Copper-Catalyzed Triboration of Terminal Alkynes Using B2pin2: Efficient Synthesis of 1,1,2-Triborylalkenes

Article abstract of DOI:10.1002/anie.201908466

We report herein the catalytic triboration of terminal alkynes with B₂pin₂ (bis(pinacolato)diboron) using readily available Cu(OAc)₂ and PⁿBu₃. Various 1,1,2-triborylalkenes, a class of compounds that have been demonstrated to be potential matrix metalloproteinase (MMP-2) inhibitors, were obtained directly in moderate to good yields. The process features mild reaction conditions, a broad substrate scope, and good functional group tolerance. This copper-catalyzed reaction can be conducted on a gram scale to produce the corresponding 1,1,2-triborylalkenes in modest yields. The utility of these products was demonstrated by further transformations of the C?B bonds to prepare gem-dihaloborylalkenes (F, Cl, Br), monohaloborylalkenes (Cl, Br), and trans-diaryldiborylalkenes, which serve as important synthons and have previously been challenging to prepare.

Full text of DOI:10.1002/anie.201908466

FULL PAPER
Strong Coupling
www.advopticalmat.de
Strong Plasmon–Exciton Interactions on Nanoantenna
Array–Monolayer WS2 Hybrid System
Lin Liu, Landobasa Y. M. Tobing, Xuechao Yu, Jinchao Tong, Bo Qiang,
Antonio I. Fernández-Domínguez, Francisco J. Garcia-Vidal, Dao Hua Zhang,*
Qi Jie Wang,* and Yu Luo*
exemplified in spontaneous emissions
of quantum emitters.[16–19] However, this
is qualitatively different when the light–
matter interaction enters the strong cou-
pling regime, where coherent oscillations
of energy between excitons and photons
dominates over the electromagnetic
damping, manifesting in clear spectral
splitting known as the Rabi splitting.[20–23]
Realization of strong coupling requires
a large density of photonic states, which
is mainly dependent on the spatiotem-
poral confinement of electromagnetic
fields characterized by the mode volume
V and resonance Q-factor, LDOS ∝ Q/V.
For this, many efforts have been devoted
Strong plasmon–exciton interactions in monolayer transition-metal dichalco-
genides (TMDs) is emerging as a promising material platform for light emis-
sions, nonlinear optics, and quantum communications, and their realizations
require highly localized electric fields parallel to the transition dipole moment
of TMD excitons. Here, a systematic study of light–matter interaction in
planar dimer nanoantenna of nanoscale gaps coupled with monolayer WS2 is
presented, where the effects of the local field enhancement and spatial mode
overlap in the plasmon–exciton coupling strength are experimentally inves-
tigated. Importantly, anticrossing behaviors in the strong coupling regime
with a Rabi splitting of Ω = 118 meV for gold bowtie antennas with 7 nm gaps
and Ω = 138 meV for gold dimer arrays with 10 nm gaps are demonstrated.
to realize photonic microcavities with high Q/V, namely, high
finesse resonators based on photonic crystals[24–26] and whis-
pering gallery modes.[27] Despite their ultrahigh Q resonances,
realizations of strong coupling in these platforms often require
their characterizations implemented in cryogenic temperatures
due to the mismatch between the photon and exciton lifetimes.
In addition, light–matter coupling occurs within an interaction
volume dictated by the diffraction-limit, thereby hindering the
characterization of light–matter interaction at a single molecule
level. Alternatively, plasmonic systems have been explored for
strong light–matter interaction at the nanoscale, particularly
because plasmonic nanocavities also exhibit high Q/V due to
their strong field confinement at the deep subwavelength scale
despite their much lower Q-factor than their dielectric counter-
parts. Different kinds of quantum emitters have been explored
for studying strong coupling with plasmonic nanostructures,
including J-aggregates,[28–32] dye molecules[33–35] and quantum
dots.[36,37] Recently, monolayers made of transition metal dichal-
cogenides (TMDs), when combined with plasmonic nanostruc-
tures, have been shown to be excellent platforms for achieving
strong coupling at room temperature due to their direct
bandgap and strong binding energy of their excitons. [38,39]
Different kinds of plasmonic systems have been explored
for plasmon–exciton coupling with TMDs, including nano-
cavities,[40] self-assembled metallic nanoparticles,[41–44] nano-
antenna arrays,[17,45–47] and metal nanohole arrays.[48] For strong
coupling, it is important to design plasmonic nanocavities
with a mode field parallel to the transition dipole moment of
the excitons in TMDs. In nanoparticle on mirror (NPoM) sys-
tems,[40] for example, multilayered TMDs are used such that
1. Introduction
The interaction of highly localized modes with nanoscale emit-
ters has gained much attention in the past decades[1–3] for its
importance in fundamental quantum physics, such as Bose-
Einstein condensation,[4–7] quantum vortex[8] and non-Hermi-
tian physics[9,10] and applications in optical transistors,[11,12]
polariton switches[13,14] and single photon sources.[15] Based
on the energy exchange rate between matter and electromag-
netic modes relative to their respective damping constants,
light–matter interaction strength can be classified into weak
and strong coupling regimes. In a weakly-coupled system,
optical processes involving light and matter are irreversible,
L. Liu, Dr. L. Y. M. Tobing, Dr. X. Yu, Dr. J. Tong, Dr. B. Qiang,
Prof. D. H. Zhang, Prof. Q. J. Wang, Prof. Y. Luo
School of Electrical & Electronic Engineering
Nanyang Technological University
Singapore 639798, Singapore
E-mail: edhzhang@ntu.edu.sg; qjwang@ntu.edu.sg; luoyu@ntu.edu.sg
Prof. A. I. Fernández-Domínguez, Prof. F. J. Garcia-Vidal
Departamento de Física Teorica de la Materia Condensada and
Condensed Matter Physics Center (IFIMAC)
Universidad′ Autonoma de Madrid
28049 Madrid, Spain
Prof. F. J. Garcia-Vidal
Donostia International Physics Center (DIPC)
20018 Donostia-San Sebastian, Spain
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adom.201901002.
DOI: 10.1002/adom.201901002
©
Adv. Optical Mater. 2019, 1901002
1901002 (1 of 8)
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advopticalmat.de
the transition dipole moment of the TMDs is reoriented toward
the out-of-plane resonance mode field. However, for monolayer
TMD in which the transition dipole moment is purely in-plane,
the plasmonic mode of the NPoM system only couples weakly
to the exciton despite its strong field enhancement in the die-
lectric spacer separating the nanoparticle and the metal film,
which is largely due to its orthogonal mode field. In the context
of synthesized nanoparticles, attempts to reduce misalignment
between the mode field and the exciton transition dipole have
been explored, notably in the use of gold bi-pyramids[43] and
ultrathin gold nanodisks.[44]
Meanwhile, top-down fabricated planar nanoantennae with
nanogaps exhibit strong in-plane mode fields, and have been
demonstrated to couple strongly with quantum emitters.[49–51]
The main advantage of the top-down fabricated antennae lies in
their lithographic flexibility, which could be used as a platform
for studying light–matter interaction with TMD, for example in
the moderate coupling of bowtie nanoantenna with monolayer
MoS2 that manifests in the emergence of Fano resonances.[17]
In this work, we present an in-depth analysis of the strong-cou-
pling anticrossing characteristics in planar gold nanoantenna
with nanogaps, namely bowtie and square dimer nanoantennae.
By fabricating nanoantennae of different gaps, sizes, and peri-
odicity on the same chip, we are able to investigate systemati-
cally the roles of E-field enhancements and spatial mode overlap
in the coupling strength. By introducing ultrathin dielectric
interlayer, we observe the effect of spatial mode overlap where
≈20–30% decrease of Rabi splitting is observed when 2 nm thick
Al2O3 layer is inserted. Specifically, we experimentally demon-
strate a direct evidence for the correlation between the local den-
sity of states and the coupling strength. We demonstrate a Rabi
splitting of Ω = 138 meV based on square dimer of sub-10 nm
gaps, which is shown to be in strong-coupling regime and con-
firmed by photoluminescence measurements of TMD-dimer
systems based on CVD-grown and exfoliated monolayer WS2.
where (α, β) are eigen coefficients satisfying |α|2 + |β|2 = 1, and



0 




2
γpl −γ
γpl −γ
1
1
2
ω = ω +ω +i
ω −ω +i
(3)

0 
+ 4gc2
pl
0
pl
0

2
2
2
are the eigenfrequencies ω of the hybrid modes, with γpl and
γ0 as the dissipation rates of uncoupled plasmon and exciton
modes. By ignoring the dissipation terms, the eigenfrequencies
can be simplified as [54,55]
1
ω =
ω
(
pl +ω0
gc2 +δ 2/4
(4)
)
2
with δ = ωpl − ω0 as the detuning between plasmon mode and
exciton energy. The vacuum Rabi splitting is the resonance
splitting at zero detuning, Ω = 2gc, which can be derived from
Ω = (ω −ω )(ω −ω )
Equation (4) at zero detuning as
. The
+
0
0
−
system is said to be in strong coupling when the rate of energy
exchange is faster than the loss. The rather generic criterion for
strong coupling is Ω > (γ+ + γ−)/2, where γ are the linewidth
of the split polaritonic modes. A more rigorous criterion for
the strong coupling has been studied in detail,[56] suggesting
that strong coupling is achieved when |Ω/(γpl − γ0)| > 1. For N
excitons coupled with a plasmon mode, the coupling strength
can be written as: gc = N µe |Epl |, where µe is the transition
diople moment of exciton and |Epl | = ω/20V is the plas-
monic field. The quality factor of the plasmonic antenna can
be calculated as Q = ωpl/γpl. Therefore, strong plasmon–exciton
gc2 /γ pl ≈ Q N/V
(
)
.
coupling is related with LDOS as
3. Results and Discussions
As shown in Figure 1, we present a detailed study of plasmon–
exciton coupling of gold nanoantenna arrays lithographi-
cally fabricated on monolayer WS2 films deposited on SiO2/Si
substrates, where square dimer and bowtie antenna designs
were chosen in this study for their strong local E-fields across
nanogaps and their small mode volume. As commented above,
the advantage of these top-down fabricated antennae over self-
assembled nanoparticle on mirror (NPoM) systems lies in
their in-plane local E-fields that align with the transition dipole
moment of the monolayer WS2. Furthermore, the lithographic
flexibility of the top-down fabrication allows us to tune the
antenna dimensions for studying plasmon–exciton coupling at
different resonance detuning. We also introduced an ultrathin
dielectric spacer between the WS2 and gold nanoantenna for
investigating the effect of the plasmon mode spatial overlap
with the WS2 to the plasmon–exciton coupling strength. The
scanning electron micrographs (SEMs) of the fabricated gold
nanostructures are shown in Figure 1c (bowtie nanoantenna)
and Figure 1d (square dimer). In order to ensure a small mode
volume, we fabricated these nanoantenna with gap spacing (g)
in ≈10 nm range (see the Experimental Section). The other
parameters such as the side length (s) and apex angle (θ) were
chosen to match the plasmon mode frequency (ωpl) close to
the WS2 exciton transition (ω0). The periodicities (Px,Py) of the
nanoantenna array were designed as Px = Py = 4s to minimize
2. Theoretical Model
The strong coupling of a plasmonic resonance with excitons
supported by a monolayer TMD can be understood from the so-
called Jaynes–Cummings (JC) Hamiltonian, which describes the
interaction of electromagnetic field with a two-level atom [52,53]
ˆ† ˆ
ˆ
ˆ†
ˆ
−
ˆ
ˆ
ˆ
HJC = ωpla a +ω0σ z /2 + gc aσ + a σ
(1)
(
)
+
where the first term describes the electromagnetic energy
characterized by the plasmon mode frequency ωpl, the second
term describes the two-level excitation characterized by exciton
frequency ω0, and the third term describes the exciton-photon
ˆ†
ˆ
interaction characterized by coupling strength gc. a and a is the
ˆ
creation and annihilation operators for the photon, while σ+ and
ˆ
σ−
ˆ
is
z
σ
are the raising and lowering operators for the exciton.
the atomic inversion operator. Rearranging the JC Hamiltonian
into an eigenvalue problem for the hybrid states gives


ωpl + iγ pl /2
gc




α
β
α
β
(2)
= ω










gc
ω0 + iγ 0 /2


©
Adv. Optical Mater. 2019, 1901002
1901002 (2 of 8)
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advopticalmat.de
Figure 1. Hybrid TMDC-nanoantenna array systems. a,b) Schematics of gold bowtie and square dimer arrays coupled with monolayer WS2 separated
by dielectric spacer layer. c,d) Scanning electron microscopy images of the fabricated gold nanoantenna with sub-10 nm gaps. e) Raman spectrum
and f) photoluminescence of monolayer WS2.
the near-field coupling between nanoantennae, which is known
to impart resonance broadening and, consequently, decreases
the Q factor. This is confirmed in our numerical simulations
which reveals that the scattering cross sections (σscat) of the
nanoantenna are smaller than the unit cell area (σscat < PxPy).
However, the far-field coupling still exists among these nano-
antenna due to the strong radiation of the dimer antenna. We
used this far-field coupling to further enhance the local E-fields,
from which our numerical simulation show that the E-field can
be ≈2.5× higher in a nanoantenna array (with Px,y = 4s) than
in an isolated antenna. Another purpose of using the nanoan-
tenna array is to investigate the mode interaction under two
mode excitations. The first type is the normal excitation (in
bright field geometry), which generates a Fano-like resonance
that results from the interaction between the localized surface
plasmon (LSP) of the gold nanoantenna and the thin film reso-
nance of the 295 nm thick SiO2 film. The second type is the
oblique excitation (in dark field geometry) which generates only
the LSP modes.
and A1g mode 417 cm−1 with a 64.1 cm−1 frequency difference
between the two modes. This value is close to the 65.5 cm−1
frequency difference for monolayer WS2 in other report,[58]
thereby confirming the monolayer characteristics of our CVD-
grown WS2. The photoluminescence of our CVD-grown WS2
(under 532 nm laser excitation) is presented in Figure 1f,
showing exciton peak at ω0 = 2 eV with γ0 = 25 meV linewidth.
We also note the two small peaks near the 532 nm excitation,
which are the two Raman signals in Figure 1e.
The mappings of the optical response of the gold nanoan-
tenna arrays were carried out by hyperspectral imaging system
(see the Experimental Section) schematically illustrated in
the Supporting Information (Figure S1). Figure 2a displays
the x-polarized scattering spectrum (in the bright-field geom-
etry) of square dimer arrays coupled with monolayer WS2 with
near zero detuning (s = 60 nm, g = 20 nm), where a weak dip
is observed near the exciton transition in between more pro-
nounced resonance dips. The weak dip signifies the mode
splitting which results from the plasmon–exciton coupling,
with the split modes indicated by the weak scattering peaks.
Such an obscured spectral splitting is attributed to the hybrid
mode resulting from the interference between the LSP mode
of the nanoantenna and the mode arising from the multiple
reflections inside the 295 nm SiO2 film. In order to verify if
this is indeed caused by multiple reflection that builds up only
under normal incidences, we excite the square dimers under
oblique illumination (in the dark-field geometry). As evident
in Figure 2b, we observed an optical response with clear mode
splitting at the exciton transition (denoted by the shaded band),
indicating the absence of the Fabry–Perot (FP)-like mode
under oblique excitation. The small bump around 546 nm is
not associated with any resonance as it is the spectral artifact
of the mercury lamp itself. The hybridization of this mode is
numerically simulated in Figure 2c, which shows that the two
dips at 500 and 700 nm stem from the modified FP resonance
The WS2 considered in this work was grown by chemical
vapor deposition (CVD) instead of realized by mechanical
exfoliation. Despite the lower optical quality of the former, we
note the importance of the large-scale uniformity of the WS2
film in this study, particularly in the context of characterizing
plasmon–exciton couplings with gold nanoantenna arrays fab-
ricated across the sample. For transition metal dichalcogenide
(MX2) layered compounds (M = Mo, W and X = S, Se, Te), there
are generally four Raman-active modes, namely the A1g, E1g, E21g
and E22g modes.[57] In order to investigate the monolayer proper-
ties of the CVD-grown WS2, we focus our attention to the in-
plane vibrational E21g mode and the out-of-plane vibrational A1g
mode, where the former shows little dependence on the film
thickness and the latter undergoes a blue shift with increasing
layer number.[58] As shown in Figure 1e, Raman characteriza-
E21g
tion of the CVD-grown WS2 shows the
mode at 353 cm−1
©
Adv. Optical Mater. 2019, 1901002
1901002 (3 of 8)
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advopticalmat.de
Figure 2. Resonance splitting near zero detuning for gold square dimer coupled with monolayer WS2. a) Bright-field and b) dark-field optical responses
of square dimers with nominal side length s = 60 nm and gap spacing g = 20 nm. The optical response of monolayer WS2 exciton (ω0 = 2 eV) is
denoted by the red band whose width represents the emission linewidth ( γ0 = 64 meV). c) Fullwave simulations depicting mode interference between
LSP resonance of gold nanoantenna and FP resonance of multiple thin film reflections.
condition due to the presence of gold nanoantenna (dashed
blue curve). The phase response (φ) of a strongly radiating plas-
monic antenna such as dimer or disk is close to φ = π at LSP
resonance (for negligible damping loss), translating to a rapid
change of the FP interference condition from constructive to
destructive, satisfying the FP resonance condition in two dif-
ferent frequencies. Thus, in addition to the LSP resonance posi-
tion that remains unchanged, we observe two reflectance dips
associated with light localization inside SiO2 film. In the pres-
ence of monolayer WS2, a mode splitting is to be expected for
the reflection peak at the exciton transition (solid blue curve).
The simulated spectra differ from the measure spectra notably
in their spectral contrasts. Such a difference in spectral contrast
is attributed to the nonideal mode excitation in our experiment,
where the plasmonic nanoantenna arrays were excited at dif-
ferent incident angles dictated by the numerical aperture of our
objective lens (which is NA = 0.55). Other factors include the
permittivity of both the physically deposited gold and the CVD-
grown WS2 monolayer, and also the surface roughness and
roundedness in the gold nanoantenna structure.
We present in Figure 3 anticrossing behaviors of gold bowtie
and square dimer arrays coupled with monolayer WS2 based on
dark-field measurements. The upper (ω+) and lower (ω−) energy
branches of the hybrid modes are plotted for different resonance
detunings (δ = ωpl − ω0), realized through geometric variations
of the nanoantenna. Red dashed line denotes the WS2 exciton
energy at 2 eV, while the blue and green solid curves denote
the fitting based on Equation (4). In Figure 3a, we present the
anticrossing behavior for gold bowtie nanoantenna, which
were fabricated with different nominal apex angles (θ = 40°,
50°, 60°) and side lengths (s = 60, 80, 100 nm). Although the
gap spacing was nominally fixed as g = 20 nm, the actual gap
spacing in the fabricated structures can be tuned by adjusting
the exposure dose and time. In our experiments, we have
realized bowtie antenna with gaps as small as g ≈ 7 nm with
good repeatability. The Rabi splitting for WS2-bowtie system
was found to be Ω = 118 meV. In Figure 3b, we present the
anticrossing behavior for gold square dimer, which were fabri-
cated with different nominal side lengths (s = 60, 80, 100 nm)
and gap spacings (g = 20, 30, 40 nm). Similar to bowtie struc-
tures, the real gap spacings were realized by tuning the expo-
sure dose and develop time, from which gold square dimer with
gap as small as g ≈ 10 nm have been realized. The Rabi split-
ting for WS2-dimer system was found to be Ω = 138 meV. It is
worth noting that wider Rabi splitting is observed for the square
dimers than for bowtie antennae despite the smaller mode
volume of the latter, which indicates the higher Q/V values and
therefore larger local density of states (LDOS) of the former.
This is verified by the smaller local E-field of the bowtie antenna
(Figure 3e), as compared to that of the square dimer (Figure 3f).
The geometry shapes of the nanoantenna arrays were designed
to have LSP resonances near WS2 transition at 620 nm (2 eV).
The details of geometrical structures, resonance wavelength, Q
factor and mode area for the two structures are given in Table S1
(Supporting Information), showing that square dimer antenna
exhibits 70% higher Q-factor with 26% larger mode area com-
pared to those of the bowtie antenna. The role of E-field enhance-
ment in the coupling strength is studied in isolated square
dimers separated by much larger periodicity ( Px,y = 2 µm). In
such situations, both the near-field and far-field couplings can
be ignored, which results in 2.5× smaller E-field enhancements
for the isolated square dimers (Figure S3c,d, Supporting Infor-
mation). The much higher field enhancement in nanoantenna
arrays is attributed to Rayleigh–Wood anomaly arising from
the diffraction condition of the pattern. Periodic patterning of
dimers into an array reduces the linewidth of the dimers LSP
resonance.[60,61] The anticrossing behavior for isolated square
dimers is given in the Supporting Information (Figure S3b),
showing narrower Rabi splitting (Ω = 114 meV) compared to
that of the square dimer arrays (Ω = 138 meV).
©
Adv. Optical Mater. 2019, 1901002
1901002 (4 of 8)
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advopticalmat.de
Figure 3. Anticrossing behaviors in strongly coupled systems for a) bowtie and b) square dimer antennae. The effect of 2 nm thick Al2O3 spacer in
reducing spatial mode overlap is demonstrated for c) bowtie and d) square dimer. Corresponding |E|-field distributions at near zero detuning for
e) bowtie (s = 60 nm, θ = 60°) and f) square dimer (s = 60 nm, g = 20 nm) arrays.
We have also investigated the role of spatial overlap between
the plasmon mode and the excitons in the coupling strength,
where an ultrathin dielectric spacer is introduced between the gold
nanoantenna array and monolayer WS2. Insertion of 2 nm thick
Al2O3 layer leads to a decrease of the Rabi splitting from Ω = 118
to 94 meV for the bowtie antenna (Figure 3c), and from Ω = 138
to 97 meV for the square dimers (Figure 3d). In addition to the
decrease of spatial overlap and more light localization within
the ultrathin film (Figure S4, Supporting Information), a decrease
in the coupling strength is also caused by the increased proportion
of the out-of-plane E-fields which does not couple with the in-plane
transition dipole moment of the monolayer WS2 excitons. The
strong coupling regimes of our results are evaluated by the two
criteria described briefly in the previous section. Using the square
dimer represents the largest Rabi splitting in this work, where
γ+ = 70.6 meV, γ− = 67.5 meV, γ0 = 64 meV, γpl = 247.8 meV,
Ω = 138 meV, we found that Ω − (γ+ + γ−)/2 ≅ 68.9 meV and
|Ω/(γpl − γ0)| ≅ 1.5, indicating that our system satisfies strong
coupling condition. In the presence of 2 nm Al2O3 film, where
γ+ = 65.7 meV, γ− = 57.9 meV, Ω = 97 meV, we found that
Ω − (γ+ + γ−)/2 ≅ 27.95 meV and |Ω/(γpl − γ0)| ≅ 1.05. This suggests
that our system is still in strong coupling region, albeit close to the
boundary between weak and strong coupling.
of the WS2 film (Figure 4a, red curve). As shown in Figure 4b,
we note that the spectral dip in the PL response coincides with
that in the dark field spectra, but with upper polariton emission
(ω+) much less pronounced than the lower one (ω−). We found
that the splitting in the PL response is narrower than the dark
field scattering counterparts, which appears to be in agreement
with other works that investigated PL splitting in J-aggregates
coupled to silver nanoprisms.[62] Our numerical calculations
also yield a smaller splitting in the absorption cross section
than in the scattering one (Figure S6, Supplementary Informa-
tion). The more pronounced PL peak in the PL splitting may
have been caused by the quality of the monolayer WS2 and
the way it couples with the gold nanoantenna. We investigated
this using mechanically exfoliated monolayer WS2 transferred
onto gold nanoantenna array (Figure S5, Supporting Informa-
tion), and found that the PL response exhibits splitting but with
more pronounced higher energy emission, contrary to what we
observed in Figure 4a. However, the detailed mechanisms for
the more pronounced PL peaks remains unclear as there are
still many factors influencing the PL response of a strongly cou-
pled system.[62,63]
Other fingerprints of the plasmon–exciton interaction can
be found in the photoluminescence (PL) response of the mon-
olayer WS2 coupled with square dimers of s = 80 nm (blue) and
s = 100 nm (green). Although the mode splitting associated
with strong coupling can be clearly observed in the dark field
scattering results, we observe no splitting in the PL response.
However, we observed a red shift in the PL emissions accom-
panied with an increasing shoulder below 620 nm in the WS2-
dimer system (Figure 4a). To further understand this, we plot
the PL contribution from the dimer structure, PLdimer = PLdimer
− WS2 − PLWS2, obtained by subtracting the PL response with that
4. Conclusion
Strong plasmon–exciton interaction in monolayer WS2 coupled
with gold nanoantenna arrays have been investigated. Bowtie
and square dimer antenna of sub-10 nm gaps are employed in
this study for their high Q/V property which stems from their
strongly confined E-fields across nanogaps. We coupled the
exciton with two different modes excited under bright field and
dark field geometries, and characterize anticrossing behavior with
Rabi splitting as large as 138 meV for gold square dimer arrays.
The role of the E-field enhancement in the coupling strength is
©
Adv. Optical Mater. 2019, 1901002
1901002 (5 of 8)
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advopticalmat.de
Figure 4. Photoluminescence of WS2-dimer system. a) PL responses of uncoupled monolayer WS2 on SiO2/Si substrate (red curve), and monolayer
WS2 coupled with square dimers with s = 80 nm, g = 30 nm (blue curve) and s = 100 nm, g = 30 nm (green curve). b) The PL response of the dimer
structure in log-scale, in comparison with the dark fields scattering response of square dimer with s = 80 nm, g = 30 nm.
focused onto the sample via 50× objective lens (NA = 0.55) under
bright-field geometry. The signals were directed to the hyperspectral
imaging system, while the piezo-controlled stage is scanned in order
to obtain the reflection mapping. The reflectance was obtained by
normalizing the reflection signals to the reflection of gold film of the
same thickness. The scattering measurements followed the same
routine but in dark-field geometry. The scattering was normalized to
the scattered signals from the substrate. In all these measurements,
polarizers were placed after the source and before the hyperspectral
imaging system in order to eliminate polarization mixing. The micro-
photoluminescence and Raman measurements were conducted
by WITec system under confocal geometry, based on 532 nm laser
focused by 100× objective lens (NA = 0.8). The grating was set as
600 g mm−1 (for PL measurement) and 1800 g mm−1 (for Raman
measurement).
studied by eliminating the near-field and far-field couplings in the
sparse nanoantenna arrays, which significantly decreases the local
E-field, leading to narrower Rabi splitting. The role of the spatial
mode overlap is investigated through inserting ultrathin dielec-
tric spacers between nanoantenna and monolayer WS2, where
a 20–30% decrease in the coupling strength is observed when a
2 nm thick Al2O3 spacer is introduced. Using the criteria for the
strong coupling, the WS2-dimer system is justified to be in the
strong coupling region, even with a 2 nm dielectric spacer. This is
mainly attributed to the fact that the in-plane transition dipole of
the monolayer WS2 aligns with the strong in-plane E-fields of the
nanoantenna confined within nanoscale gaps. The signature of
the strong coupling is also characterized in the PL measurements
of our WS2-dimer system, with a more pronounced lower energy
emission. The cause for such asymmetrical PL splitting remains
unclear, and future investigations are still needed to understand
the PL splitting properties of strongly coupled system.
Numerical Calculations: The full-wave simulations were done by a
commercial finite-difference time-domain software (FDTD Solutions,
Lumerical Inc) based on normal planewave excitation and periodic
boundary conditions in x and y directions. The permittivities of the
gold, silicon, and Al2O3 were all taken from the software database,
with the Johnson-Christy model adopted for the permittivity of gold.
The monolayer WS2 was modelled as a 0.618 nm thick dielectric
layer, whose permittivity modelled as Lorentzian oscillator[64,65]
5. Experimental Section
N
∈(E) = 1+
fk /(E2 −E2 − iγ E)
. Here, fk and γk are the oscillator
∑
k
k = 1
k
Sample Fabrication: The full-area WS2 was grown on 295 nm thick
thermal oxide on silicon substrate (from 2DSemiconductors Inc), with
the monolayer thickness properties of the WS2 confirmed by atomic force
microscopy (AFM) and Raman spectroscopy. The emission wavelength
of the CVD-grown and mechanically exfoliated WS2 were characterized
by micro-photoluminescence. The ultrathin 2 nm thick Al2O3 spacer
layer for studying the near field coupling between plasmon and WS2
was deposited by atomic layer deposition (Cambridge Nanotech). The
gold dimer and bowtie nanoantenna arrays were fabricated by electron
beam lithography (20 keV energy and 30 pA beam current), followed
by sonicated cold development in a mixture of n-pentyl acetate and
isopropanol at 6 °C temperature; and 30 nm thick gold evaporation
at 0.05 nm s−1 rate. The lift-off pattern transfer was then conducted in
n-methyl pyrrolidone for 10 min, followed by isopropanol rinse.
strength and the linewidth of the kth oscillator and Ek is the oscillation
energy that runs over the full spectral range.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
Acknowledgements
Optical Characterization: The samples were characterized in
microscope setting coupled to hyperspectral imaging system.
In the reflection measurements, mercury lamp (X-Cite 120) was
L.L. and L.Y.M.T. contributed equally to this work. This work is
supported in part by Singapore Ministry of Education under Grant No.
RG177/17, 2017-T1-001-239 (RG91/17 (S)), MOE2018-T2-2-189 (S),
©
Adv. Optical Mater. 2019, 1901002
1901002 (6 of 8)
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advopticalmat.de
MOE2016-T2-2-159, MOE2016-T2-1-128; A*Star AME programmatic
grant no. A18A7b0058, A*Star (SERC 1720700038 and SERC
A1883c0002); National Research Foundation Competitive Research
Program: NRF-CRP18-2017-02, and the Spanish MINECO under
[15] J. McKeever, A. Boca, A. D. Boozer, J. R. Buck, H. J. Kimble, Nature
2003, 425, 268.
[16] D. G. Baranov, M. Wersäll, J. Cuadra, T. J. Antosiewicz, T. Shegai,
ACS Photonics 2018, 5, 24.
[17] B. Lee, J. Park, G. H. Han, H. S. Ee, C. H. Naylor, W. Liu,
A. T. C. Johnson, R. Agarwal, Nano Lett. 2015, 15, 3646.
[18] K. Børkje, A. Nunnenkamp, J. D. Teufel, S. M. Girvin, Phys. Rev. Lett.
2013, 111, 053603.
[19] A. S. Mishchenko, N. Nagaosa, Phys. Rev. Lett. 2004, 93, 036402.
[20] X. Liu, T. Galfsky, Z. Sun, F. Xia, E. C. Lin, Y. H. Lee, S. Kéna-Cohen,
V. M. Menon, Nat. Photonics 2015, 9, 30.
[21] R. F. Ribeiro, L. A. Martínez-Martínez, M. Du, J. Campos-Gonzalez-
Angulo, J. Yuen-Zhou, Chem. Sci. 2018, 9, 6325.
[22] V. A. Shahnazaryan, V. A. Saroka, I. A. Shelykh, W. L. Barnes,
M. E. Portnoi, ACS Photonics 2019, 6, 904.
[23] E. Cohen-Hoshen, G. W. Bryant, I. Pinkas, J. Sperling, I. Bar-Joseph,
Nano Lett. 2012, 12, 4260.
contracts
MAT2014-53432-C5-5-R and FIS2015-64951-R. Author
contributions: Y.L., D.H.Z., and Q.J.W. supervised the project. Y.L.
and F.J.G.V. conceived the idea, while L.L. and L.Y.M.T. designed the
experiments. L.L. characterized the bright-field and dark-field responses
of the WS2-nanoantenna systems and performed the numerical analysis.
L.Y.M.T. fabricated the gold bowtie and square dimer antenna arrays
with sub-10 nm gaps and performed data analysis. L.L. performed the
photoluminescence and Raman scattering measurements of the CVD-
grown and mechanically exfoliated WS2. J.T. performed atomic layer
deposition of Al2O3, while X.Y. mechanically exfoliated and transferred
monolayer WS2 on the sample. B.Q. fabricated and performed SEM
inspection of gold nanoantenna arrays coupled with exfoliated WS2. L.L.
and L.Y.M.T. wrote the manuscript with the inputs from all authors. All
authors discussed the results and commented on the manuscript.
[24] G. Khitrova, A. Scherer, C. Ell, H. M. Gibbs, J. Hendrickson,
T. Yoshie, O. B. Shchekin, D. G. Deppe, G. Rupper, Nature 2004,
432, 200.
[25] A. Laucht, F. Hofbauer, N. Hauke, J. Angele, S. Stobbe, M. Kaniber,
G. Böhm, P. Lodahl, M. C. Amann, J. J. Finley, New J. Phys. 2009,
11, 023034.
Conflict of Interest
The authors declare no conflict of interest.
[26] I. J. Luxmoore, R. Toro, O. Del Pozo-Zamudio, N. A. Wasley,
E. A. Chekhovich, A. M. Sanchez, R. Beanland, A. M. Fox,
M. S. Skolnick, H. Y. Liu, A. I. Tartakovskii, Sci. Rep. 2013, 3, 20.
[27] M. Cai, O. Painter, K. J. Vahala, Phys. Rev. Lett. 2000, 85, 74.
[28] J. Dintinger, S. Klein, F. Bustos, W. L. Barnes, T. W. Ebbesen, Phys.
Rev. B 2005, 71, 035424.
Keywords
localized surface plasmons, photoluminescence, polariton, strong
coupling, WS2
[29] J. Dintinger, S. Klein, T. W. Ebbesen, Adv. Mater. 2006, 18, 1267.
[30] Y. Sugawara, T. A. Kelf, J. J. Baumberg, M. E. Abdelsalam,
P. N. Bartlett, Phys. Rev. Lett. 2006, 97, 266808.
[31] G. A. Wurtz, P. R. Evans, W. Hendren, R. Atkinson, W. Dickson,
R. J. Pollard, A. V Zayats, W. Harrison, C. Bower, Nano Lett. 2007,
7, 1297.
Received: June 14, 2019
Revised: August 19, 2019
Published online:
[32] J. Bellessa, C. Symonds, K. Vynck, A. Lemaitre, A. Brioude, L. Beaur,
J. C. Plenet, P. Viste, D. Felbacq, E. Cambril, P. Valvin, Phys. Rev. B
2009, 80, 033303.
[33] I. Pockrand, J. D. Swalen, R. Santo, A. Brillante, M. R. Philpott,
J. Chem. Phys. 1978, 69, 4001.
[34] R. J. Moerland, H. T. Rekola, Nano Lett. 2014, 14, 1721.
[35] T. K. Hakala, J. J. Toppari, A. Kuzyk, M. Pettersson, H. Tikkanen,
H. Kunttu, P. To, Phys. Rev. Lett. 2009, 103, 053602.
[36] D. E. Gómez, K. C. Vernon, P. Mulvaney, T. J. Davis, Nano Lett.
2010, 10, 274.
[1] K. Huang, Proc. R. Soc. London, Ser. A 1951, 208, 352.
[2] J. J. Hopfield, Phys. Rev. 1958, 112, 1555.
[3] Y. J. Chen, E. Burstein, D. L. Mills, Phys. Rev. Lett. 1975, 34, 1516.
[4] Y. Yamamoto, H. Deng, H. Haug, Bose-Einstein Condensation of
Exciton-Polaritons 2010.
[5] H. Deng, H. Haug, Y. Yamamoto, Rev. Mod. Phys. 2010, 82, 1489.
[6] H. Deng, G. Weihs, C. Santori, J. Bloch, Y. Yamamoto, Science 2002,
298, 199.
[7] T. Byrnes, N. Y. Kim, Y. Yamamoto, Nat. Phys. 2014, 10, 803.
[8] L. Dominici, G. Dagvadorj, J. M. Fellows, D. Ballarini, M. De Giorgi,
F. M. Marchetti, B. Piccirillo, L. Marrucci, A. Bramati, G. Gigli,
M. H. Szyman´ska, D. Sanvitto, Sci. Adv. 2015, 1, e1500807.
[9] L. Chang, X. Jiang, S. Hua, C. Yang, J. Wen, L. Jiang, G. Li, G. Wang,
M. Xiao, Nat. Photonics 2014, 8, 524.
[10] R. El-Ganainy, K. G. Makris, M. Khajavikhan, Z. H. Musslimani,
S. Rotter, D. N. Christodoulides, Nat. Phys. 2018, 14, 11.
[11] D. Ballarini, M. De Giorgi, E. Cancellieri, R. Houdre, E. Giacobino,
R. Cingolani, A. Bramati, G. Gigli, D. Sanvitto, in 2013 Conf. Lasers
Electro-Optics Eur. Int. Quantum Electron. Conf. CLEO/Europe-IQEC
2013IEEE, Munich, Germany 2013.
[12] A. V. Zasedatelev, A. V. Baranikov, D. Urbonas, F. Scafirimuto,
U. Scherf, T. Stöferle, R. F. Mahrt, P. G. Lagoudakis, Nat. Photonics
2019, 13, 378.
[13] T. Gao, P. S. Eldridge, T. C. H. Liew, S. I. Tsintzos, G. Stavrinidis,
G. Deligeorgis, Z. Hatzopoulos, P. G. Savvidis, Phys. Rev. B 2012,
85, 235102.
[37] D. E. Gómez, K. C. Vernon, P. Mulvaney, T. J. Davis, Appl. Phys. Lett.
2010, 96, 073108.
[38] K. F. Mak, J. Shan, Nat. Photonics 2016, 10, 216.
[39] X. Duan, C. Wang, A. Pan, R. Yu, X. Duan, Chem. Soc. Rev. 2015, 44,
8859.
[40] M. E. Kleemann, R. Chikkaraddy, E. M. Alexeev, D. Kos, C. Carnegie,
W. Deacon, A. C. De Pury, C. Große, B. De Nijs, J. Mertens,
A. I. Tartakovskii, J. J. Baumberg, Nat. Commun. 2017, 8, 1296.
[41] J. Wen, H. Wang, W. Wang, Z. Deng, C. Zhuang, Y. Zhang, F. Liu,
J. She, J. Chen, H. Chen, S. Deng, N. Xu, Nano Lett. 2017, 17, 4689.
[42] S. Lepeshov, M. Wang, A. Krasnok, O. Kotov, T. Zhang, H. Liu,
T. Jiang, B. Korgel, M. Terrones, Y. Zheng, A. Alú, ACS Appl. Mater.
Interfaces 2018, 10, 16690.
[43] M. Stührenberg, B. Munkhbat, D. G. Baranov, J. Cuadra,
A. B. Yankovich, T. J. Antosiewicz, E. Olsson, T. Shegai, Nano Lett.
2018, 18, 5938.
[44] M. Geisler, X. Cui, J. Wang, T. Rindzevicius, L. Gammelgaard,
B. S. Jessen, P. A. D. Gonçalves, F. Todisco, P. Bøggild, A. Boisen,
M. Wubs, N. A. Mortensen, S. Xiao, N. Stenger, ACS Photonics
2019, 6, 994.
[14] T. G. Tiecke, J. D. Thompson, N. P. De Leon, L. R. Liu, V. Vuletic´,
M. D. Lukin, Nature 2014, 508, 241.
©
Adv. Optical Mater. 2019, 1901002
1901002 (7 of 8)
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advopticalmat.de
[45] W. Liu, B. Lee, C. H. Naylor, H. S. Ee, J. Park, A. T. C. Johnson,
R. Agarwal, Nano Lett. 2016, 16, 1262.
[46] Z. Li, Y. Li, T. Han, X. Wang, Y. Yu, B. Tay, Z. Liu, Z. Fang, ACS Nano
2017, 11, 1165.
[47] S. Zu, B. Li, Y. Gong, Z. Li, P. M. Ajayan, Z. Fang, Adv. Opt. Mater.
2016, 4, 1463.
[48] S. Wang, S. Li, T. Chervy, A. Shalabney, S. Azzini, E. Orgiu,
J. A. Hutchison, C. Genet, P. Samorì, T. W. Ebbesen, Nano Lett.
2016, 16, 4368.
[56] S. R.-K. Rodriguez, Eur. J. Phys. 2016, 37, 025802.
[57] Y. Ding, Y. Wang, J. Ni, L. Shi, S. Shi, W. Tang, Phys. B 2011, 406,
2254.
[58] H. Zeng, G. Liu, J. Dai, Y. Yan, B. Zhu, R. He, L. Xie, S. Xu, X. Chen,
W. Yao, X. Cui, Sci. Rep. 2013, 3, 1608.
[59] R. Esteban, J. Aizpurua, G. W. Bryant, New J. Phys. 2014, 16,
013052.
[60] B. Lee, J. Park, G. H. Han, H. Ee, C. H. Naylor, W. Liu,
A. T. C. Johnson, R. Agarwal, Nano Lett. 2015, 15, 3646.
[61] A. Darweesh, S. Bauman, D. Debu, J. Herzog, Nanomaterials 2018,
8, 809.
[49] A. Kinkhabwala, Z. Yu, S. Fan, Y. Avlasevich, K. Müllen,
W. E. Moerner, Nat. Photonics 2009, 3, 654.
[50] K. Santhosh, O. Bitton, L. Chuntonov, G. Haran, Nat. Commun.
2016, 7, ncomms11823.
[51] R. Q. Li, D. Hernángomez-Pérez, F. J. García-Vidal, A. I. Fernández-
Domínguez, Phys. Rev. Lett. 2016, 117, 107401.
[52] S. J. D. Phoenix, P. L. Knight, Phys. Rev. A 1991, 44, 6023.
[53] B. W. Shore, P. L. Knight, J. Mod. Opt. 1993, 40, 1195.
[54] G. Zengin, M. Wersäll, S. Nilsson, T. J. Antosiewicz, M. Käll,
T. Shegai, Phys. Ref. Lett. 2015, 114, 157401.
[62] M. Wersä, J. Cuadra, T. J. Antosiewicz, S. Balci, T. Shegai,
M. Wersäll, J. Cuadra, T. J. Antosiewicz, S. Balci, T. Shegai, Nano
Lett. 2017, 17, 551.
[63] X. Liu, T. Galfsky, Z. Sun, F. Xia, E. Lin, Y.-H. Lee, S. Kéna-Cohen,
V. M. Menon, Nat. Photonics 2015, 9, 30.
[64] Y. Li, A. Chernikov, X. Zhang, A. Rigosi, H. M. Hill, A. M. Van Der
Zande, D. A. Chenet, E. Shih, J. Hone, T. F. Heinz, Phys. Rev. B
2014, 90, 205422.
[55] R. Liu, Z. K. Zhou, Y. C. Yu, T. Zhang, H. Wang, G. Liu, Y. Wei,
H. Chen, X. H. Wang, Phys. Rev. Lett. 2017, 118, 237401.
[65] A. Laturia, M. L. Van de Put, W. G. Vandenberghe, npj 2D Mater.
Appl. 2018, 2, 6.
©
Adv. Optical Mater. 2019, 1901002
1901002 (8 of 8)
2019 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim