Light-Emitting Plexciton: Exploiting PlasmonâExciton Interaction in the

Subscriber access provided by Kaohsiung Medical University

Article

Light-Emitting Plexciton: Exploiting Plasmon-Exciton Interaction in the Intermediate Coupling Regime Jiawei Sun, Huatian Hu, Di Zheng, Daxiao Zhang, Qian Deng, Shunping Zhang, and Hongxing Xu ACS Nano, Just Accepted Manuscript • DOI: 10.1021/acsnano.8b05880 • Publication Date (Web): 17 Sep 2018 Downloaded from http://pubs.acs.org on September 17, 2018

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 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 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.

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 30 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

ACS Nano

Light-Emitting Plexciton: Exploiting Plasmon-Exciton Interaction in the Intermediate Coupling Regime Jiawei Sun,†,# Huatian Hu,†,# Di Zheng,‡ Daxiao Zhang,‡ Qian Deng,‡ Shunping Zhang,*, ‡

and Hongxing Xu,*, †, ‡

†

The Institute for Advanced Studies, Wuhan University, Wuhan 430072, China.

‡

School of Physics and Technology, Center for Nanoscience and Nanotechnology, and

Key Laboratory of Artificial Micro- and Nano-structures of Ministry of Education, Wuhan University, Wuhan 430072, China. Corresponding Author *E-mail: [email protected]. Phone: +8627 68752219. *E-mail: [email protected]. Phone: +8627 68752253.

ABSTRACT The interaction between plasmons in metal nanostructures and excitons in layered materials attracts recent interests due to its fascinating properties inherited from the two constituents, e.g., the high tunability on its spectral or spatial properties from the plasmonic component, and the large optical nonlinearity or light emitting properties from the excitonic counterpart. Here, we demonstrate the light-emitting plexcitons from the coupling between the neutral excitons in monolayer WSe2 and highly-confined nanocavity plasmons in nanocube-over-mirror system. We observe, simultaneously, an anti-crossing dispersion curve of the hybrid system in the dark-field scattering spectrum and a 1700 times enhancement in the photoluminescence. We attribute the large photoluminescence enhancement to the increased local density of states by both the plasmonic and excitonic constituents in the intermediate coupling regime. What’s more, increasing the confinement of the hybrid systems is achieved by shrinking down the size of hot spot within the gap between the nanocube and the metal film. Numerical calculations reproduce the experimental observations and provide the effective number of excitons taking part in the interaction. This highly compact system provides a room temperature testing platform for quantum cavity electromagnetics at the deep subwavelength scale. 1

ACS Paragon Plus Environment

ACS Nano 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

KEYWORDS: plasmonics, plexciton, enhanced fluorescence, nanocavity, transition metal dichalcogenides

Light-matter interation at nanometer scale is the central issue in nanophotonics. Typically, a quantum emitter in free space couples weakly to a photon due to the large mismatch in their length scale. This mismatch can be greatly alleviated by introducing a metalic nanoparticle as a nanocavity that localizes the light.1,2 This is a nanoscopic scenario of cavity quantum electrodynanics (cQED) that exploits the deep subwavelength mode volume of the cavity resonances, i.e., the localized surface plasmons. In the weak coupling regime, the spontaneous emission rate of the emitter can be accelerated by a factor called Purcell factor (in the single mode approximation), leading to a phenomenon known as plasmon-enhanced fluorescence.3-5 In the strong coupling regime, when the rate of the energy exchange between the two components is larger than the respective dissipation rates, hybrid light-matter states appear with a characteristic energy difference named as Rabi splitting.6 Owing to their part-light part-matter nature, the group velocity of the hybrid states can be significantly sped up due to the light component, and at the same time, the system possesses a strong nonlinearity inherited from the emitter counterpart.7 In the intermediate coupling regime, however, the transmission spectrum of the hybrid system features a Fano interference.8,9 The fluorescence brightness of hybrid systems is predicted to reach its maximum just in this regime.2 Transition metal dichalcogenides (TMDs) arouse great interests in the researches of light-matter interaction owing to their fascinating optical properties and flexible features for integration.10-15 When the TMDs are thinned down to monolayer,11 they become direct band-gap materials. Because of the reduced screening, giant excitonic effect dominates the band-gap transition in monolayer TMDs, possessing a huge transition dipole moment µ that facilitates the light-matter interaction.16,17 Therefore, monolayer TMDs represent ideal platforms for the formation of exciton polaritons that survive at room temperature. These hybrid states have been demonstrated by embedding the excitonic materials into Fabry-Pérot type cavities.18-20 Another elegant way of constructing hybrid system is simply 2


Page 2 of 30

Page 3 of 30 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

ACS Nano

depositing a metallic nanoparticle on top of a TMD monolayer. In contrast to photonic cavities, plasmonic nanocavities are bad cavities (low quality factor) but harbor high mode confinement since they store the optical energy partly in the form of the collective motions of the electrons. Strong coupling between TMDs excitons and plasmons has been realized in periodic metallic nanoparticle arrays21,22 and in individual metallic nanoparticles,23-26 at room temperature. This can be applied to imitate some conventional cQED experiments onto a highly compact solid state test bed.27 The tunability of plasmons by the size, shape, and interparticle separation of the nanostructures provides a convenient route to manipulate the coherent hybrid states via their plasmonic components. For example, reducing the spatial extension of the hybrid states, which enhances the nonlinearity of the system, can be accomplished by increasing the confinement of plasmons. Compared with individual nanoparticles, plasmons in nanogaps between adjacent metallic nanostructures have higher mode concentration in the gap,28,29 which have been widely used to amplify the light-matter interaction in molecular systems.30-32 Integration of TMDs with plasmonic nanogaps have been illustrated to show a pronounced enhancement of the photoluminescence (PL) in monolayer TMDs,33,34 indicating that the hybrid system is in the weak coupling regime. Rabi splitting in the scattering spectrum was only realized in multi-layer TMDs inserted in the gap.35 However, it sacrifices the PL efficiency due to the in-direct band-gap nature of multi-layer TMDs. Here, we demonstrate a highly luminescent plexcitonic system by combining the neutral (X0) excitons in monolayer WSe2 with the nanocavity plasmons in a nanocube-over-mirror (NCOM) system. Anti-crossing dispersion curve reveals a Rabi splitting of 36.7 meV, which is comparable with the excitonic linewidth but smaller than the plasmon linewidth. Meanwhile, the off-resonance excitated PL spectra show 1700 times enhancement for the monolayer WSe2 within the hotspot. Theoretical calculations reveal an increased local density of states (LDOS) due to both the confinement from the nanocavity plasmon and the large excitonic effect in monolayer WSe2. Noting that such plexciton enhanced fluorescence only occurs in the intermediate coupling regime, and it will evolve into plasmon enhanced fluorescence (plexciton lumininescence) in the weak (strong) coupling regime, sacrificing the giant light-emitting property. The confinement of the plexcitons is increased by 3


ACS Nano 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

shrinking down the size of hot spot in the NCOM system through reducing the size of nanocubes and closing the nanogap. The number of excitons involved in the plexcitonic states is decreased for a nanocube with a shorter edge length.

RESULTS AND DISCUSSION Our hybrid NCOM sample consists of a monolayer WSe2 inserted into the gap between a silver nanocube and a gold film (Figure 1a). The ultra-smooth gold film was prepared following the standard template stripping method,36 ending up with a root-mean-square roughness of 0.34 nm. An alumina layer was deposited onto the gold film by atomic layer deposition (ALD), and then the monolayer WSe2 was mechanically exfoliated onto it. Characterization of monolayer WSe2 was shown in SI Figure S1. Silver nanocubes were prepared by polyol synthesis using polyvinylpyrrolidone (PVP) as the capping reagent, and they were covered by a PVP layer of 2 – 3 nm (details see Methods). Finally, silver nanocubes were dropped and dried onto the monolayer WSe2, constructing the hybrid system as shown in the inset of Figure 1a. The alumina layer and the PVP cladding are expected to prevent the direct contact between the WSe2 and the metal, in order to eliminate possible charge transfer between metals and WSe2.37,38 The design and composition of the NCOM systems enable exciton-plasmon coupling. At room temperature, the band-gap transition of monolayer WSe2 is dominated by the intense X0 exciton peak at 745 nm (details see SI Figure S1). The transition dipole moment is well in-plane (xy- plane)33,39 and this interband transition couples with circular polarized light according to the valley selection rule.12 The coupling strength between a single exciton and a plasmon is proportional to the inner product of the transition dipole moment of the exciton and the in-plane component of the electric field of the plasmon. Since the monolayer WSe2 is inserted between the gold film and silver nanocube, the X0 excitons will experience the most intense plasmon near-field in the nanogap, satisfying the requirement of spatial overlap for optimizing the plasmon-exciton coupling. The optical properties of the NCOM system were also calculated using full wave finite element method (COMSOL Multiphysics, V5.2a). The permittivities of gold and silver followed the 4


Page 4 of 30

Page 5 of 30 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

ACS Nano

experimental results by Johnson and Christy.40 The refractive index of alumina was taken as 1.5. The thickness of monolayer WSe2 was taken as 0.7 nm. The in-plane permittivity of monolayer WSe2 followed the Lorentz model ε ip ( E ) = ε ∞ − f

E 02 E − E 02 + iE Γ 0 2

(1)

where ε∞ = 17.5 is the high frequency contribution to the permittivity, f is the reduced oscillator strength of the exciton. E0 and Γ0 are taken from the absorption spectra of WSe2 on silica substrate (details see our recent work).23 The out-of-plane permittivity εout of WSe2 was taken as 2.9.41 The calculated scattering spectrum of a bare NCOM system (without excitonic transitions) is shown in Figure 1b. The edge length of the nanocube was set to be 84.6 nm, with the sharp edges and corners rounded by a 17 nm-curverture to match with our nanocube sample. For a bare NCOM system, the osillator strength f of the X0 exciton was set to zero so that the WSe2 layer was solely treated as a dielectric medium with the in-plane permittivity of 17.5 and the out-of-plane permittivity of 2.9. A NCOM system can be viewed as a truncated metal-insulator-metal cavity with effective reflections at the bottom edges of the nanocube.42 The lowest order mode, a round trip accumulating 2π phase shift of the gap surface plasmon polaritons, is also called magnetic plasmon since it enhances the local magnetic field at the center of the system.43 Interestingly, the emission of this mode is shown to be highly directional along the film normal, making it addressable by free space radiation.30 Under the illumination of a normal incident x-polarized plane wave (as shown in Figure 1b inset), the magnetic plasmon mode (denoted by 'M') is excited at the wavelength close to the X0 exciton in WSe2. According to Ref.,44 the scattered field of a plasmonic resonator can map the distribution of the plasmon. Figures 1c-e show the scattered electric field of x, y, z- component of the M mode normalized to the incident electric field E0. The maximum enhancement factors (electric field amplitude) of x, y, z- component of electric field are 12.1, 7.1 and 58.5 respectively. The scattering spectra of the NCOM system were measured using a home-built oblique illumination dark-field scattering spectrometer. We found that the scattering peak of the M mode always deviates from a pure Lorentz function for non-polarized light excitation. In some samples, the spectra displayed closely-spaced double peaks, but in most samples the spectra showed a prominent peak 5


ACS Nano 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

superposed on a broader peak. These features agree with previous studies on the NCOM configuration using dark field spectroscopy,42 but have not received much attention. To interpret these features correctly, in the following, we will show that these line profiles come from two different aspects: (i) the splitting of the degenerate M modes into two modes due to the imperfect cubic shape; and (ii) the excitation of the bonding dipole (BD) mode contributing to the broad peak around the M mode. Before investigating the hybrid system, we first characterized a bare NCOM system without the WSe2 layer. To simplify the mode identification, we added a polarizer into the illumination path to selectively excite a certain plasmon mode and rotated the sample with respect to the incident light. As shown in SI Figure S3, when the edge of the nanocube is aligned along the direction of the incident light (the projection of incident direction on the xy-plane, more precisely), the peak is separated into two peaks coresponding respectively to the P- or S-polarization. Correlated scanning electron microscope (SEM) image shows that this is because the length and width of the nanocube is not equal – the nanocube is not a perfect quadrate when projecting onto the xy-plane. Therefore, the degenerate M mode splits into two modes that can be excited separately under P- and S-polarized light excitation. Being aware of this, all the dark-field scattering measurements in the following were taken under polarized light excitation and the silver nanocubes were orientated so that only one of the M mode is excited. The experimental and numerical results of the scattering spectra of a representative hybrid NCOM system under P- or S-polarization excitation are shown in Figure 2. The system consisted of a 75.4 nm nanocube on a monolayer WSe2, which was separated from the gold film by a 1.6 nm alumina layer. An additional alumina layer of 8 nm was deposited on top of the sample to prevent the silver nanocube from oxidization or sulfuration. The coincidence of the M mode under P and S excitations indicates that this nanocube has equal length and width. A close comparison between the two normalized scattering spectra reveals that the M peak for S-polarization behaves more like a single Lorentz line shape, while there is tail in the red side of the M mode for P-polarization excitation. An intentional subtraction of the spectrum for S-polarization from the one for P-polarization shows a peak around 800 nm. This is actually the excitation of the BD mode in NCOM, analogous to the bonding 6


Page 6 of 30

Page 7 of 30 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

ACS Nano

dipole-dipole mode in a nanocube dimer.45 Since this mode is even parity, it cannot be excited by normal incident light or inclinded S-polarized light. The M mode, however, is odd symmetry and can be excited by both P- and S-polarized light. FEM simulations confirm the above interpretations. As shown in Figure 2b, the scattering spectrum for P-polarized light matches quite well with the experimental result. The surface charge distribution at the peak position (725 nm) provides a clear identification of the M mode. For S-polarization, the tail at the red side of the M mode disappears because of the absence of the BD mode. A vertical electric dipole excitation at the center of the gap is specially applied in order to avoid exciting the M mode. The result exhibits a broad peak near 800 nm, and the surface charge distribution explicitly shows the BD mode features. Combining the experimental and calculation results, we confirm that the asymmetric line shape is due to the co-excitation of the M mode and the BD mode by P-polarized light. It should be noted that the BD modes are not always at the red side of the M mode, but depend on the gap distance, the size of the nanocubes and the edge rounding.45 Strong light-matter interaction requires simultaneous spatial and spectral overlap. The spatial overlap is realized by inserting the monolayer WSe2 into the region where the near-field of the plasmon mode is maximized. The spectral overlap, however, can be accomplished by tuning the plasmon energy across the excitonic transition by continuously depositing alumina layers over the nanocubes (shown in SI Figure S4a), as used in our recent work.23 Since the refractive index of alumina is larger than air, the deposited alumina layers will redshift the surface plasmon due to the dielectric screening effect.46 Also, the layer-by-layer grown procedure in ALD allows us to tune the plasmon energy in a decent and controllable manner. Here for a nanocube sitting on the monolayer WSe2 separated from the gold film by a 4 nm alumina, we deposited alumina to shift the plasmon peak in order to investigate the coupling between the X0 exciton in WSe2 and the M plasmon mode in NCOM. For the sample with 14 nm alumina coating, the dark-field scattering image is shown in Figure 3a and the single silver nanocube is marked out by the red dash circle. The edge length of this silver nanocube is 84.6 nm, measured from the SEM image in Figure 3b (the SEM image was taken after all the optical measurements). Normalized dark field 7


ACS Nano 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 30

scattering spectra for the silver nanocube under P-polarized excitation are shown in Figure 3c. As the alumina deposition thickness changed from 10 nm to 32 nm, the M mode peak redshifted from 739.1 nm to 774.4 nm, across the X0 exciton. The dip (highlighted by a grey line) at the wavelength of X0 exciton peak indicates the coherent coupling between the excitons and the plasmon. Considering that multiple excitons are involved in this coupling process, the phenomenological coupled harmonic oscillators model can be used to describe it. In this picture, the multiple excitons can be treated as a ‘super oscillator’, so the whole system can be simply described by two coupled oscillators:47  E pl − iΓ pl / 2  g 

g α  α    = E   , E 0 − iΓ 0 / 2   β  β 

(2)

where Epl and E0 are the energy of the plasmon and the X0 exciton, and g is the coupling strength. Γpl and Γ0 respectively represent the dissipation rates of plasmon and excitons. E stands for the energy of quasi-particles, α and β are eigenvector components. Solving the secular equation Eq. (2) and neglecting the high order parts in the widths of exciton and plasmon (i.e., assuming the widths of exciton and plasmon are small compared to their energies), we can get the solution E± =

1 ( E 0 + E pl ) ± 2

g2 +

1 2 δ , where δ = Epl – E0 is the detuning. 4

According to the coupled oscillator model, the scattering spectrum of the hybrid system follows48 ( E 2 − E0 2 + iE Γ 0 ) σ scat ( E ) = AE ( E 2 − E0 2 + iE Γ 0 )( E 2 − Epl 2 + iE Γ pl ) − 4 E 2 g 2 4

2

,

(3)

where A represents the scattering amplitude, E0 and Epl respectively stands for the resonance energy of the WSe2 X0 exciton and the magnetic plasmon mode, while Γ0 and Γpl are the dissipation terms. In order to extract the coupling strength and plasmon energy, we fit the scattering spectra in Figure 3c using Eq. (3). The results show an anti-crossing behavior (Figure 3d). When the detuning δ equals zero, the energy difference between the upper and lower plexciton branch defines the vacuum Rabi splitting Ω = 2g = 36.7 meV, which is comparable with the excitonic linewidth (45 meV) but smaller than the plasmon linewidth (130 meV). The extinction spectra of the hybrid system also show a Rabi splitting (details see SI Section 5). Therefore, the formation of plexcitons is valid even if considering the more rigorous criteria for plasmon-exciton interaction based on absorption or extinction.49 Figure 8


Page 9 of 30 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

ACS Nano

3e shows that there is a slight decline of the coupling strength against the thickness of the alumina coating, indicating that the near-field of the M mode within the gap region exhibits a small reduction upon alumina deposition. The Rabi splitting obtained here is smaller (but not far smaller) than both the exciton and plasmon linewidth, indicating that the coupling strength is on the border between the weak and the strong coupling regime. This is a very interesting regime where the hybrid system possesses the best light-emitting capacity.2 In the following, we will demonstrate this aspect by inspecting the PL emission from the hybrid system. First, to understand the enhancement of PL in the system of NCOM coupled with monolayer WSe2, we did the PL imaging (shown in Figure 4a). The silver nanocube measured in Figure 3c is marked out by the red dash circles and a bright spot indicates that the PL is tremendously enhanced. Then, we excited the sample with a 633 nm laser and collected the PL spectra for every alumina coating. The laser spot was about 6.3 µm2 and we only collected the PL signal from the central region (1.6 µm2) containing the silver nanocube. For example, Figure 4b shows the PL spectra from a region with and without a silver nanocube, for alumina coating of 14 nm. A 4.2 folds enhancement in spectra was obtained. Considering that PL is enhanced only at region close to the nanocube and the collected PL signal also contains contribution from the background WSe2, the actual PL enhancement around the nanocube should be large. To quantify the real PL enhancement around the silver nanocube, we define the PL enhancement factor as: EF =

I NC − I 0 S 0 , I0 S NC

(4)

where INC is the PL intensity from the WSe2 area with the silver nanocube and I0 is the PL intensity from the background WSe2 without the nanocube. S0 defines the area of WSe2 we collected (1.62 µm2), while SNC represents the area of WSe2 within the hotspot around the nanocube. For brevity, we assume SNC to be the area under the silver nanocube (0.0072 µm2) at first step. The PL enhancement factor evaluated by Eq. (4) is plot against the resonant wavelength of the M plasmon in Figure 4c for every coating. A maximum of 940 times enhancement is achieved when the M plasmon is on resonance with the X0 excitons. The PL enhancement agrees with the above assignment of intermediate coupling regime where the Purcell effect still applies. No splitting was observed in the PL spectra, also in agree 9


ACS Nano 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

with the fact that the Rabi splitting is smaller than the exciton linewidth.49 To understand the origin of the large PL enhancement, we calculated the LDOS in this NCOM system, since the spontaneous emission of the emitter is always governed by its photonic environment through the LDOS. For a perfect nanocube with equal length and width, the projected LDOS ρy(x) is degenerate in the x or y- direction. As shown in Figure 5a, the in-plane LDOS is enhanced in the gap between the silver nanocube and the gold film, at the edge of the nanocube. Compared to an emitter in vacuum, this enhancement is solely from the plasmonic part since the excitonic effect is removed by setting reduced oscillator strength f = 0 in Eq. (1). Apparently, the maximum LDOS enhancement (~ 140) is insufficient to explain the large PL enhancement obtained in the experiment. However, if the excitonic effect is included (non-zero f), the LDOS can be further increased. By carefully comparing the experimental measured scattering spectra with the FEM calculated scattering spectra, f should be as large as 0.8 so that the calculation can reproduce the experiment. This value is close to the one used in our previous work.23 Figure 5b (Figure 5c) shows the LDOS (absorption within the silver nanocube) as a function of wavelength, calculated by setting a dipole at (x, y) = (22 nm, -38 nm). As the reduced oscillator strength f increases, the LDOS increases accordingly with slight blue shift in the peak position. The maximum LDOS enhancement for f = 0.8 can reach as high as 5300, which can now explains the large PL enhancement observed in the experiment. The absorption spectra, however, gradually show an evolution from a single peak to a dip on top of a peak. The presence of the absorption dip indicates that the energy is extracted out of the nanocube as the reduced oscillator strength f increases. This is, exactly, the consequence of the energy transfer from the plasmon to the exciton. Meanwhile, a larger Rabi splitting in the absorption spectra can be observed when increasing f, which is attributed to the increasing coupling strength g.50 To account for these phenomena, the ratio of the energy in WSe2 to the energy in the whole system, defined as Fx, was calculated (Figure 5d). A monotonous increase dependent on the reduced oscillator strength f is shown. As illustrated in SI Section 6, the relation between coupling strength g and energy ratio Fx can be written as

g ∝

Fx

. Therefore, the f-dependent splitting behavior in Figure 5c directly

arises from the increasing energy ratio inside the WSe2 layer. A larger Fx leads to a stronger vacuum 10


Page 10 of 30

Page 11 of 30 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

ACS Nano

electric field in the WSe2, resulting in a higher spontaneous decay rate γ for an external dipole.51 This is why the calculated LDOS increases as f increases in Figure 5b. We should note that the LDOS won’t be further enhanced but will saturate and then decrease instead if f is getting larger and larger. It is because the hybrid system will enter the strong coupling regime where both the UPB and the LPB depart from the original exciton resonance as f increases. Understanding and being able to tailor the number of excitons involved in the coupling process is an important issue when designing light-emitting plexcitonic systems.52 In the following, we will estimate how many excitons are involved in the formation of plexciton in the NCOM system and devise ways to decrease this number by increasing the spatial confinement of the plexciton. We can do this, on one hand, by shortening the edge lengths of the nanocubes, which will confine the near-field of the plasmon to a smaller region and reduce the spatial overlap region with the WSe2 accordingly. Consequently, there will be fewer excitons interacting with the M mode for smaller nanocubes. On the other hand, we can blue-shift the M mode by increasing the gap distance between the silver nanocube and the gold film, whereas redshift the M mode by increasing the size of the silver nanocubes.42 To achieve the spectral overlap between the X0 excitons in WSe2 and the M plasmon mode within a large range of cube sizes, we changed the thickness of the spacer layer so that the M mode always had slightly higher energy compared to the X0 excitons. Three different alumina thickness were used, i.e., 1.6 nm, 3 nm and 4 nm. Note that the effect of the spacer thickness on the exciton transition was neglected, the scattering spectra of the background (WSe2 without cubes) show that the X0 exciton peak still exists for a thin alumina layer (shown in SI Figure S4b-c). Our experimental investigations of 9 nanocubes with different sizes, show that the Rabi splitting decreases when reducing the size of the cubes (Figure 6a). This is because when the size of the cubes decreases, the energy density tends to be distributed in the metals (Au film and Ag cube) rather than in the WSe2 layer (see SI Section 6 for details). To estimate how many excitons are taking part in the formation of plexciton, we need to calculate the scaling behavior of the mode volume of the plasmon mode. The effective mode volume V of the M mode in a bare NCOM, according to the usual definition, can be calculated by53,54 11


ACS Nano 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

V=

Page 12 of 30

2∫ W ( r ) dr 2

max(ε E )

,

(5)

 d ε ( r ) ω   2 2 1 1 where W ( r ) = Re    E ( r ) + µ 0 H ( r ) is the energy density. |E(r)| and |H(r)| are the 4  dω 4 

module of the electric and magnetic field, ε(r) is the material permittivity and µ0 is the vacuum magnetic permeability. In our calculations, the scattered electric and magnetic field obtained under normal incident plane wave excitation were used to represent the near-field of the M mode in the NCOM. Since the mode volume of a plasmonic cavity tends to show a linear divergence with increasing integral volume due to the leaky radiation,53 the mode volumes shown in Figure 6b are calculated by subtracting the linear divergence (details see SI Figure S8). With the size of the cubes decreasing gradually, the mode volumes shrink tremendously due to the highly confined property of plasmonic modes. It follows approximately by V ∝ L3 (L is the edge length of the nanocubes), similar to other single particle plasmonic cavities. The definition of mode volume in Eq.(5) assumes that the exciton locates exactly at the maximum of the plasmon near-field, outside the metal. Thus, the use of Evac = hω 2εV corresponds to the vacuum electric field right at this position. However, in the circumstance of plasmon-exciton coupling which involves more than one exciton around a nanocavity, the vacuum electric field is highly spatial dependent. As a result, the excitons around the nanoparticle show different coupling strength to the plasmon according to g 0 ( r ) = µ ⋅ E ip ( r ) , where E ip ( r ) is the in-plane vacuum electric field. The calculated maximum single exciton coupling strength for an 84.6 nm nanocube is g 0 = 0.59 meV , which is much smaller than the g from the experiment, indicating that multiple excitons are involved in this process. To estimate the number of excitons involved into the coupling process,23 we assumed that excitons were dispersed uniformly in the central plane of WSe2 layer with a Bohr radius defined density n e = η π a 2 . This quantity η, the number of excitons in a unit area defined by the exciton Bohr radius

a, is introduced to fit the calculated results to the experiment. And further derivations confirm that the coupling strength g is independent on the Bohr radius a (see SI Section 6). The summation of the 12


Page 13 of 30 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

ACS Nano

coupling strength for individual excitons gives birth to the total coupling strength according to the relationship g =

∑g i

2 0

(ri ) , where g0(ri) is the coupling strength for the i-th exciton at the position ri.

∫

2 2 In the calculation, the summation can be replaced by an integration g = g0 (ri ) ⋅ dn , where

dn =η ⋅ dS ( πa2 ) is the number of exciton within an elemental area dS. As the integral volume increases, g will saturates to the experimental result. Let Ne defines the exciton number on an area that contributes to 1 – 1/e2 times of the total coupling strength g. We obtain Ne = 1259 excitons for the hybrid NCOM shown in Figure 3 and 4. The same calculation was performed for the smallest nanocube (L = 63.0 nm) measured in the experiment and the number of excitons involved is 816. As we can see, with the size of the cubes decreasing, the number of the excitons involved into the coupling process is decreasing accordingly due to the smaller mode volume. For a rough estimation from the linear fitting in Figure 6a, the coupling strength for a 30 nm silver nanocube is 5.15 meV, and the number of excitons involved is only 25 (details see SI Section 8). According to the above assumption of uniform distribution of excitons, the Ne excitons that are actually coupled with the M mode occupy a hot area of S hot = π a 2 N e (3955 nm2).55 This area is smaller than the geometric area under the nanocube SNC = 0.0072 µm2, yet the PL enhancement actually originates from this area. With this consideration, we calculated the PL enhancement again using Eq. (4) but replacing SNC by Shot. The real PL enhancement induced by the M mode can reach 1700 times. Excitation enhancement is responsible for 4.1 times among the total one (details see SI Section 9). Thus, the emission enhancement is about 410 times, which is substantial compared with other works.33,34 In this study, we realized the coupling between the magnetic plasmon mode in a NCOM system and neutral (X0) excitons in monolayer WSe2. However, plasmon-exciton coupling can also occur between the BD mode and the excitons. Yet we conclude that the coupling strength can be ignored because the BD mode has a much smaller in-plane electric field within the WSe2 layer than the M mode. A strong evidence for this is that the dark-field scattering spectra measured using S-polarized incident light reveals a Rabi splitting of 35.0 meV, contributing the major part of the one measured using 13


ACS Nano 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

P-polarized light (see Figure S12 in SI). There are three major advantages of using the M mode in NCOM instead of the BD mode in typical nanoparticle over mirror (NPOM) for plasmon-exciton coupling. First, the M mode is more sensitive to the parameters (size, gap distance, refractive index, etc.) of the coupled system, which can be applied to design active-tuning hybrid system.45,56 Second, the M mode has narrower linewidth because of reduced radiative damping compared to the BD mode, which is a crucial parameter to realize strong plasmon-exciton coupling. Third, the M mode radiates along the surface normal that makes it optically accessible.30 The BD mode, however, has a vertical electric dipole nature with a maximum emission (excitation efficiency) at around 60 degrees with respect to the surface normal.57 As a NPOM system, the hybrid system can always be constructed directly by dropping chemically synthesized nanocubes, nanobars and nanorods etc. onto a TMD covered gold film. Alternately, it can also be fabricated using top-down techniques combining film deposition, chemical vapor deposition of TMDs and lithography defined patterning, etc. On the other hand, the drawback of using the NCOM coupled with the bright exciton in monolayer TMDs is that it has a much weaker in-plane field enhancement than its vertical one. So it couples weakly with X0 excitons possessing nearly in-plane transition dipole moment.39 The imperfect alignment of the electric field with the exciton dipole moment explains the overall small coupling strength in this work, and other similar configurations,23,30 even though the dipole moments of the excitons in TMDs are large in general. To obtain a larger coupling strength between the exciton and the localized plasmons, one has to carefully align the dipole orientation of the exciton with respect to maximum field component.58 Alternatively, further increase the spatial confinement of the plasmonic resonances by using sharp tips can also increase the coupling strength26. To take the advantage of strong out-of-plane electric field enhancement in the nanogap, one can use multi-layer TMDs which possesses a non-zero out-of-plane dipole moment.35 Also, recent studies on TMDs reveal the presence of dark excitons with an out-of-plane transition dipole moment.59,60 Therefore, the NCOM with considerable z- component electric field enhancement is expected to couple effectively with the dark excitons.

14


Page 14 of 30

Page 15 of 30 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

ACS Nano

CONCLUSION In this work, we investigated the coupling between the WSe2 X0 excitons and magnetic plasmon mode in NCOM system. The formation of plexciton was confirmed by mapping the plexciton dispersion in individual NCOM, revealing a Rabi splitting of 36.7 meV that was not found in previous studies using the same configuration. The splitting is comparable with the linewidth of the exciton but smaller than the linewidth of plasmon, implying that the coupling between plasmons and excitons is at the border of the weak and strong coupling regime. PL of the WSe2 X0 excitons is greatly enhanced, by a factor of 1700, including a 410-fold emission enhancement. This ultra-bright plexciton system takes full advantage of the plasmon-exciton coupling in the intermediate coupling regime – the LDOS at the excitonic wavelength is enhanced by both the plasmonic confinement and the strong excitonic effect that attracts electromagnetic energy to the excitonic material. This plexciton enhanced fluorescence will disappear if the hybrid system enters either the rigorously weak or strong coupling regime. What’s more, increasing the confinement of plexciton was accomplished by scaling the hybrid system, which is a critical aspect to study the plexciton-plexciton interaction, plexciton condensation, and integrated plexcitonic devices.

METHODS Silver nanocubes synthesis and sample preparation. The alumina layers were grown by atomic layer deposition at 120 °C. Through the CCD image contrast, the monolayer WSe2 can be identified. PL and Raman spectra were also used to characterize the sample. Silver nanocubes were prepared by polyol synthesis using polyvinylpyrrolidone (PVP) as the capping reagent, with the presence of a trace amount of NaHS.61 The size of these silver nanocubes can be identified from the SEM images. The resultant nanocubes had edge lengthes of 50 – 100 nm, covered by a PVP layer of 2 – 3 nm. Finally, solution with silver nanocubes were dropped directly on the position with monolayer WSe2, and then the silver nanocubes were dried naturally over the monolayer WSe2 after about 1 minute, constructing the hybrid NCOM system. Dark-field scattering spectroscopy and PL image. Oblique light illumination dark-field 15


ACS Nano 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

spectroscopy was used to acquire the scattering spectra of individual nanocubes. The non-polarized white light from a halogen lamp (100W, Olympus) passed through a polarizer (Olympus) and obliquely illuminated on the sample with an angle of 80° with respect to the substrate normal. An Olympus objective (100X, N.A. = 0.8) was utilized to collect the scattered light from the sample, and then the light was directed to a spectrometer (Renishaw inVia) equipped with an air-cooled CCD and a blazed grating (300 lines/mm, blazed at 1 µm). The PL image was obtained using a mercury lamp as excitation after passing through a 530-550 nm band pass filter. Then the signal passed through a 575 nm long pass filter before collected by an electron-multiplying CCD (Q-capture). The morphologies of nanocubes were characterized by SEM after all the optical measurements.

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: Monolayer WSe2 characterization, schematic modes excitation under P- or S- polarized light, scattering spectra of magnetic mode under polarized light excitation, characterization of individual NCOM and monolayer WSe2, FEM simulation of absorption and extinction spectrum, the derivation of the coupling strength, simulation of the mode volume, calculations of the number of excitons involved and the PL excitation enhancement, and the experimental dark-field scattering spectra as well as the anti-crossing behavior under S-polarized excitation. (PDF)

AUTHOR INFORMATION Corresponding Authors *E-mail: [email protected] *E-mail: [email protected] Author Contributions #

J.W.S. and H.T.H contributed equally to this work. 16


Page 16 of 30

Page 17 of 30 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

ACS Nano

H.X.X. and S.P.Z. conceived the idea and designed the project. J.W.S. prepared the samples and performed the experiments. H.T.H. performed the theoretical simulations. D.Z. and Q.D. help with the optical measurements. D.X.Z prepared the part of the nanocubes. S.P.Z, J.W.S. and D.Z. analyzed the data. S.P.Z, J.W.S., and H.T.H. wrote the manuscript. All authors discussed and commented on the manuscript. The authors declare no competing financial interest.

ACKNOWLEDGMENTS This work was supported by the National Key R&D Program of China (Grant No. 2017YFA0303504, 2017YFA0205800), the National Key Basic Research Program (Grant No. 2015CB932400), the National Natural Science Foundation of China (Grants Nos. 11674255 and 11674256) and the China Postdoctoral Science Foundation (Grant No. 2014T70727).

REFERENCES

(1) Marquier, F.; Sauvan, C.; Greffet, J. J. Revisiting Quantum Optics with Surface Plasmons and Plasmonic Resonators. ACS Photonics 2017, 4, 2091-2101. (2) Hugall, J. T.; Singh, A.; van Hulst, N. F. Plasmonic Cavity Coupling. ACS Photonics 2018, 5, 43-53. (3) Anger, P.; Bharadwaj, P.; Novotny, L. Enhancement and Quenching of Single-Molecule Fluorescence. Phys. Rev. Lett. 2006, 96, 113002. (4) Noda, S.; Fujita, M.; Asano, T. Spontaneous-Emission Control by Photonic Crystals and Nanocavities. Nat. Photonics 2007, 1, 449-458. (5) Pelton, M. Modified Spontaneous Emission in Nanophotonic Structures. Nat. Photonics 2015, 9, 427-435. (6) Khitrova, G.; Gibbs, H. M.; Kira, M.; Koch, S. W.; Scherer, A. Vacuum Rabi Splitting in Semiconductors. Nat. Phys. 2006, 2, 81-90. (7) Basov, D. N.; Fogler, M. M.; Garcia de Abajo, F. J. Polaritons in Van Der Waals Materials. Science 2016, 354, 195-203. 17


ACS Nano 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

(8) Zhang, W.; Govorov, A. O.; Bryant, G. W. Semiconductor-Metal Nanoparticle Molecules: Hybrid Excitons and the Nonlinear Fano Effect. Phys. Rev. Lett. 2006, 97, 146804. (9) Antosiewicz, T. J.; Apell, S. P.; Shegai, T. Plasmon–Exciton Interactions in a Core–Shell Geometry: From Enhanced Absorption to Strong Coupling. ACS Photonics 2014, 1, 454-463. (10) Splendiani, A.; Sun, L.; Zhang, Y.; Li, T.; Kim, J.; Chim, C. Y.; Galli, G.; Wang, F. Emerging Photoluminescence in Monolayer MoS2. Nano Lett. 2010, 10, 1271-1275. (11) Mak, K. F.; Lee, C.; Hone, J.; Shan, J.; Heinz, T. F. Atomically Thin MoS2: A New Direct-Gap Semiconductor. Phys. Rev. Lett. 2010, 105, 136805. (12) Xiao, D.; Liu, G. B.; Feng, W.; Xu, X.; Yao, W. Coupled Spin and Valley Physics in Monolayers of MoS2 and Other Group-VI Dichalcogenides. Phys. Rev. Lett. 2012, 108, 196802. (13) Zeng, H.; Dai, J.; Yao, W.; Xiao, D.; Cui, X. Valley Polarization in MoS2 Monolayers by Optical Pumping. Nat. Nanotechnol. 2012, 7, 490-493. (14) Cao, T.; Wang, G.; Han, W.; Ye, H.; Zhu, C.; Shi, J.; Niu, Q.; Tan, P.; Wang, E.; Liu, B.; Feng, J. Valley-Selective Circular Dichroism of Monolayer Molybdenum Disulphide. Nat. Commun. 2012, 3, 887. (15) Mak, K. F.; He, K.; Shan, J.; Heinz, T. F. Control of Valley Polarization in Monolayer MoS2 by Optical Helicity. Nat. Nanotechnol. 2012, 7, 494-498. (16) Ugeda, M. M.; Bradley, A. J.; Shi, S. F.; da Jornada, F. H.; Zhang, Y.; Qiu, D. Y.; Ruan, W.; Mo, S. K.; Hussain, Z.; Shen, Z. X.; Wang, F.; Louie, S. G.; Crommie, M. F. Giant Bandgap Renormalization and Excitonic Effects in a Monolayer Transition Metal Dichalcogenide Semiconductor. Nat. Mater. 2014, 13, 1091-1095. (17) He, K.; Kumar, N.; Zhao, L.; Wang, Z.; Mak, K. F.; Zhao, H.; Shan, J. Tightly Bound Excitons in Monolayer WSe2. Phys. Rev. Lett. 2014, 113, 026803. (18) Liu, X.; Galfsky, T.; Sun, Z.; Xia, F.; Lin, E.-c.; Lee, Y.-H.; Kéna-Cohen, S.; Menon, V. M. Strong Light–Matter Coupling in Two-Dimensional Atomic Crystals. Nat. Photonics 2014, 9, 30-34. (19) Liu, X.; Bao, W.; Li, Q.; Ropp, C.; Wang, Y.; Zhang, X. Control of Coherently Coupled Exciton Polaritons in Monolayer Tungsten Disulphide. Phys. Rev. Lett. 2017, 119, 027403. (20) Flatten, L. C.; He, Z.; Coles, D. M.; Trichet, A. A.; Powell, A. W.; Taylor, R. A.; Warner, J. H.; Smith, J. M. Room-Temperature Exciton-Polaritons with Two-Dimensional WS2. Sci. Rep. 2016, 6, 33134. (21) Liu, W.; Lee, B.; Naylor, C. H.; Ee, H. S.; Park, J.; Johnson, A. T.; Agarwal, R. Strong Exciton-Plasmon Coupling in MoS2 Coupled with Plasmonic Lattice. Nano Lett. 2016, 16, 1262-1269. (22) Wang, S.; Li, S.; Chervy, T.; Shalabney, A.; Azzini, S.; Orgiu, E.; Hutchison, J. A.; Genet, C.; Samori, P.; Ebbesen, T. W. Coherent Coupling of WS2 Monolayers with Metallic Photonic Nanostructures at Room Temperature. Nano Lett. 2016, 16, 4368-4374. (23) Zheng, D.; Zhang, S.; Deng, Q.; Kang, M.; Nordlander, P.; Xu, H. Manipulating Coherent Plasmon-Exciton Interaction in a Single Silver Nanorod on Monolayer WSe2. Nano Lett. 2017, 17, 3809-3814. (24) Wen, J.; Wang, H.; Wang, W.; Deng, Z.; Zhuang, C.; Zhang, Y.; Liu, F.; She, J.; Chen, J.; Chen, H.; Deng, S.; Xu, N. Room-Temperature Strong Light-Matter Interaction with Active Control in Single Plasmonic Nanorod Coupled with Two-Dimensional Atomic Crystals. Nano Lett. 2017, 17, 4689-4697. (25) Cuadra, J.; Baranov, D. G.; Wersäll, M.; Verre, R.; Antosiewicz, T. J.; Shegai, T. Observation of Tunable Charged Exciton Polaritons in Hybrid Monolayer WS2-Plasmonic Nanoantenna System. Nano Lett. 18


Page 18 of 30

Page 19 of 30 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

ACS Nano

2018, 18, 1777-1785. (26) Stührenberg, M.; Munkhbat, B.; Baranov, D. G.; Cuadra, J.; Yankovich, A. B.; Antosiewicz, T. J.; Olsson, E.; Shegai, T. Strong Light-Matter Coupling between Plasmons in Individual Gold Bi-Pyramids and Excitons in Mono- and Multilayer WSe2. Nano Lett. 2018, DOI: 10.1021/acs.nanolett.8b02652. (27) Majumdar, A.; Dodson, C. M.; Fryett, T. K.; Zhan, A.; Buckley, S.; Gerace, D. Hybrid 2D Material Nanophotonics: A Scalable Platform for Low-Power Nonlinear and Quantum Optics. ACS Photonics 2015, 2, 1160-1166. (28) Xu, H.; Bjerneld, E. J.; Käll, M.; Börjesson, L. Spectroscopy of Single Hemoglobin Molecules by Surface Enhanced Raman Scattering. Phys. Rev. Lett. 1999, 83, 4357-4360. (29) Xu, H.; Aizpurua, J.; Käll, M.; Apell, P. Electromagnetic Contributions to Single-Molecule Sensitivity in Surface-Enhanced Raman Scattering. Phys. Rev. E 2000, 62, 4318-4324. (30) Akselrod, G. M.; Argyropoulos, C.; Hoang, T. B.; Ciracì, C.; Fang, C.; Huang, J.; Smith, D. R.; Mikkelsen, M. H. Probing the Mechanisms of Large Purcell Enhancement in Plasmonic Nanoantennas. Nat. Photonics 2014, 8, 835-840. (31) Zhang, R.; Zhang, Y.; Dong, Z. C.; Jiang, S.; Zhang, C.; Chen, L. G.; Zhang, L.; Liao, Y.; Aizpurua, J.; Luo, Y.; Yang, J. L.; Hou, J. G. Chemical Mapping of a Single Molecule by Plasmon-Enhanced Raman Scattering. Nature 2013, 498, 82-86. (32) Xu, H.; Wang, X. H.; Persson, M. P.; Xu, H. Q.; Käll, M.; Johansson, P. Unified Treatment of Fluorescence and Raman Scattering Processes near Metal Surfaces. Phys. Rev. Lett. 2004, 93, 243002. (33) Akselrod, G. M.; Ming, T.; Argyropoulos, C.; Hoang, T. B.; Lin, Y.; Ling, X.; Smith, D. R.; Kong, J.; Mikkelsen, M. H. Leveraging Nanocavity Harmonics for Control of Optical Processes in 2D Semiconductors. Nano Lett. 2015, 15, 3578-3584. (34) Wang, Z.; Dong, Z.; Gu, Y.; Chang, Y. H.; Zhang, L.; Li, L. J.; Zhao, W.; Eda, G.; Zhang, W.; Grinblat, G.; Maier, S. A.; Yang, J. K.; Qiu, C. W.; Wee, A. T. Giant Photoluminescence Enhancement in Tungsten-Diselenide-Gold Plasmonic Hybrid Structures. Nat. Commun. 2016, 7, 11283. (35) Kleemann, M. E.; Chikkaraddy, R.; Alexeev, E. M.; Kos, D.; Carnegie, C.; Deacon, W.; de Pury, A. C.; Grosse, C.; de Nijs, B.; Mertens, J.; Tartakovskii, A. I.; Baumberg, J. J. Strong-Coupling of WSe2 in Ultra-Compact Plasmonic Nanocavities at Room Temperature. Nat. Commun. 2017, 8, 1296. (36) Nagpal, P.; Lindquist, N. C.; Oh, S. H.; Norris, D. J. Ultrasmooth Patterned Metals for Plasmonics and Metamaterials. Science 2009, 325, 594-597. (37) Kang, Y.; Najmaei, S.; Liu, Z.; Bao, Y.; Wang, Y.; Zhu, X.; Halas, N. J.; Nordlander, P.; Ajayan, P. M.; Lou, J.; Fang, Z. Plasmonic Hot Electron Induced Structural Phase Transition in a MoS2 Monolayer. Adv. Mater. 2014, 26, 6467-6471. (38) Najmaei, S.; Mlayah, A.; Arbouet, A.; Girard, C.; Léotin, J.; Lou, J. Plasmonic Pumping of Excitonic Photoluminescence in Hybrid MoS2-Au Nanostructures. ACS Nano 2014, 8, 12682-12689. (39) Schuller, J. A.; Karaveli, S.; Schiros, T.; He, K.; Yang, S.; Kymissis, I.; Shan, J.; Zia, R. Orientation of Luminescent Excitons in Layered Nanomaterials. Nat. Nanotechnol. 2013, 8, 271-276. (40) Johnson, P. B.; Christy, R. W. Optical Constants of the Noble Metals. Phys. Rev. B 1972, 6, 4370-4379. (41) Thilagam, A. Exciton Complexes in Low Dimensional Transition Metal Dichalcogenides. J. Appl. Phys. 2014, 116, 053523. 19


ACS Nano 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

(42) Lassiter, J. B.; McGuire, F.; Mock, J. J.; Ciraci, C.; Hill, R. T.; Wiley, B. J.; Chilkoti, A.; Smith, D. R. Plasmonic Waveguide Modes of Film-Coupled Metallic Nanocubes. Nano Lett. 2013, 13, 5866-5872. (43) Bowen, P. T.; Smith, D. R. Coupled-Mode Theory for Film-Coupled Plasmonic Nanocubes. Phys. Rev. B 2014, 90, 195402. (44) Kristensen, P. T.; Hughes, S. Modes and Mode Volumes of Leaky Optical Cavities and Plasmonic Nanoresonators. ACS Photonics 2014, 1, 2-10. (45) Zhang, S.; Xu, H. Tunable Dark Plasmons in a Metallic Nanocube Dimer: Toward Ultimate Sensitivity Nanoplasmonic Sensors. Nanoscale 2016, 8, 13722-13729. (46) Xu, G.; Tazawa, M.; Jin, P.; Nakao, S.; Yoshimura, K. Wavelength Tuning of Surface Plasmon Resonance Using Dielectric Layers on Silver Island Films. Appl. Phys. Lett. 2003, 82, 3811-3813. (47) Törma, P.; Barnes, W. L. Strong Coupling between Surface Plasmon Polaritons and Emitters: A Review. Rep. Prog. Phys. 2015, 78, 013901. (48) Wu, X.; Gray, S. K.; Pelton, M. Quantum-Dot-Induced Transparency in a Nanoscale Plasmonic Resonator. Opt. Express 2010, 18, 23633-23645. (49) Wersäll, M.; Cuadra, J.; Antosiewicz, T. J.; Balci, S.; Shegai, T. Observation of Mode Splitting in Photoluminescence of Individual Plasmonic Nanoparticles Strongly Coupled to Molecular Excitons. Nano Lett. 2017, 17, 551-558. (50) Thomas, R.; Thomas, A.; Pullanchery, S.; Joseph, L.; Somasundaran, S. M.; Swathi, R. S.; Gray, S. K.; Thomas, K. G. Plexcitons: The Role of Oscillator Strengths and Spectral Widths in Determining Strong Coupling. ACS Nano 2018, 12, 402-415. (51) Lodahl, P.; Mahmoodian, S.; Stobbe, S. Interfacing Single Photons and Single Quantum Dots with Photonic Nanostructures. Rev. Mod. Phys. 2015, 87, 347-400. (52) Zengin, G.; Wersäll, M.; Nilsson, S.; Antosiewicz, T. J.; Käll, M.; Shegai, T. Realizing Strong Light-Matter Interactions between Single-Nanoparticle Plasmons and Molecular Excitons at Ambient Conditions. Phys. Rev. Lett. 2015, 114, 157401. (53) Koenderink, A. F. On the Use of Purcell Factors for Plasmon Antennas. Opt. Lett. 2010, 35, 4208-4210. (54) Huang, S.; Ming, T.; Lin, Y.; Ling, X.; Ruan, Q.; Palacios, T.; Wang, J.; Dresselhaus, M.; Kong, J. Ultrasmall Mode Volumes in Plasmonic Cavities of Nanoparticle-on-Mirror Structures. Small 2016, 12, 5190-5199. (55) Chen, W.; Zhang, S.; Kang, M.; Liu, W.; Ou, Z.; Li, Y.; Zhang, Y.; Guan, Z.; Xu, H. Probing the Limits of Plasmonic Enhancement Using a Two-Dimensional Atomic Crystal Probe. Light: Sci. Appl. 2018, 7, 56. (56) Chen, W.; Zhang, S.; Deng, Q.; Xu, H. Probing of Sub-Picometer Vertical Differential Resolutions Using Cavity Plasmons. Nat. Commun. 2018, 9, 801. (57) Mubeen, S.; Zhang, S.; Kim, N.; Lee, S.; Krämer, S.; Xu, H.; Moskovits, M. Plasmonic Properties of Gold Nanoparticles Separated from a Gold Mirror by an Ultrathin Oxide. Nano Lett. 2012, 12, 2088-2094. (58) Chikkaraddy, R.; de Nijs, B.; Benz, F.; Barrow, S. J.; Scherman, O. A.; Rosta, E.; Demetriadou, A.; Fox, P.; Hess, O.; Baumberg, J. J. Single-Molecule Strong Coupling at Room Temperature in Plasmonic Nanocavities. Nature 2016, 535, 127-130. (59) Slobodeniuk, A. O.; Basko, D. M. Spin–Flip Processes and Radiative Decay of Dark Intravalley 20


Page 20 of 30

Page 21 of 30 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

ACS Nano

Excitons in Transition Metal Dichalcogenide Monolayers. 2D Mater. 2016, 3, 035009. (60) Dery, H.; Song, Y. Polarization Analysis of Excitons in Monolayer and Bilayer Transition-Metal Dichalcogenides. Phys. Rev. B 2015, 92, 125431. (61) Siekkinen, A. R.; McLellan, J. M.; Chen, J.; Xia, Y. Rapid Synthesis of Small Silver Nanocubes by Mediating Polyol Reduction with a Trace Amount of Sodium Sulfide or Sodium Hydrosulfide. Chem. Phys. Lett. 2006, 432, 491-496.

21


ACS Nano 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

Figure Captions Figure 1. (a) Schematic of NCOM system-silver nanocube over an ultra-smooth gold film substrate, separated by an alumina deposition, monolayer WSe2 and the PVP layer around the nanocube. In-plane excitons (linked red and blue spheres) are situated in the monolayer WSe2. (b) The calculated scattering spectrum of a NCOM system excited by a normally incident x-polarized plane wave. The character ‘M’ denotes the magnetic plasmon resonance ultilized throughout this work. (c)-(e) The distribution of the x, y, z-component of the scattered electric field at the central plane of the WSe2 layer, normalized to the incident electric field E0. The maps correspond to the M peak at 743.4 nm shown in (b).

Figure 2. (a) Normalized dark field scattering spectra excited by P- (red) and S- (black) polarized light, and the difference (blue) between them shows a peak around 800 nm. (b) Calculated normalized scattering spectrum excited by P-polarized plane wave (PW, red line) and normalized absorption spectrum excited by a dipole along z direction (blue line). Red and blue triangles indicate the M mode at 725 nm and BD mode at 800 nm. The upper inset shows charge distribution corresponding to the M mode while the lower inset corresponds to the BD mode.

Figure 3. (a) Dark field scattering image of the NCOM coupled with monolayer WSe2. The single silver nanocube was marked out by the red dash circles. White dash line represents the boundary of monolayer WSe2. (b) SEM image of the single silver nanocube marked out in (a). (c) Normalized scattering spectrum for a single silver nanocube (marked out by the white circles in (a)) over ultra-smooth gold film coupled with monolayer WSe2, with alumina coating ranging from 10 nm to 32 nm. P-polarized light was utilized as the excitation. (d) Energy of upper plexciton branch (UPB) and lower plexciton branch (LPB) against detuning. (e) Coupling strength g against the thickness of alumina coating over the silver nanocube. The gray line denotes the averaged value of g.

Figure 4. (a) PL imaging of the NCOM coupled with monolayer WSe2. The red dash circle marks the same silver nanocube measured in Figure 3. The white dash line highlights the boundary of monolayer WSe2. (b) The PL spectra of the monolayer WSe2 with (red) and without (black) the silver nanocube. (c) The PL enhancement against the resonance wavelength of the M mode.

22


Page 22 of 30

Page 23 of 30 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

ACS Nano

Figure 5. (a) The LDOS mapping of a bare NCOM in x or y- direction, evaluated at 746.4 nm corresponding to the exciton peak. (b) LDOS spectra of the NCOM coupled with monolayer WSe2. The inset shows the top view of the silver nanocube with a red dot representing the position of a y-polarized dipole. (c) Normalized absorption energy within the silver nanocube for different reduced oscillator strength f. (d) Energy ratio harbored in WSe2 and other part as a function of the reduced oscillator strength f.

Figure 6. (a) The Rabi splitting Ω fitting from dark-field scattering spectra for 9 silver nanocubes of various sizes. (b) Mode volumes of NCOM systems for silver nanocubes varying from 30 nm to 110 nm.

23


ACS Nano 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

Figure 1 170x76mm (300 x 300 DPI)


Page 24 of 30

Page 25 of 30 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

ACS Nano

Figure 2 85x34mm (300 x 300 DPI)


ACS Nano 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

Figure 3 160x98mm (300 x 300 DPI)


Page 26 of 30

Page 27 of 30 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

ACS Nano

Figure 4 160x48mm (300 x 300 DPI)


ACS Nano 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

Figure 5 120x106mm (300 x 300 DPI)


Page 28 of 30

Page 29 of 30 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

ACS Nano

Figure 6 85x42mm (300 x 300 DPI)


ACS Nano 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

TOC 76x50mm (300 x 300 DPI)


Page 30 of 30

Light-Emitting Plexciton: Exploiting PlasmonâExciton Interaction in the

Recommend Documents

Light-Emitting Plexciton: Exploiting PlasmonâExciton Interaction in the