Room-temperature Tamm-plasmon exciton-polaritons with a WSe2 monolayer

Abstract

Solid-state cavity quantum electrodynamics is a rapidly advancing field, which explores the frontiers of light-matter coupling. Metal-based approaches are of particular interest in this field, as they carry the potential to squeeze optical modes to spaces significantly below the diffraction limit. Transition metal dichalcogenides are ideally suited as the active material in cavity quantum electrodynamics, as they interact strongly with light at the ultimate monolayer limit. Here, we implement a Tamm-plasmon-polariton structure and study the coupling to a monolayer of WSe₂ , hosting highly stable excitons. Exciton-polariton formation at room temperature is manifested in the characteristic energy-momentum dispersion relation studied in photoluminescence, featuring an anti-crossing between the exciton and photon modes with a Rabi-splitting of 23.5 meV. Creating polaritonic quasiparticles in monolithic, compact architectures with atomic monolayers under ambient conditions is a crucial step towards the exploration of nonlinearities, macroscopic coherence and advanced spinor physics with novel, low-mass bosons.

ARTICLE

Received 10 Apr 2016 | Accepted 21 Sep 2016 | Published 31 Oct 2016

Nils Lundt1, Sebastian Klembt1, Evgeniia Cherotchenko2, Simon Betzold1, Oliver Iff1, Anton V. Nalitov2, Martin Klaas1, Christof P. Dietrich1, Alexey V. Kavokin2,3, Sven Hing1,4 & Christian Schneider1

Solid-state cavity quantum electrodynamics is a rapidly advancing eld, which explores the frontiers of lightmatter coupling. Metal-based approaches are of particular interest in this eld, as they carry the potential to squeeze optical modes to spaces signicantly below the diffraction limit. Transition metal dichalcogenides are ideally suited as the active material in cavity quantum electrodynamics, as they interact strongly with light at the ultimate monolayer limit. Here, we implement a Tamm-plasmon-polariton structure and study the coupling to a monolayer of WSe2, hosting highly stable excitons. Exciton-polariton formation at room temperature is manifested in the characteristic energymomentum dispersion relation studied in photoluminescence, featuring an anti-crossing between the exciton and photon modes with a Rabi-splitting of 23.5 meV. Creating polaritonic quasiparticles in monolithic, compact architectures with atomic monolayers under ambient conditions is a crucial step towards the exploration of nonlinearities, macroscopic coherence and advanced spinor physics with novel, low-mass bosons.

DOI: 10.1038/ncomms13328 OPEN

Room-temperature Tamm-plasmon exciton-polaritons with a WSe2 monolayer

1 Technische Physik and Wilhelm-Conrad-Rntgen Research Center for Complex Material Systems, Universitat Wrzburg, Am Hubland, D-97074 Wrzburg, Germany. 2 Physics and Astronomy School, University of Southampton, Higheld, Southampton SO171BJ, UK. 3 SPIN-CNR, Viale del Politecnico 1, I-00133 Rome, Italy. 4 SUPA, School of Physics and Astronomy, University of St Andrews, St Andrews KY 16 9SS, UK. Correspondence and requests for materials should be addressed to N.L. (email: mailto:[email protected]

Web End [email protected] ) or to C.S. (email: mailto:[email protected]

Web End [email protected] ).

NATURE COMMUNICATIONS | 7:13328 | DOI: 10.1038/ncomms13328 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 1

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13328

With the rst reports of atomic monolayer materials exfoliated from a graphite block, a genuine breakthrough in physics was triggered. However, from

the point of view of opto-electronic applications, pristine graphene has its limitations, as it does not have a (direct) bandgap. Two-dimensional atomic crystals of transition metal dichalcogenides (TMDCs), compounds of a MX2 stoichiometry (M being a transition metal, X a chalcogenide), seem to be much more promising16, as monolayers of some TMDCs have a direct bandgap on the order of 1.62.1 eV2,7. Furthermore, the combination of large exciton-binding energies up to 550 meV2,8, large oscillator strength, the possible absence of structural disorder and intriguing spinor and polarization properties912 has recently placed sheets of TMDCs in the focus of solid-state cavity quantum electrodynamics and polaritonics. Polariton formation can be observed in the strong lightmatter coupling regime, which becomes accessible in high quality, or ultra-compact photonic structures, such as dielectric microcavities or plasmonic architectures with embedded emitters comprising large oscillator strengths13,14. Once the lightmatter coupling strength in such a system exceeds dissipation and dephasing, the hybridization of light and matter excitations leads to the formation of exciton-polaritons13. These composite quasi-particles have very appealing physical properties. Polaritons can travel over macroscopic distances at high speed (B1% of the speed of light15) and, due to the inherited matter component, interactions between polaritons are notable. This puts them in the focus of nonlinear optics, collective bosonic phenomena and integrated photonics. Exciton-polaritons are bosons with a very low (and tailorable) effective mass and are therefore almost ideal candidates to study BoseEinstein condensation phenomena at elevated temperatures. A serious drawback for this eld, which circumvents a more efcient exploitation of polaritons, are the limited thermal stability of excitons in most of the III/V materials, strong disorder and defects in IIVI multilayer structures and exciton bleaching in organic polymers. TMDCs monolayers have the potential to overcome these drawbacks and are therefore considered as a highly promising material platform for light matter interaction experiments1.

Recently, the physics of strong lightmatter coupling between a single ake of MoS2 and a cavity resonance in a FabryPerot resonator structure was discussed16. However, the comparably broad photoluminescence (PL) emission of the monolayer used for these ndings (60 meV16), which was grown by chemical

vapour deposition, render the unambiguous identication of the full characteristic polariton dispersion relation, in particular in non-resonant PL experiments challenging. Unambiguous polariton formation with a single monolayer of MoSe2 was subsequently demonstrated at cryogenic temperatures17, enabled by its narrow linewidth (11 meV at 4 K and 35 meV at 300 K). Exfoliated WSe2 monolayers exhibit comparable linewidths and have a strongly enhanced luminescence yield under ambient conditions18, suggesting their suitability for room-temperature polaritonics. However, not even at cryogenic temperatures, strong coupling has been demonstrated in WSe2 monolayers, yet.

Here, to demonstrate strong coupling at ambient conditions, we have embedded a WSe2 monolayer in a compact

Tamm-plasmon photonic microstructure19,20 composed of a dielectric distributed Bragg reector (DBR), a polymer layer and a thin gold cap. We map out the characteristic energymomentum dispersion relations of the upper and the lower polariton branch at ambient conditions by angle-resolved PL and reection measurements. Our experimental ndings are supported by modelling our device in a coupled oscillator framework, showing an excellent agreement between theory and experiment.

ResultsDevice fabrication. Figure 1a depicts a graphic illustration of our investigated device: it consists of a SiO2/TiO2 DBR (ten pairs), which supports a very high reectivity of 99.97% in a wide spectral range between 580 and 780 nm. A single layer of WSe2,

mechanically exfoliated via commercial adhesive tape (Tesa brand) from a bulk crystal was transferred onto the top SiO2 layer with a polymer stamp. The monolayer was identied by PL measurements, which also conrmed its excellent optical quality under ambient conditions (see Fig. 1b). Here we observe the characteristic prole from the A-valley exciton with a linewidth of37.5 meV. The monolayer was capped by a 130 nm-thick layer of poly(methyl methacrylate) (PMMA) and the device was completed by a 35 nm-thick gold layer (see Supplementary Note 1 and Supplementary Figs 1 and 2 for details on the inuence of PMMA on the emission properties of the monolayer). The layer thicknesses were designed to support a Tamm-plasmon resonance at the energy of the room-temperature emission energy of the A-valley exciton (1.650 eV). Figure 1c shows the vertical optical mode prole obtained by a transfer matrix calculation, the

a b c

Energy (eV)

1.75 1.7 1.65 2.52.01.51.00.50.0

1.6 1.55 1.5

780 800 820 840

Field distribution (arb. units)

Refractive index

Au

Intensity (arb. unit)

PMMA

WSe2

SiO2

TiO2

1.0

Reflectivity

0.8

0.6

0.4

0.2

0.0

1.56 1.60 1.64 1.68 Energy (eV)

700 720 740 760

0.0 0.5 1.0 1.5 2.0

Wavelength (nm) Position (m)

Figure 1 | Tamm-monolayer device. (a) Schematic illustration of the Tamm-plasmon device with the embedded WSe2 monolayer. The monolayer is capped with PMMA, whose thickness primarily determines the frequency of the devices optical resonance. (b) PL spectrum of the WSe2 monolayer before capping, recorded under ambient conditions. The dominant emission is identied to stem from the A-valley exciton. Inset: false-colour optical microscopy image of the used WSe2 ake (monolayer in red shaded area; scale bar, 20 mm). (c) Calculation of the electromagnetic eld intensity in the heterostructure and the optical resonance (inset). The Tamm-plasmon features a strongly enhanced eld maximum close to the surface of the structure, which coincides with the vertical position of the monolayer in the device. The optical resonance features a quality factor on the order of 110.

2 NATURE COMMUNICATIONS | 7:13328 | DOI: 10.1038/ncomms13328 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13328 ARTICLE

corresponding refractive indices of the layer sequence and the resulting reectivity spectrum (without embedded monolayer). The successful implementation of this device was conrmed by reectivity measurements (see Supplementary Note 2 and Supplementary Fig. 3). This type of photonic microstructure features a strong eld enhancement close to the metallic interface, which has proven to sufce for promoting polariton formation with embedded InGaAs-21, GaAs-22 and II/VI-23 based quantum well structures at cryogenic temperatures. We point out that the large refractive index difference in the dielectric Bragg reector leads to Tamm resonances, which provide a very strong eld connement, yet quality factors on the order of 110 can be obtained, as can be seen from the simulated reectivity spectrum in Fig. 1c.

Optical characterization. In the following, we will discuss the case of a device, which contains a single layer of WSe2 embedded in the resonant structure: we employ the characteristic energymomentum dispersion relation of the vertically conned photon eld in the Tamm device, to map out the coupling of the A-valley exciton of the WSe2 monolayer and the Tammplasmon-polariton. To cover a large emission angle, thus accessing a sizeable spectral tuning range, we use a high magnication ( 100) microscope objective with a numerical aperture

of 0.7 in the PL experiment. As the polariton in-plane momentum k|| is proportional to sin(y), with y being the PL emission angle, this allows us to project a momentum range of up to 4.2 mm 1 onto the CCD (charge-coupled device) chip of our spectrometer in the far-eld imaging conguration (see Methods section for further details). The sample is held at 300 K and the embedded monolayer is excited via a non-resonant continuous wave laser at a wavelength of 532 nm at an excitation power of 3 mW, measured in front of the microscope objective. In Fig. 2a, we plot the PL spectra extracted from our device at various in-plane momenta. At an in-plane momentum of 1.84 mm 1 (corresponding to an emission angle of 12.67), we can observe a minimum peak distance between the two prominent features, which we identify as the lower and upper polariton branch. These two branches feature the characteristic anti-crossing behaviour with a Rabi splitting of 23.5 meV, the key signature of the strong coupling regime. We note that the strong coupling regime is primarily a result of the tight mode connement provided by the Tamm structure24 (see Supplementary Note 3 and Supplementary Fig. 4 for details). Figure 2b depicts the fully mapped out energymomentum dispersion relation of the two polariton resonances by plotting the corresponding peak energies as a function of the in-plane momentum. As expected from two coupled oscillators with strongly varying effective masses, we observe the characteristic potential minimum in the lower polariton branch with a modest negative detuning of D EC EX 11.7 meV. This negative detuning condition

leads to an effective polariton mass of 1.45 10 5 me at the

bottom of the lower polariton, where me is the free electron mass. We can furthermore observe the characteristic transition from a light particle close to k|| 0 to a heavy, exciton-like

particle at large k|| values. The corresponding Hopeld coefcients, which characterize the excitonic and photonic fraction of the lower polariton (|X|2 versus |C|2, respectively) are plotted as a function of the in-plane momentum in Fig. 2c. The potential minimum, which is formed in the lower polariton branch, is another key signature of an exciton polariton in the presence of vertically conned eld. It furthermore provides a well-dened nal energy state with a distinct effective mass, which is crucial for advanced parametric and stimulated scattering experiments25. A key advantage of exciton

polaritons, as compared with other composite bosons (such as excitons), is the possibility to conveniently tune the depth of this attractive potential, and simultaneously the particles effective masses and light-versus-matter composition by changing the detuning between the light and the matter oscillator. In addition, we carried out reectivity measurements to provide further evidence that our device works in the strong coupling regime. The results are presented in Fig. 2d and were analysed and tted the same way as for the PL data. Owing to experimental limitations (see Supplementary Note 4) we have realized a white light reectivity measurement with a 20

magnication objective with a numerical aperture of 0.40. Similarly, compared with the PL experiment, we observe a clear appearance of two normal modes that can be well described by a coupled oscillator model. For the sake of clarity, we have inverted the reectivity spectra; thus, the reection dips appear as positive signals in the graph. In Fig. 2e, we plot the extracted values of the reection resonances as a function of the in-plane momentum, which allows us to reconstruct the polariton dispersion relation. The dispersion features the characteristic avoided crossing behaviour with a Rabi splitting of 14.7 meV. The somewhat reduced Rabi splitting measured in reectivity is expected for a photonic structure with a modest quality factor and the slightly shifted resonance of the excitonic oscillator is most likely to be resulting from sample ageing, which is both detailed in the Supplementary Note 4.

Modelling. To interpret our experimental data, we can t the dispersions with a coupled oscillator model:

Eph i Gph O=2

O=2 Eex i Gex D

1

where Eph and Eex are photon and exciton energies, respectively, D is the detuning between the two modes, and Gph and Gex are photon and exciton mode broadening, respectively. The eigenvectors represent the weighting coefcients of exciton and photon fraction and :O represents the Rabi splitting in the system. Solving the dispersion equation:

det Eph i Gph E O=2

O=2 Eex i Gex D E

a

b

E

a

b

0 2

one can obtain two polariton branches. The result of this modelling is shown in Fig. 2b,e (solid lines) along with the experimental data (symbols). The tting was carried out via solving the optimization problem with detuning, Rabi splitting and photon mass used as parameters. As the exciton mass is several orders of magnitude larger than the photon mass, it does not affect the result of the simulation and its value is taken to be0.8 me, as dened in ref. 26. The dashed lines show photon and exciton energies as a function of the in plane wave vector k||.

For details on the tting procedure of momentum-resolved spectra see Supplementary Note 5 and Supplementary Figs 5 and 6.

We will now address the occupation of the polariton states in our device, operated under ambient conditions. The overall, momentum-resolved PL spectrum of the structure is plotted in Fig. 3a. In stark contrast to previous reports discussing polariton emission with TMDC materials at room temperature16, we observe a pronounced occupation of the low-energy states in the lower polartion branch and a reduced occupation of the excited polariton states. The following model was used to analyse the luminescence experiment: in a rst approximation, owing to the comparably low particle numbers and high temperatures, we assume a Boltzmann distribution law for our particles: NiBexp( Ei/kBT), where Ni and Ei denote i-state population

a

b

NATURE COMMUNICATIONS | 7:13328 | DOI: 10.1038/ncomms13328 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 3

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13328

a d

Photoluminescence Reflectivity

Norm. intensity (arb. unit)

Norm. intensity (arb. unit)

k = 3.5 m

k = 2.5 m

k = 1.5 m

k = 0.5 m

k = 3.0 m1

k = 2.0 m

k = 1.0 m

k = 0.0 m

k = 2.0 m

k = 1.6 m

k = 1.2 m

k = 0.8 m

k = 0.4 m

k = 0.0 m

1.60 1.64 1.68 1.72 1.76

1.62 1.64 1.66 1.68 1.70

Energy (eV)

Energy (eV)

b

e

UPB

Energy (eV)

Fractions LPB

1.71

1.70

1.69

1.68

1.67

1.66

1.65

1.64

1.63

C

X

Energy (eV)

1.67

1.66

1.65

1.64

1.63

LPB

4 2 0 2 4

In plane wave vector (m1)

In plane wave vector (m1)

In plane wave vector (m1)

c

1.00.80.60.40.20.0

|C|2

|X|2

4 2 0 2 4

3 2 1 0 1 2 3

Figure 2 | Exciton-polariton formation with Tamm-plasmons. (a) PL spectra recorded from the coupled device at room temperature at various in-plane momenta (depicted in a waterfall representation). Two pronounced resonances evolve in the system, which feature the characteristic anti-crossing behaviour of exciton-polaritons. (b) Energymomentum dispersion relation of the lower and upper polariton branch at room temperature: the polariton energies are extracted by tting spectra at various in-plane momenta (solid symbols). A coupled oscillator approach is employed to t the data and to demonstrate excellent agreement between experiment and theory (lines). (c) Plot of the exciton and photon fraction of the lower polariton branch as a function of the in-plane momentum extracted from coupled oscillator t. (d) Inverted reectivity spectra at different in-plane momenta. (e) Energy momentum dispersion relation extracted from the reectivity spectra.

1.72

1.70

1.68

1.66

1.64

1.62

4 3 2 1 0 1 2 3 4

In plane wave vector (m1)

a b

Energy (eV)

Intensity (arb. units)

0.0 0.5 1.0 1.5 2.0 2.5

Energy (eV)

1.72

1.70

1.68

1.66

1.64

1.62

4 3 2 1 0 1 2 3 4

In plane wave vector (m1)

Intensity (arb. units)

0.0 0.5 1.0 1.5 2.0 2.5

Experiment Theory

Figure 3 | Experimental and theoretical polariton dispersion relations in the studied Tamm structure. (a) Room-temperature false colour intensity prole of the full polariton dispersion relation extracted from the PL measurements. (b) Model of the full dispersion by assuming a Boltzmann distribution of the quasiparticles with an effective temperature of 300 K.

and energy, and kB is the Boltzmann constant. The modelled PL is thus generated by a polariton gas at room temperature (T 300 K). We further assume that the emission stems from

the photonic mode only and is broadened in energy according to a Lorentz distribution. This allows us to relate the PL intensity to

the photonic Hopeld coefcients via:

I k; E

X

i

Ciph

2 G2ph

3

=kBT

E Ei k

2 exp Ei k

4 NATURE COMMUNICATIONS | 7:13328 | DOI: 10.1038/ncomms13328 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13328 ARTICLE

where Gph is the broadening of the photonic mode and the i-index spans over the two polariton branches. We extract the value of Gph 15 meV from the experimental data. The

experimental results and the theoretically calculated dispersion relation are plotted in Fig. 3a,b, respectively.

In fact, we achieve very good agreement between theory and experiment. Although our model is purely phenomenological and cannot account for any dynamic and microscopic effects in our system, it already serves as a rst indicator that, despite the pronounced dissipation in our system, polariton relaxation is indeed signicant. Additional simulations of our system with higher temperatures, which result in stronger luminescence from the upper polariton branch, can be found in Supplementary Note 6 and Supplementary Fig. 7 of the manuscript.

DiscussionIn conclusion, we have observed clear evidence for the formation of exciton polaritons in a hybrid dielectric and polymer Tamm-plasmon-polariton device featuring an integrated single atomic layer of the TMDC WSe2. We mapped out the distinct polariton dispersion relation in angle-resolved PL and reectivity measurements, and resolved both polariton branches including the characteristic parabolic energy minimum and the attening towards the exciton band. Our experimental data are supported by a coupled harmonic oscillator model and we achieve very good agreement both for the energy evolution of the polariton resonances and for the population of polariton eigenstates. We believe that our work represents a signicant step towards the implementation of polariton condensates and nonlinear experiments in the strong coupling regime based on single layers or stacks of several layers of TMDCs. Moreover, it will be of particular interest for TMDC polaritonic experiments to harness the unique spinor and valley physics inherited by the atomic monolayers. Combining plasmonics27,28 and two-dimensional active media in the strong lightmatter coupling regime certainly carries great potential for building new architectures of highly integrated, nonlinear optical circuits and logic devices, which are operated at ultra-low powers and close to terahertz frequencies.

Methods

Sample design and fabrication. The sample was designed based on transfer matrix calculations, where the plasmon-polariton resonance was tuned to match the A exciton resonance of WSe2 monolayer at room temperature (1.650 eV).

The bottom mirror consists of a commercially available DBR based on a fused silica substrate topped with ten pairs of TiO2/SiO2 layers (72/117 nm thickness, respectively, corresponding to a central stopband wavelength of 680 nm). The stopband ranges from 580 to 780 nm. The WSe2 monolayer was mechanically exfoliated onto a polymer gel lm (polydimethylsiloxane) and was then transferred onto the bottom DBR. One hundred and thirty nanometres of PMMA were deposited by spin coating onto the structure. Finally, a 35 nm-thick gold layer was thermally evaporated onto the sample.

Experimental setup. We took advantage of an optical setup in which both spatially (near-eld) and momentum-space (far-eld)-resolved spectroscopy and imaging are accessible. PL is collected through a 0.7 numerical aperture microscope objective lens and directed into an imaging spectrometer with 1,200 groves mm 1 grating via a set of relay lenses, projecting the proper projection plane onto the monochromators entrance slit. The systems angular resolution is B0.05 mm 1 (B0.5) and spectral resolution is B0.050 meV with a nitrogen-cooled Si-CCD as detector.

Analysis of the PL spectra. As the ne spectrometer grating (1,200groves mm 1) does not cover the full spectral range of the PL signal, the angle-resolved PL spectra were taken at three different energies with B60% overlap and were subsequently combined. This procedure used a fast Fourier transform (FFT) smoothing algorithm to account for small-intensity offsets in the overlap regions. Finally, single spectra were tted with a two-Gaussian t, to deduce the peak energy.

Data availability. The data that support the ndings of this study are available from the corresponding author upon request.

References

1. Cao, L. Two-dimensional transition-metal dichalcogenide materials: toward an age of atomic-scale photonics. MRS Bull. 40, 592599 (2015).

2. Chernikov, A. et al. Exciton binding energy and nonhydrogenic Rydberg series in monolayer WS2. Phys. Rev. Lett. 113, 076802 (2014).

3. Wang, Q. H., Kalantar-Zadeh, K., Kis, A., Coleman, J. N. & Strano, M. S. Electronics and optoelectronics of two-dimensional transition metal dichalcogenides. Nat. Nanotechnol. 7, 699712 (2012).

4. Xu, M., Liang, T., Shi, M. & Chen, H. Graphene-like two-dimensional materials. Chem. Rev. 113, 37663798 (2013).

5. Novoselov, K. S. et al. Two-dimensional atomic crystals. Proc. Natl Acad. Sci. USA 102, 1045110453 (2005).

6. Wang, Z. M. MoS2 - Materials, Physics and Devices (Springer, 2014).7. Tonndorf, P. et al. Photoluminescence emission and raman response of MoS2,

MoSe2, and WSe2 nanolayers. Opt. Express 21, 237268 (2013).8. Ugeda, M. M. et al. Giant bandgap renormalization and excitonic effects in a monolayer transition metal dichalcogenide semiconductor. Nat. Mater. 13, 10911095 (2014).

9. Xu, X., Yao, W., Xiao, D. & Heinz, T. F. Spin and pseudospins in layered transition metal dichalcogenides. Nat. Phys. 10, 343350 (2014).

10. Wang, G. et al. Polarization and time-resolved photoluminescence spectroscopy of excitons in MoSe2 monolayers. Appl. Phys. Lett. 106, 112101 (2015).

11. Wang, G. et al. Double resonant Raman scattering and valley coherence generation in monolayer WSe2. Phys. Rev. Lett. 115, 16 (2015).

12. Srivastava, A. et al. Valley Zeeman effect in elementary optical excitations of monolayer WSe2. Nat. Phys. 11, 141147 (2015).

13. Weisbuch, C., Nishioka, M., Ishikawa, A. & Arakawa, Y. Observation of the coupled exciton-photon mode splitting in a semiconductor quantum microcavity. Phys. Rev. Lett. 69, 33143317 (1992).

14. Kavokin, A. V., Baumberg, J. J., Malpuech, G. & Laussy, F. P. Microcavities (Oxford Univ. Press, 2007).

15. Amo, A. et al. Collective uid dynamics of a polariton condensate in a semiconductor microcavity. Nature 457, 291295 (2009).

16. Liu, X. et al. Strong lightmatter coupling in two-dimensional atomic crystals. Nat. Photonics 9, 3034 (2015).

17. Dufferwiel, S. et al. Exciton-polaritons in van der Waals heterostructures embedded in tunable microcavities. Nat. Commun. 6, 8579 (2015).

18. Wang, G. et al. Spin-orbit engineering in transition metal dichalcogenide alloy monolayers. Nat. Commun. 6, 19 (2015).

19. Kaliteevski, M. A., Brand, S., Chamberlain, J. M., Abram, R. A. & Nikolaev, V. V. Effect of longitudinal excitations on surface plasmons. Solid State Commun. 144, 413417 (2007).

20. Sasin, M. E. et al. Tamm plasmon polaritons: slow and spatially compact light. Appl. Phys. Lett. 92, 4851 (2008).

21. Symonds, C., Lemtre, A., Homeyer, E., Plenet, J. C. & Bellessa, J. Emission of Tamm plasmon/exciton polaritons. Appl. Phys. Lett. 95, 20072010 (2009).

22. Gessler, J. et al. Electro optical tuning of Tamm-plasmon exciton-polaritons. Appl. Phys. Lett. 105, 181107 (2014).

23. Sebald, K. et al. Tailoring the optical properties of wide-bandgap based microcavities via metal lms. Appl. Phys. Lett. 107, 062101 (2015).

24. Lundt, N. et al. Monolayered MoSe2: A candidate for room temperature polaritonics. Preprint at: arXiv:1603.05562 (2016).

25. Savvidis, P. G. et al. Angle-resonant stimulated polariton amplier. Phys. Rev. Lett. 84, 15471550 (2000).

26. Vasilevskiy, M. I., Santiago-Prez, D. G., Trallero-Giner, C., Peres, N. M. R. & Kavokin, A. Exciton-polaritons in 2D dichalcogenide layers placed in a planar microcavity: tuneable interaction between two Bose-Einstein condensates. Phys. Rev. B 92, 245435 (2015).

27. Brongersma, M. L. & Kik, P. G. Surface Plasmon Nanophotonics (Springer, 2007).

28. Barnes, W. L., Dereux, A. & Ebbesen, T. W. Surface plasmon subwavelength optics. Nature 424, 824830 (2003).

Acknowledgements

This work has been supported by the State of Bavaria. We thank S. Tongay for providing initial samples for the project. We further thank S. Stoll and I. Kim for assistance in the exfoliation and pre-characterization of the high-quality monolayer material. Moreover, we thank A. Wolf for fabrication assistance and S. Brodbeck for assistance with the transfer matrix calculations. C.S. thanks L. Worschech for encouraging him at the very early stage of the work. A.K. and S.H. acknowledge the partial nancial support from the EPSRC Hybrid Polaritonics Programme. C.S. acknowledges nancial support by the European Research Council (unLiMIt-2D project).

NATURE COMMUNICATIONS | 7:13328 | DOI: 10.1038/ncomms13328 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications 5

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms13328

Author contributions

C.S and S.H. initiated the study and guided the work. N.L., C.P.D. and C.S. designed the Tamm device. N.L. and O.I. exfoliated, identied and transferred the monolayer. N.L. fabricated the Tamm structure. N.L., S.K. and M.K. performed experiments. N.L., S.K., S.B. and C.S. analysed and interpreted the experimental data, supported by all co-authors. E.C., A.V.N. and A.V.K. provided the theory. C.S., N.L., S.K. and E.C. wrote the manuscript, with input from all co-authors.

Additional information

Supplementary Information accompanies this paper at http://www.nature.com/naturecommunications

Web End =http://www.nature.com/ http://www.nature.com/naturecommunications

Web End =naturecommunications

Competing nancial interests: The authors declare no competing nancial interests.

Reprints and permission information is available online at http://npg.nature.com/reprintsandpermissions/

Web End =http://npg.nature.com/ http://npg.nature.com/reprintsandpermissions/

Web End =reprintsandpermissions/

How to cite this article: Lundt, N. et al. Room-temperature Tamm-Plasmon exciton-polaritons with a WSe2 monolayer. Nat. Commun. 7, 13328 doi: 10.1038/ncomms13328 (2016).

Publishers note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afliations.

This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the articles Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Web End =http://creativecommons.org/licenses/by/4.0/

r The Author(s) 2016

6 NATURE COMMUNICATIONS | 7:13328 | DOI: 10.1038/ncomms13328 | http://www.nature.com/naturecommunications

Web End =www.nature.com/naturecommunications