obara, M., Randomly-grown high-dielectric-constant
ZnO nanorods for near-field enhanced Raman scattering. Applied Physics A: Materials Science & Processing 2011, 102, 661-665. 11.
Ahmadi, A.; Mosallaei, H., Physical configuration and performance modeling of all-dielectric
metamaterials. Physical Review B 2008, 77, 045104. 12.
Tanaka, Y.; Obara, M., Comparison of Resonant Plasmon Polaritons with Mie Scattering for Laser-Induced
Near-Field Nanopatterning: Metallic Particle vs Dielectric Particle. Japanese Journal of Applied Physics 2009, 48. 13.
Schuller, J. A.; Brongersma, M. L., General properties of dielectric optical antennas. Optics Express 2009,
17, 24084-24095. 14.
Wang, X.; Shi, W.; She, G.; Mu, L., Surface-Enhanced Raman Scattering (SERS) on transition metal and
semiconductor nanostructures. Physical Chemistry Chemical Physics 2012, 14, 5891-5901. 15.
Inoue, M.; Ohtaka, K., Enhanced Raman-Scattering by Two-Dimensional Array of Polarizable Spheres.
Journal of the Physical Society of Japan 1983, 52, 1457-1468. 16.
Sigalas, M. M.; Fattal, D. A.; Williams, R. S.; Wang, S. Y.; Beausoleil, R. G., Electric field enhancement
between two Si microdisks. Optics Express 2007, 15, 14711-14716. 17.
Khorasaninejad, M.; Walia, J.; Saini, S. S., Enhanced first-order Raman scattering from arrays of vertical
silicon nanowires. Nanotechnology 2012, 23, 275706. 18.
Akimov, Y. A.; Koh, W. S.; Sian, S. Y.; Ren, S., Nanoparticle-enhanced thin film solar cells: Metallic or
dielectric nanoparticles? Applied Physics Letters 2010, 96, 073111-3. 19.
Letnes, P. A.; Simonsen, I.; Mills, D. L., Substrate influence on the plasmonic response of clusters of
spherical nanoparticles. Physical Review B 2011, 83, 075426. 20.
Ni, X.; Liu, Z.; Kildishev, A. V. PhotonicsDB: Optical Constants.
http://nanohub.org/resources/PhotonicsDB. 21.
Okamoto, T.; Yamaguchi, I., Optical Absorption Study of the Surface Plasmon Resonance in Gold
Nanoparticles Immobilized onto a Gold Substrate by Self-Assembly Technique. The Journal of Physical Chemistry B 2003, 107, 10321-10324. 22.
Xu, H.; Käll, M., Polarization-Dependent Surface-Enhanced Raman Spectroscopy of Isolated Silver
Nanoaggregates. ChemPhysChem 2003, 4, 1001-1005.
ACS Paragon Plus Environment
23
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 24 of 24
Table of contents image:
ACS Paragon Plus Environment
24