
==== Front
Nat Commun
Nat Commun
Nature Communications
2041-1723
Nature Publishing Group UK London

51612
10.1038/s41467-024-51612-2
Article
Room temperature polariton spin switches based on Van der Waals superlattices
Zhao Jiaxin 1
http://orcid.org/0000-0002-8912-1137
Fieramosca Antonio antonio.fieramosca@gmail.com

2
http://orcid.org/0000-0001-9768-4864
Bao Ruiqi 1
Dini Kevin 1
http://orcid.org/0000-0002-2808-0327
Su Rui 13
Sanvitto Daniele 2
http://orcid.org/0000-0002-2555-4363
Xiong Qihua Qihua_xiong@tsinghua.edu.cn

4567
Liew Timothy C. H. tchliew@gmail.com

1
1 https://ror.org/02e7b5302 grid.59025.3b 0000 0001 2224 0361 Division of Physics and Applied Physics, School of Physical and Mathematical Sciences, Nanyang Technological University, Singapore, Singapore
2 grid.494551.8 0000 0004 6477 0549 CNR NANOTEC Institute of Nanotechnology, via Monteroni, Lecce, Italy
3 https://ror.org/02e7b5302 grid.59025.3b 0000 0001 2224 0361 School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore, Singapore
4 grid.12527.33 0000 0001 0662 3178 State Key Laboratory of Low-Dimensional Quantum Physics and Department of Physics, Tsinghua University, Beijing, P.R. China
5 grid.12527.33 0000 0001 0662 3178 Frontier Science Center for Quantum Information, Beijing, P.R. China
6 https://ror.org/04nqf9k60 grid.510904.9 0000 0004 9362 2406 Beijing Academy of Quantum Information Sciences, Beijing, P.R. China
7 https://ror.org/03cve4549 grid.12527.33 0000 0001 0662 3178 Beijing Innovation Center for Future Chips, Tsinghua University, Beijing, P.R. China
1 9 2024
1 9 2024
2024
15 760127 10 2023
13 8 2024
© The Author(s) 2024
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Transition-metal dichalcogenide monolayers possess large exciton binding energy and a robust valley degree of freedom, making them a viable platform for the development of spintronic devices capable of operating at room temperature. The development of such monolayer TMD-based spintronic devices requires strong spin-dependent interactions and effective spin transport. This can be achieved by employing exciton-polaritons. These hybrid light-matter states arising from the strong coupling of excitons and photons allow high-speed in-plane propagation and strong nonlinear interactions. Here, we demonstrate the operation of all-optical polariton spin switches by incorporating a WS2 superlattice into a planar microcavity. We demonstrate spin-anisotropic polariton nonlinear interactions in a WS2 superlattice at room temperature. As a proof-of-concept, we utilize these spin-dependent interactions to implement different spin switch geometries at ambient conditions, which show intrinsic sub-picosecond switching time and small footprint. Our findings offer new perspectives on manipulations of the polarization state in polaritonic systems and highlight the potential of atomically thin semiconductors for the development of next generation information processing devices.

Exciton-polaritons result from the strong coupling of excitons and photons, exhibiting strong nonlinearity. Here, Zhao et al demonstrate room-temperature optical polariton spin-switching in a tungsten disulfide superlattice.

Subject terms

Nanocavities
Nonlinear optics
Nanophotonics and plasmonics
https://doi.org/10.13039/501100001459 Ministry of Education - Singapore (MOE) MOE-T2EP50121-0020 MOE2018-T3-1-002 Liew Timothy C. H. (MOE2018- T3-1-002the Singapore Ministry of Education via AcRF Tier 2 project (MOE-T2EP50121-0020)issue-copyright-statement© Springer Nature Limited 2024
==== Body
pmcIntroduction

Carrying information in both the charge and spin of electrons allows for the realization of complex devices with a high degree of functionality. Starting from the observation of spin-dependent electron transport phenomena and the discovery of giant magnetoresistance in solid-state devices1, spin-based electronics (spintronics) has become a major research area in condensed-matter physics and information technology, leading to the realization of devices with reduced dimensionality that are now ubiquitous in modern life2. Spintronics holds significant potential for the development of next generation ultrafast and cost effective technologies3,4. The majority of spintronic applications necessitate robust spin-dependent interactions within devices, long-distance transfer of spin information, high switching efficacy, high processing speed and integrability.

To properly fulfill these requirements, the concept of spintronics with electrons has been extended to different research areas, spanning from optics5 to optoelectronics with bounded electron-hole pairs, i.e., excitons6. All-optical schemes present long-distance transport and high switching rates but weak nonlinearity. On the other hand, exciton-based optoelectronic devices have good integrability and switching efficacy but limited spin transport, hindered by the low exciton diffusion. In this respect, exciton-polaritons, i.e., hybrid light-matter excitations arising from the strong coupling of excitons and photons, have emerged as a promising platform for spintronic devices7,8. Owing to their hybrid nature, polaritons exhibit small effective mass and pronounced nonlinearities, leading to long in-plane propagation with high velocities as well as to high switching efficacy processing speed9–14. They are spin degenerate and the interaction strength between parallel and antiparallel spin states is different in GaAs-based microcavities15. This spin anisotropic interaction has enabled the realization of proof-of-concept polarization-sensitive devices7,8,16,17, as well as the demonstration of spin multistability18, which typically rely on a near negligible interaction between antiparallel spins compared to parallel spins. However, the practical implementation of cost-effective and technologically relevant polariton-based spintronic devices requires room temperature operation, which poses a significant challenge for traditional inorganic semiconductors due to their small exciton binding energy19. Moreover, efficient devices demand both miniaturization and fast operational time and most of proof-of-concept GaAs-based spintronic devices have large footprint and slow operation time7,11.

In this respect, transition metal dichalcogenides (TMDs), characterized by large binding energy and strong oscillator strength, have been proposed as a good platform to investigate many body physics and realize technological applications at room temperature. The weak van der Waals interlayer interactions in TMDs allow the exfoliation down to the monolayer (ML) level and the fabrication of complex heterostructures, as well as the integration with optical microcavities, facilitating the design and manufacture of various on-chip photonic and polaritonic devices, such as laser and light emitting diodes20. Furthermore, the broken inversion symmetry and strong spin-orbit interaction in TMDs result in pronounced optical selection rules at two energy-degenerate valleys21. When TMDs are strongly coupled with microcavities, the valley properties are preserved even at room temperature22–24. Demonstrations of the optical valley Hall effect25, valley selective optical Stark effect26, optical parametric scattering27, and valley manipulation under magnetic field28, further emphasize the significance of TMD polaritons for the development of realistic devices.

Prior studies have been focused mainly on the creation of valley polaritons using non-resonant photoluminescence and largely overlooked the polarization-dependent nonlinear optical properties in TMD polaritons, which are fundamental for the development of innovative polariton-based devices, such as optical spin switches, transistors, and logic gates. Moreover, in comparison to MLs, superlattices of TMDs, where multiple MLs are stacked together with a dielectric spacer in between, exhibit larger optical density and higher absorbance, therefore leading to larger Rabi splitting. This makes them a unique platform for the control and manipulation of optical properties in TMD-based devices29–32.

Here, we investigated the spin-dependent interactions in WS2 superlattice microcavities at room temperature, revealing a robust full spin anisotropy. The presence of spin-dependent polariton interactions is exploited to realize all-optical spin switches. We demonstrate that the ON/OFF polarization switching can be realized within 1 picosecond, in an area of around 25 μm2. Our results indicate that TMD superlattices are extremely suitable for the implementation of novel designs fully exploiting the spin anisotropy of excitonic-polaritons for the realization of ultrafast and small footprint spintronic devices.

Results

We investigated WS2 superlattices embedded in a planar dielectric microcavity, as schematically shown in Fig. 1a. The microcavity consists of two distributed Bragg reflectors (DBRs) made of seven pairs of SiO2/TiO2 layers each (see “Methods”). The bottom mirror was grown on a glass substrate for the purpose of transmission measurements. The superlattice N3 is composed of alternating layers of WS2 ML and insulating spacers of SiO2 with thickness around 4 nm and deposited by E-beam evaporation (see “Methods” for details), as schematically shown in Fig. 1a. Optical microscope images of the sample on the bottom DBR are shown in Supplementary Fig. S1. In the superlattice areas, electronic interaction between each ML is absent when neighboring WS2 MLs are separated by a SiO2 layer, thick enough to avoid tunneling. Thus, the superlattice areas preserve the ML optical and electronic properties, while having a higher oscillator strength30.Fig. 1 Optical characterization of TMDs microcavities.

a Schematic of the TMD based microcavities, consisting of WS2-superlattices sandwiched between two DBRs. b The angle-resolved reflectivity map of the superlattice microcavity, collected at room temperature by using a white light source. The two dashed yellow curves are the UPB and LPB dispersions obtained by using a two-coupled oscillator model. The dashed red line and dashed parabola represent the exciton energy and the dispersion of cavity mode, respectively.

To establish the existence of a strong coupling between cavity photons and material excitons, we performed angle-resolved white light reflectance measurements. The microcavity dispersion shows a clear anti-crossing behavior at the exciton resonances and the formation of two new branches, i.e., the lower and upper polariton branches (LPB and UPB), as shown in Fig. 1b for N3 superlattice. The experimental dispersion was accurately described by a two-coupled oscillator model, with a Rabi splitting of around 66 meV, as indicated by the dashed yellow curves in Fig. 1b. The reflectance maps for both ML and N2 superlattice region, as well as the Rabi splitting energy as a function of number of stacked layers are shown in Supplementary Fig. S2. These values are comparable to previous works and proportional to the square root of the number of stacked layers29,31.

Spin-dependent nonlinearities are critical for the development of polariton physics as well as for the realization of polariton-based spintronic devices. To further explore the potential applications of TMD polaritons, we have studied the polarization-dependent interactions under pulsed resonant pumping. Specifically, linear polarization can be considered as a coherent superposition of two counter-polarized circular components. Consequently, under linearly polarized laser excitation, only half of the excited polaritons have the same spin, while under circularly polarized laser excitation, all polaritons are created with the same spin. Therefore, a stronger interaction is expected for the circular polarization while the interaction is halved for the linear polarization, as it has been observed in GaAs-based polaritons15 and 2D perovskite at room temperature33–36. We measured the sample in a transmission configuration by using a femtosecond pulsed laser (100 fs, 1 kHz) that resonantly excites the ground state of the LPB (see “Methods” for details). This allows us to treat the system as a pure polariton system, avoiding the formation of an incoherent and uncoupled exciton reservoir as well as minimizing thermal heat effects.

The normalized transmission spectra for the superlattice microcavity as a function of polariton density with linearly (red curves) and circularly (blue and purple curves) polarized pump are displayed in Fig. 2a. Under both excitation conditions, we observed a continuous blueshift by increasing the polariton density (see “Methods”). The energy shift of the ground state (△ELP) for linear (red dots) and circular (light blue dots for σ− and dark blue for σ+) polarized pump is reported in Fig. 2b. The dashed red and blue line represent a linear fit of the experimental data for linear and circular polarization, respectively. As can be immediately seen, the energy shift of the ground state for circularly polarized excitation (blue curves) is higher than that of the linear case (red curves) for the superlattice, therefore confirming a clear modulation in response to changes in the pump polarization state. Moreover, we also checked whether the observed polarization-dependent nonlinearity is affected by a rotation in the linear polarization. Specifically, we measured the blueshift of the ground state as a function of the pumping power for the two orthogonal linear components, TE and TM, respectively. As shown in Supplementary Fig. S3, the blueshift is almost identical for the two linear polarizations.Fig. 2 Spin-dependent nonlinearities in TMD microcavities.

a Resonant transmission spectra of the LPB ground state in superlattice microcavity (k = 0) for different excitation powers for linear (red), circular σ+ (blue) and σ− (purple) polarized excitations, respectively. b The blueshift of the ground state as a function of polariton density for linear (red dots) and circular (light blue dots for σ− and dark blue for σ+) polarized excitations for the superlattice microcavity. The dashed red and blue lines are linear fits of the experimental data. The error bars of polariton density in b are explained in the “Methods” section. The error bars on the energy shift data correspond to the 95% confidence interval of a Gaussian fit.

We then analyze the above reported polarization-resolved nonlinear transmission measurements considering the interaction strengths between polaritons with parallel (α1), or antiparallel (α2) spins. The energy shift of the ground state is proportional to α1 and (α1+α2)/2 for circularly and linearly polarized pumps respectively (with the same total intensity). The blueshift ratio measured in different polarizations is then given by (1+α2/α1)/2 and is equal to the ratio between the two slopes (L/C) experimentally observed, leading to a α2/α1 ratio of around 0.14 ±0.01 for superlattices. The measured value for the superlattice is consistent with the typical picture of spin-dependent polariton-polariton interactions, where the repulsive interaction is stronger for polaritons with the same spin (α1>0) while the interaction for polaritons with different spins is much weaker (α2/α1≪1). Moreover, from the above reported data, we also evaluated the exciton-exciton interaction strength for the circular excitation case. The blueshifts of the ground state were well-fitted by a linear function where the slope represents the polariton-polariton interaction constant, according to △Epol=α1npol. By considering the excitonic fraction and the following relation gxx=α1/X4 (see Supplementary Fig. S4), we then retrieved an exciton-exciton interaction constant of 0.129 μeVμm2. Our present results indicate that our superlattice exhibits a significant spin-anisotropy of polariton interactions, which is a necessary foundation to consider polariton-based spintronic applications.

In traditionally studied GaAs-based systems the dominant mechanism of polariton-polariton interactions comes from the exciton exchange interaction, where pairs of excitons exchange electrons or holes. While this preserves the spin of excitons with the same spin, it results in a conversion of bright excitons to dark excitons when excitons of opposite spin interact, which must be suppressed as dark excitons have a different resonance energy37. Consequently, opposite spin excitons tended to rely on a different mechanism of interaction, with scattering through biexcitons understood as a dominant source.

In TMD-based systems, the mechanisms of polariton-polariton interaction are still a topic of debate. While the exchange interaction is still expected, polaron mediated interaction38,39, trion mediated interaction40,41, and phase-space filling are discussed42. Our present study does not aim to separate these mechanisms, but only point out that a strong spin-anisotropic interaction response is possible with TMD-based microcavities and that a spin-switch can be realized based on it (at room temperature). While the different interaction mechanisms are expected to be weaker in multilayered superlattice samples, as the interacting particles are spread across multiple TMD layers, we believe that phase-space filling can remain significant as the Rabi splitting that it reduces is larger in superlattice samples31. We note that phase-space filling should be fully anisotropic, where the filling of an exciton state causes a reduction of the Rabi splitting for the same spin polarization33.

The observation of spin-dependent interactions in superlattice microcavities at room temperature is a crucial advance toward the practical implementation of polariton spintronics. Specifically, by increasing the Rabi splitting of the system, the LPB shifts far in energy from the exciton resonance, therefore limiting the inhomogeneous broadening and keeping a good visibility and suitable excitonic fraction. This is particularly important for the realization of nonlinear polariton devices that usually rely on the resonant excitation of the LPB8. It is noteworthy that an increase of the Rabi splitting can also be achieved by decreasing the volume of the optical mode, i.e., by employing different optical platforms such as Bloch surface waves, bound states in the continuum, or plasmons43–45. On the other hand, as shown in Supplementary Fig. S2, the ML region of our sample, shows a positive detuning (Δ=0–5 meV) and the LPB is quite broad, which made it difficult to establish an initially OFF state. Although an increase in the detuning would mitigate the inhomogeneous broadening of the ML sample, the excitonic fraction, and therefore the nonlinear character, would decrease dramatically, making it challenging to observe a significant blueshift. For these reasons, the heterostructure offers the best compromise, naturally mitigating the inhomogeneous broadening and maintaining suitable excitonic fractions far from the exciton resonance.

We take advantage of the spin-dependent interactions and increased Rabi splitting of superlattice polaritons for the realization of polariton spin switches. With the aim of testing the efficiency of our spin switch device, we employed an experimental configuration closely resembling that reported in GaAs-based microcavities. Nevertheless, we conducted our experiments under pulsed excitation conditions, which could be more relevant for switching applications. The microcavity is pumped with two beams: a strong pump beam focused at a given in-plane momentum with energy slightly blue-detuned with respect to the LPB and a weak control beam always resonant with the LPB at k = 0, which triggers the transmission of the pump. Both beams were obtained from the same laser pulse that was split into two excitation paths and focused on a spot with a diameter of around 5 μm. Two bandpass filters were used to reduce the laser bandwidth while a motorized stage placed on the control excitation line was employed to synchronize in time the two beams on the sample plane (see “Methods” and Supplementary Fig. S5 for further details). The pump laser is circularly polarized and initially out of resonance (OFF state), therefore with almost zero transmitted intensity, while the control beam is either co- or cross-circular polarized with respect to the pump and always resonant with the ground state of the LPB, as schematically shown in Fig. 3a. The polarization state of both beams has been carefully characterized as shown in Supplementary Fig. S6. The detection line is always co-circular polarized with respect to the pump while the control beam transmitted signal is removed from the detection line by using a filter in momentum space. At low control fluences the system is always in the OFF state for both polarizations while, by increasing the control beam fluence, the presence of polariton-polariton interactions blueshifts the LPB, thus gradually bringing the mode into resonance with the pump beam, therefore resulting in the ON state with high transmitted intensity for the co-circular excitation scheme. This behavior is summarized by the red dotted line in Fig. 3b. The yellow/red region in the left/right part of Fig. 3b indicates the OFF/ON state regions. However, as we discussed above, in systems with spin-dependent interactions, polaritons with parallel spin exhibit stronger interactions than those with antiparallel spin. This peculiarity allows the ON/OFF switch of the pump by either changing the control beam fluence or by simply changing the polarization state of the control from co- to cross-circular polarization, as indicated by the blue dotted line in Fig. 3b. We evaluated the ON-OFF ratio by dividing the transmitted pump intensity in the high-control power regime (ON state, red region) by that in the low-control power regime (OFF state, yellow region), obtaining an on-off ratio of ~100. It is important to note that this ratio is influenced by critical experimental parameters, including the linewidth of the pump beam, the quality of the microcavity, and the visibility of the LPB, which varies with the wavevector k. The real space images of the transmitted pump in the ON state region for the four different combinations, i.e., pump σ + /control σ + , pump σ + /control σ-, pump σ-/ control σ + , and pump σ-/ control σ- are shown in Fig. 3c–f, respectively. A significant increase in the population of the transmitted pump is only visible when the control beam is co-circularly polarized with the pump (Fig. 3c, f and Supplementary Fig. S7a) while only a weak residual transmission is visible in the cross-polarization schemes (Figs. 3d, e and S7b). This residual emission may be attributed to local imperfections of the sample. The high nonlinearities of TMD microcavities allow the realization of the spin switch at low pumping fluences. The above reported data has been also obtained with a slightly different injection k for the pump, as shown in Supplementary Fig. S8. The above presented data immediately suggest how the device requires a small footprint, comparable to that used in GaAs-based samples, and low pumping power. Indeed, with a pumping power of the probe exceeding 100 nW, the system reaches the ON state.Fig. 3 Spin switch based on the superlattice microcavity.

a Schematic of the experimental configuration with two beams focused on the same spatial position with two different incident angles and energies. The yellow dashed curve indicates the LPB. b Integrated transmitted real space signal as a function of the control beam power and polarization, σ+ (red dots) andσ− (blue dots), respectively. The threshold for the co-circularly polarized excitation scheme is clearly visible. The yellow and red regions indicate the OFF and ON state region, respectively. The top-left label indicates the polarization of the pump while the detection line is co-circular with the pump polarization. The reported counts are integrated over the whole spot size. c–f Results of the combined effect of co-circularly and cross-circularly polarized pump and control, respectively. When the control polarization is changed to the cross-circular polarization, no enhancement of transmission is observed. The bottom-right label indicates the polarization of the detection line. Scale bar is 1 μm. g Integrated transmitted intensity of the pump beam as a function of the ellipticity of the control beam. The point of maximum/minimum transmission indicates the co/cross-circular polarization state. The top-right label indicates the polarization of the pump while the detection line is co-circular with the pump polarization. h Transient transmission signal versus the time delay between the two beams.

We investigated the sensitivity of the spin switch to the polarization of the control beam and time delay between the two beams with a fixed pump fluence while keeping the control beam fluence in the ON region depicted in Fig. 3b. The transmitted pump emission integrated in real space is plotted as a function of the ellipticity of the control, which is defined by the parameter θ, where θ=0(π/2) corresponds to σ+(σ−) circular polarization, as shown in Fig. 3g. When the two beams are co-circularly polarized (θ=0), a strong signal is observed, indicating that the weak control is sufficient to switch the system from OFF to ON regime. By further increasing θ, the transmitted pump gradually decreases and reaches its minimum at 0.5 rad where the control beam is cross-circularly polarized with respect to the pump. A detailed analysis of the detected polarization state is presented in Supplementary Fig. S9. Here, the transmitted pump intensity for a co-σ+ excitation scheme is plotted as a function of the ellipticity of the detection line, thus confirming that the σ- state is almost empty. In addition, we also monitored the temporal response of the spin switch by scanning the time delay between the two beams kept in the co-σ+ excitation scheme, as shown in Fig. 3h. The transmitted pump reaches its maximum and completely vanishes within 1 ps (at 0.5 ps the transmitted pump signal is decreased by 80%), thus confirming the ultrafast nature of the spin switch. It is important to note that the intrinsic short lifetime of our sample allows an ultrafast switching time that is two orders of magnitude faster than GaAs-based microcavities11. Although some proof-of-concept polariton transistors and switches have been proposed by using perovskites and organics12,13, none of these designs exploited the polarization degree of freedom. This additional degree of freedom, along with the high stability of TMDs, will certainly push them as a leading material platform for room-temperature polaritonic applications, surpassing other active materials that often exhibit significant instability under optical pumping conditions.

An alternative scenario to the circular excitation schemes discussed is for the pump beam to be linearly polarized (superposition of a clockwise and anticlockwise circular polarization), while the control beam is either σ+ or σ-, as schematically shown in Fig. 4a and b, respectively. Even in this case, at high control fluences the system is switched from the OFF to ON state. However, when the system reaches the ON state, a rotation of the control beam polarization state from clockwise to anticlockwise leads to a similar rotation of the transmitted pump polarization. To underline this effect, we detected the real space images of the transmitted pump kept in the linear polarization (vertical) by changing either the polarization state of the control or of the detection line, which are presented in Fig. 4c–f for control σ + /det σ + , control σ + /det σ-, control σ-/det σ+ and control σ-/det σ-, respectively. We performed theoretical simulations (Fig. 4g–j) by solving the spin-dependent Gross-Pitaevskii equations for polaritons, which quantitatively reproduce these experimental results under the same pulsed conditions. Our results clearly demonstrate the feasibility of polariton-based spintronic devices working at room temperature and are a fundamental step for future applications. These results not only mark a significant advancement in the field but also pave the way for the development of polaritonic devices embedding TMDs that harness spin-dependent interaction-driven phenomena.Fig. 4 Spin switch in different polarizations.

a, b Schematic of the experimental configuration of the spin switch with a linearly polarized pump and circularly polarized control. The transmitted pump signal (with linear polarization being a superposition of σ− and σ+) is always co-polarized with the control beam. c–f Real-space experimental images of the transmitted pump signal. In the presence of a circularly polarized control beam either σ+ (c and d) or σ− (e and f), the emission of the whole spot become co-polarized with the control beam. The bottom-right label indicates the polarization of the detection line. g–j Results from the numerical solution of the Gross- Pitaevskii equations, corresponding to the conditions of c–f respectively.

Discussion

A roadmap on excitonic computing was published recently46, explicitly identifying spintronic logic with excitons and exciton-polaritons as one of the most promising routes. We refer to this work for a survey of the existing state-of-the-art and note that it also identifies the need for cryogenic cooling in existing proof-of-concept experiments7 as the critical issue in exciton spintronics. This is precisely what we address in the present work. It is also important to note that recent reports have shown the potential of valley-polaritons for the implementation of polarization-sensitive devices22–24. Although a strong polarization response of the photoluminescence emission has been achieved, these works do not exploit the nonlinear regime, a fundamental requirement for the realization of polariton devices. Our work shows the practical development of devices that exploit polariton nonlinearities, a feature that underscores the unique appeal of polaritons. However, the application of TMD polaritons is currently constrained by the methodology employed, with the largest limitation being the restricted in-plane propagation observed in state-of-the-art TMD microcavities. Indeed, any logic architecture needs to cascade, allowing for the sequential processing of signals from one stage to the next. This fundamental requirement cannot be satisfied without a suitable in-plane propagation, i.e. long polariton lifetime, and represents a concrete limitation of our samples. Additionally, the constrained spatial extension of the ML flakes, resulting from mechanical exfoliation, poses another significant challenge. In this respect, chemical vapor deposition (CVD) offers a scalable fabrication approach for depositing 2D materials over wafer-scale areas30. This advancement could significantly broaden the possibilities for designing TMD polaritons, offering new pathways to overcome the current limitations.

In this work, we present an investigation of polarization-dependent interactions between polaritons in WS2-superlattice microcavities embedding up to three separated layers. Our results reveal that polaritons formed in multilayer structures exhibit extremely robust spin-dependent interactions at room temperature, a fundamental property to realize polariton-based spin devices. The presence of spin-dependent interactions has been exploited to demonstrate a proof-of-concept polariton spin switch where a weak control beam can effectively switch ON and actively modify the polarization of a pump beam. Unlike other previously proposed proof-of-concept devices operating at room temperature, our approach represents the first demonstration exploiting spin-dependent polariton nonlinearities for the realization of polariton devices effectively harnessing the spin degree of freedom of polaritons. It is important to note that the intrinsic short lifetime of the sample allows an ultrafast switching time that is two orders of magnitude faster than GaAs-based microcavities. Our results pave the way for novel applications for polariton-based spintronic devices. Further developments in this direction could also be achieved by using van der Waals heterostructure and wafer-scale growth methods to develop new on-chip integrated electronic devices.

Methods

Superlattice and microcavity fabrication

The WS2 superlattice microcavity is composed of a λExc/2 microcavity that is sandwiched between two DBR mirrors. The bottom DBR is composed of seven alternating pairs of titanium dioxide (TiO2) and silicon dioxide (SiO2) deposited by using electron-beam evaporation (Cello, Ohmiker-50B) onto a glass substrate. A transparent substrate is required to perform transmission measurements. Subsequently, the first λExc/4-SiO2 spacer was deposited onto the bottom DBR mirror using the same fabrication technique. A mixed-dimensional superlattice consisting of alternating layers of ML WS2 and SiO2 insulating layers with a thickness of ~4 nm was then assembled. The same electron-beam evaporator was used to deposit the thin SiO2 layer. All the WS2 MLs were mechanically exfoliated from the same bulk crystal, procured from HQ-Graphene with 2H phase. For details, the mechanically exfoliated ML was prepared using the scotch tape method and transferred onto a bottom DBR via a dry transfer technique with the aid of a polydimethylsiloxane (PDMS, PF-X4, Gel-Pack) stamp. Following this, a thin layer of SiO2 was deposited via an E-beam evaporator. The second ML was then exfoliated onto a PDMS stamp. Using a home-built transfer stage, a top WS2/PDMS film was meticulously aligned with the first ML and subsequently placed onto the surface. Upon removal of the PDMS layer, the second ML was aligned with the first ML and positioned on top of the SiO2 layer. The superlattice (N3) structure was fabricated by repeating this process three times. A layer of poly (methyl methacrylate, PMMA) was spin-coated onto the bottom DBR containing the superlattice structure. Finally, as for the top mirror, a small piece of 6.5-pair DBR was transferred onto the PMMA layer, precisely positioned over the superlattice structure, using the same PDMS micro-transfer process. Further information regarding this transfer technique can be found in ref. 47.

Polariton nonlinearities and optical switching measurements

Two microscope objectives with large numerical apertures of 100× (NA = 0.9) and 50× (NA = 0.45), were used in reflection and transmission configuration, respectively. The dispersion presented in Fig. 1 was obtained by collecting white light in the reflection configuration. The polariton dispersion was resolved by imaging the back aperture of the microscopy objective (Fourier plane) onto the camera (see Supplementary Fig. S5). Specifically, the signal from the microcavity was collected through the narrow slit of a Horiba iHR550 spectrometer equipped with three gratings (150 lines/mm, 600 lines/mm, 1800 lines/mm) and coupled to a 2D charge-coupled device array (Horiba, Symphony II), to achieve optimal spatial and spectral resolution.

To investigate spin-dependent interactions, a pulsed laser was employed in our experimental setup. The excitation pulsed source was derived from a Ti: sapphire laser that was equipped with an ultrafast amplifier (Spectra-Physics) and a computer-controlled Optical Parametric Amplifier for wavelength modulation. The repetition rate was 1 KHz, and the pulse width was ~100 fs. The output beam was split into two paths, where one beam was directed through a mechanical delay line (Newport, M-ILS 150CC DC Servo Linear Stage) and served as the control beam to introduce the blueshift of the polariton mode. This beam excited the microcavity at k = 0 and was spectrally filtered using a band-pass filter. The precise energetic position of the filter was adjusted by controlling the tilt angle.

Referring to data concerning the spin-dependent interactions, we calculate the incident pulsed energy Etot = P/R, where P denotes the excitation power and R = 1 kHz is the repetition rate of our pulsed laser. The excitation spot area was S ~ 20 µm2, with a diameter of ~5 µm. As displayed in Supplementary Fig. S5, the quarter-waveplate, half-waveplate, and linear polarized components were utilized to modify the polarization of the control beam. To estimate the polariton density, a conservative approximation was made, considering the total injected energy was completely absorbed. The energy absorbed by the LPB at k = 0 can be evaluated by calculating the integral of the absorption peak under the injected laser profile. The injected photon energy was estimated as follows:1 EINJ=Etot∫Aabs,LP∫Alaser,LP

Herein, the ∫Aabs,LP represents the integral of the dip of the ground state, while ∫Alaser,LP denotes the integral of the laser profile. Specifically, for the dip of the dispersion, we have estimated the injected photon number, which can be related to the polariton density through the photonic fraction (C) of the ground state, according to the following relation: nph=npolC. By making the assumptions of laser injection, the polariton density can be estimated as npol=nphC=EinjCAbeamELP, where ELP is the energy of the ground state, and Abeam is the beam area, respectively. For a superlattice, we have calculated the polariton density per layer, thus dividing the overall polariton density by the total number of layers.

During the optical spin switch measurements, a second beam was employed, which was slightly off-resonance and served as the pump beam. A notch filter was utilized to spectrally filter this beam. The experimental results depicted in Fig. 3 were obtained by utilizing a circularly polarized pump and a circularly polarized control beam, whereas those presented in Fig. 4 were acquired by employing linearly polarized pump and circularly polarized control beams. The pump and control beams were focused on the microcavity with a beam size of ~5 µm.

Theoretical calculations

With coherent pump injected, the dynamics of exciton-polaritons can be expressed by the Gross-Pitaevskii equation as:2 iℏ∂Ψ±∂t=E^LP−iγ+α1Ψ±2+α2Ψ∓2Ψ±+F±+P±

Here Ψ± is the wavefunction of polaritons in two-dimensional real space for which the ± symbols correspond to two circular polarizations. E^LP is the operator for the kinetic energy of polaritons in the lower polariton band, which can be written as:3 E^LPk=Eck+Exk−Eck−Exk2+Ω22

Ec,Ex ﻿are the energies for cavity photon and exciton modes, respectively, which are described by parabolic dispersion with effective masses mc,mx. Ω is the Rabi splitting, which equals to 66 meV in the simulation (the same as the experiment). γ is the polariton decay rate. α1,α2 are the strengths of polariton-polariton interactions between parallel and antiparallel spins. In our theoretical model, we consider two coherent laser pulses corresponding to pump, F±, and control, P±:4 F±=F0,±eikpxe−iωte−x−x02dx2e−y−y02dy2e−t−t02dt2

5 P±=P0,±e−iωote−x−x02dx2e−y−y02dy2e−t−t02dt2

F0,±,P0,± are the amplitudes of the coherent laser pulses; kp is the pump in-plane wavevector; ω is the angular frequency of the pulse, x0,y0 is the center position of the pulses; dx,dy are the widths of the pulses and dt is the pulse duration. For parameters in the simulation, we take mc=2×10−5me (me is the free electron mass), γ=0.6meV, α1=0.1ueVum2,α2=−0.1α1 which are taken in agreement with experimental values. For the pulse’s parameters, we take directly from our experimental values.

Supplementary information

Supplementary Information

Peer Review File

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-024-51612-2.

Acknowledgements

J.Z., A.F., K.D., and T.C.H.L. gratefully acknowledge the support from the Singapore Ministry of Education via the AcRF Tier 2 project (MOE-T2EP50121-0020) and AcRF Tier 3 project (MOE2018- T3-1-002). Q.X. gratefully acknowledges strong funding support from the National Key Research and Development Program of China (Grant No. 2022YFA1204700), National Natural Science Foundation of China (No. 122507101126 and 12020101003), and support from the State Key Laboratory of Low-Dimensional Quantum Physics of Tsinghua University and the Tsinghua University Initiative Scientific Research Program. D.S. and A.F. gratefully acknowledge “Quantum Optical Networks based on Exciton-polaritons” (Q-ONE, N. 101115575, HORIZON-EIC-2022-PATHFINDER CHALLENGES EU project), “National Quantum Science and Technology Institute” (NQSTI, N. PE0000023, PNRR MUR project), “Integrated Infrastructure Initiative in Photonic and Quantum Sciences” (I-PHOQS, N. IR0000016, PNRR MUR project), Neuromorphic Polariton Accelerator” (PolArt, N. 101130304, Horizon-EIC-2023-Pathfinder Open EU project). J.Z. gratefully acknowledges the Presidential Postdoctoral Fellowship support from the Nanyang Technological University.

Author contributions

A.F., J.Z., and T.L. conceived the ideas and designed the experiments. J.Z. prepared superlattice microcavity samples. J.Z. and A.F. carried out the optical spectroscopy measurements and analyzed data. R.B. and K.D. performed theoretical calculations. J.Z., A.F., R.S., D.S. Q.X., and T.L. wrote the manuscript with contributions from all authors. All authors discussed the results and commented on the manuscript.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.

Data availability

All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Additional data related to this paper may be requested from the authors.

Competing interests

The authors declare no competing interests.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

These authors contributed equally: Jiaxin Zhao, Antonio Fieramosca, Ruiqi Bao.
==== Refs
References

1. Žutić I Fabian J Das Sarma S Spintronics: fundamentals and applications Rev. Mod. Phys. 2004 76 323 410 10.1103/RevModPhys.76.323
Žutić, I., Fabian, J. & Das Sarma, S. Spintronics: fundamentals and applications. Rev. Mod. Phys. 76, 323–410 (2004).10.1103/RevModPhys.76.323
2. Soumyanarayanan A Reyren N Fert A Panagopoulos C Emergent phenomena induced by spin–orbit coupling at surfaces and interfaces Nature 2016 539 509 517 10.1038/nature19820 27882972
Soumyanarayanan, A., Reyren, N., Fert, A. & Panagopoulos, C. Emergent phenomena induced by spin–orbit coupling at surfaces and interfaces. Nature 539, 509–517 (2016).27882972 10.1038/nature19820
3. O’Brien JL Furusawa A Vučković J Photonic quantum technologies Nat. Photonics 2009 3 687 695 10.1038/nphoton.2009.229
O’Brien, J. L., Furusawa, A. & Vučković, J. Photonic quantum technologies. Nat. Photonics 3, 687–695 (2009).10.1038/nphoton.2009.229
4. Wang J Sciarrino F Laing A Thompson MG Integrated photonic quantum technologies Nat. Photonics 2020 14 273 284 10.1038/s41566-019-0532-1
Wang, J., Sciarrino, F., Laing, A. & Thompson, M. G. Integrated photonic quantum technologies. Nat. Photonics 14, 273–284 (2020).10.1038/s41566-019-0532-1
5. Nishikawa Y Tackeuchi A Yamaguchi M Muto S Wada O Ultrafast all-optical spin polarization switch using quantum-well etalon IEEE J. Sel. Top. Quantum Electron. 1996 2 661 667 10.1109/2944.571766
Nishikawa, Y., Tackeuchi, A., Yamaguchi, M., Muto, S. & Wada, O. Ultrafast all-optical spin polarization switch using quantum-well etalon. IEEE J. Sel. Top. Quantum Electron. 2, 661–667 (1996).10.1109/2944.571766
6. Pearton SJ Advances in wide bandgap materials for semiconductor spintronics Mater. Sci. Eng.: R: Rep. 2003 40 137 168 10.1016/S0927-796X(02)00136-5
Pearton, S. J. et al. Advances in wide bandgap materials for semiconductor spintronics. Mater. Sci. Eng.: R: Rep. 40, 137–168 (2003).10.1016/S0927-796X(02)00136-5
7. Amo A Exciton–polariton spin switches Nat. Photonics 2010 4 361 10.1038/nphoton.2010.79
Amo, A. et al. Exciton–polariton spin switches. Nat. Photonics 4, 361 (2010).10.1038/nphoton.2010.79
8. Sanvitto D Kéna-Cohen S The road towards polaritonic devices Nat. Mater. 2016 15 1061 1073 10.1038/nmat4668 27429208
Sanvitto, D. & Kéna-Cohen, S. The road towards polaritonic devices. Nat. Mater. 15, 1061–1073 (2016).27429208 10.1038/nmat4668
9. Amo A Superfluidity of polaritons in semiconductor microcavities Nat. Phys. 2009 5 805 810 10.1038/nphys1364
Amo, A. et al. Superfluidity of polaritons in semiconductor microcavities. Nat. Phys. 5, 805–810 (2009).10.1038/nphys1364
10. Deng H Haug H Yamamoto Y Exciton-polariton bose-einstein condensation Rev. Mod. Phys. 2010 82 1489 10.1103/RevModPhys.82.1489
Deng, H., Haug, H. & Yamamoto, Y. Exciton-polariton bose-einstein condensation. Rev. Mod. Phys. 82, 1489 (2010).10.1103/RevModPhys.82.1489
11. Adrados C Motion of Spin Polariton Bullets in Semiconductor Microcavities Phys. Rev. Lett. 2011 107 146402 10.1103/PhysRevLett.107.146402 22107218
Adrados, C. et al. Motion of Spin Polariton Bullets in Semiconductor Microcavities. Phys. Rev. Lett. 107, 146402 (2011).22107218 10.1103/PhysRevLett.107.146402
12. Zasedatelev AV A room-temperature organic polariton transistor Nat. Photonics 2019 13 378 383 10.1038/s41566-019-0392-8
Zasedatelev, A. V. et al. A room-temperature organic polariton transistor. Nat. Photonics 13, 378–383 (2019).10.1038/s41566-019-0392-8
13. Feng J All-optical switching based on interacting exciton polaritons in self-assembled perovskite microwires Sci. Adv. 2021 7 eabj6627 10.1126/sciadv.abj6627 34757800
Feng, J. et al. All-optical switching based on interacting exciton polaritons in self-assembled perovskite microwires. Sci. Adv. 7, eabj6627 (2021).34757800 10.1126/sciadv.abj6627
14. Chen F Optically controlled femtosecond polariton switch at room temperature Phys. Rev. Lett. 2022 129 057402 10.1103/PhysRevLett.129.057402 35960578
Chen, F. et al. Optically controlled femtosecond polariton switch at room temperature. Phys. Rev. Lett. 129, 057402 (2022).35960578 10.1103/PhysRevLett.129.057402
15. Vladimirova M Polariton-polariton interaction constants in microcavities Phys. Rev. B 2010 82 075301 10.1103/PhysRevB.82.075301
Vladimirova, M. et al. Polariton-polariton interaction constants in microcavities. Phys. Rev. B 82, 075301 (2010).10.1103/PhysRevB.82.075301
16. Kammann E Nonlinear optical spin hall effect and long-range spin transport in polariton lasers Phys. Rev. Lett. 2012 109 036404 10.1103/PhysRevLett.109.036404 22861878
Kammann, E. et al. Nonlinear optical spin hall effect and long-range spin transport in polariton lasers. Phys. Rev. Lett. 109, 036404 (2012).22861878 10.1103/PhysRevLett.109.036404
17. Dreismann A A sub-femtojoule electrical spin-switch based on optically trapped polariton condensates Nat. Mater. 2016 15 1074 1078 10.1038/nmat4722 27500807
Dreismann, A. et al. A sub-femtojoule electrical spin-switch based on optically trapped polariton condensates. Nat. Mater. 15, 1074–1078 (2016).27500807 10.1038/nmat4722
18. Paraïso TK Wouters M Léger Y Morier-Genoud F Deveaud-Plédran B Multistability of a coherent spin ensemble in a semiconductor microcavity Nat. Mater. 2010 9 655 60 10.1038/nmat2787 20601942
Paraïso, T. K., Wouters, M., Léger, Y., Morier-Genoud, F. & Deveaud-Plédran, B. Multistability of a coherent spin ensemble in a semiconductor microcavity. Nat. Mater. 9, 655–60 (2010).20601942 10.1038/nmat2787
19. Ghosh S Microcavity exciton polaritons at room temperature Photonics Insights 2022 1 R04 10.3788/PI.2022.R04
Ghosh, S. et al. Microcavity exciton polaritons at room temperature. Photonics Insights 1, R04 (2022).10.3788/PI.2022.R04
20. Wang QH Kalantar Zadeh K Kis A Coleman JN Strano MS Electronics and optoelectronics of two-dimensional transition metal dichalcogenides Nat. Nanotechnol. 2012 7 699 10.1038/nnano.2012.193 23132225
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, 699 (2012).23132225 10.1038/nnano.2012.193
21. Zeng H Dai J Yao W Xiao D Cui X Valley polarization in MoS2 monolayers by optical pumping Nat. Nanotechnol. 2012 7 490 493 10.1038/nnano.2012.95 22706701
Zeng, H., Dai, J., Yao, W., Xiao, D. & Cui, X. Valley polarization in MoS2 monolayers by optical pumping. Nat. Nanotechnol. 7, 490–493 (2012).22706701 10.1038/nnano.2012.95
22. Sun Z Optical control of room-temperature valley polaritons Nat. Photonics 2017 11 491 10.1038/nphoton.2017.121
Sun, Z. et al. Optical control of room-temperature valley polaritons. Nat. Photonics 11, 491 (2017).10.1038/nphoton.2017.121
23. Dufferwiel S Valley-addressable polaritons in atomically thin semiconductors Nat. Photonics 2017 11 497 10.1038/nphoton.2017.125
Dufferwiel, S. et al. Valley-addressable polaritons in atomically thin semiconductors. Nat. Photonics 11, 497 (2017).10.1038/nphoton.2017.125
24. Chen Y-J Cain JD Stanev TK Dravid VP Stern NP Valley-polarized exciton–polaritons in a monolayer semiconductor Nat. Photonics 2017 11 431 10.1038/nphoton.2017.86
Chen, Y.-J., Cain, J. D., Stanev, T. K., Dravid, V. P. & Stern, N. P. Valley-polarized exciton–polaritons in a monolayer semiconductor. Nat. Photonics 11, 431 (2017).10.1038/nphoton.2017.86
25. Lundt N Optical valley Hall effect for highly valley-coherent exciton-polaritons in an atomically thin semiconductor Nat. Nanotechnol. 2019 14 770 775 10.1038/s41565-019-0492-0 31332345
Lundt, N. et al. Optical valley Hall effect for highly valley-coherent exciton-polaritons in an atomically thin semiconductor. Nat. Nanotechnol. 14, 770–775 (2019).31332345 10.1038/s41565-019-0492-0
26. LaMountain T Valley-selective optical Stark effect of exciton-polaritons in a monolayer semiconductor Nat. Commun. 2021 12 4530 10.1038/s41467-021-24764-8 34312389
LaMountain, T. et al. Valley-selective optical Stark effect of exciton-polaritons in a monolayer semiconductor. Nat. Commun. 12, 4530 (2021).34312389 10.1038/s41467-021-24764-8
27. Zhao J Nonlinear polariton parametric emission in an atomically thin semiconductor based microcavity Nat. Nanotechnol. 2022 17 396 402 10.1038/s41565-022-01073-9 35288672
Zhao, J. et al. Nonlinear polariton parametric emission in an atomically thin semiconductor based microcavity. Nat. Nanotechnol. 17, 396–402 (2022).35288672 10.1038/s41565-022-01073-9
28. Dufferwiel S Valley coherent exciton-polaritons in a monolayer semiconductor Nat. Commun. 2018 9 4797 10.1038/s41467-018-07249-z 30442886
Dufferwiel, S. et al. Valley coherent exciton-polaritons in a monolayer semiconductor. Nat. Commun. 9, 4797 (2018).30442886 10.1038/s41467-018-07249-z
29. Dufferwiel S Exciton–polaritons in van der Waals heterostructures embedded in tunable microcavities Nat. Commun. 2015 6 8579 10.1038/ncomms9579 26446783
Dufferwiel, S. et al. Exciton–polaritons in van der Waals heterostructures embedded in tunable microcavities. Nat. Commun. 6, 8579 (2015).26446783 10.1038/ncomms9579
30. Kumar P Light–matter coupling in large-area van der Waals superlattices Nat. Nanotechnol. 2022 17 182 189 10.1038/s41565-021-01023-x 34857931
Kumar, P. et al. Light–matter coupling in large-area van der Waals superlattices. Nat. Nanotechnol. 17, 182–189 (2022).34857931 10.1038/s41565-021-01023-x
31. Zhao J Exciton polariton interactions in Van der Waals superlattices at room temperature Nat. Commun. 2023 14 1512 10.1038/s41467-023-36912-3 36932078
Zhao, J. et al. Exciton polariton interactions in Van der Waals superlattices at room temperature. Nat. Commun. 14, 1512 (2023).36932078 10.1038/s41467-023-36912-3
32. Hu Z Energy transfer driven brightening of MoS2 by ultrafast polariton relaxation in microcavity MoS2/hBN/WS2 heterostructures Nat. Commun. 2024 15 1747 10.1038/s41467-024-45554-y 38409100
Hu, Z. et al. Energy transfer driven brightening of MoS2 by ultrafast polariton relaxation in microcavity MoS2/hBN/WS2 heterostructures. Nat. Commun. 15, 1747 (2024).38409100 10.1038/s41467-024-45554-y
33. Ballarini, D. et al. Transition from the strong-to the weak-coupling regime in semiconductor microcavities: polarization dependence. Applied physics letters 90, 201905 (2007).
34. Fieramosca A Two-dimensional hybrid perovskites sustaining strong polariton interactions at room temperature Sci. Adv. 2019 5 eaav9967 10.1126/sciadv.aav9967 31172027
Fieramosca, A. et al. Two-dimensional hybrid perovskites sustaining strong polariton interactions at room temperature. Sci. Adv. 5, eaav9967 (2019).31172027 10.1126/sciadv.aav9967
35. Liu TY Dynamics of spin‐dependent polariton–polariton interactions in two‐dimensional layered halide organic perovskite microcavities Laser Photonics Rev. 2022 16 2200176 10.1002/lpor.202200176
Liu, T. Y. et al. Dynamics of spin‐dependent polariton–polariton interactions in two‐dimensional layered halide organic perovskite microcavities. Laser Photonics Rev. 16, 2200176 (2022).10.1002/lpor.202200176
36. Fieramosca A Origin of exciton–polariton interactions and decoupled dark states dynamics in 2D hybrid perovskite quantum wells Nano Lett. 2024 24 8240 8247 10.1021/acs.nanolett.4c00418 38925628
Fieramosca, A. et al. Origin of exciton–polariton interactions and decoupled dark states dynamics in 2D hybrid perovskite quantum wells. Nano Lett. 24, 8240–8247 (2024).38925628 10.1021/acs.nanolett.4c00418
37. Shelykh IA Kavokin AV Rubo YG Liew T Malpuech G Polariton polarization-sensitive phenomena in planar semiconductor microcavities Semiconductor Sci. Technol. 2009 25 013001 10.1088/0268-1242/25/1/013001
Shelykh, I. A., Kavokin, A. V., Rubo, Y. G., Liew, T. & Malpuech, G. Polariton polarization-sensitive phenomena in planar semiconductor microcavities. Semiconductor Sci. Technol. 25, 013001 (2009).10.1088/0268-1242/25/1/013001
38. Sun K Shen K Gelin MF Zhao Y Exciton dynamics and time-resolved fluorescence in nanocavity-integrated monolayers of transition-metal dichalcogenides J. Phys. Chem. Lett. 2022 14 221 229 10.1021/acs.jpclett.2c03511 36583951
Sun, K., Shen, K., Gelin, M. F. & Zhao, Y. Exciton dynamics and time-resolved fluorescence in nanocavity-integrated monolayers of transition-metal dichalcogenides. J. Phys. Chem. Lett. 14, 221–229 (2022).36583951 10.1021/acs.jpclett.2c03511
39. Choo, K., Bleu, O., Levinsen, J. & Parish, M. M. Polaronic polariton quasiparticles in a dark excitonic medium. Phys. Rev. B 109, 195432 (2024).
40. Emmanuele R Highly nonlinear trion-polaritons in a monolayer semiconductor Nat. Commun. 2020 11 3589 10.1038/s41467-020-17340-z 32680995
Emmanuele, R. et al. Highly nonlinear trion-polaritons in a monolayer semiconductor. Nat. Commun. 11, 3589 (2020).32680995 10.1038/s41467-020-17340-z
41. Song, K. W., Chiavazzo, S., Shelykh, I. A. & Kyriienko, O. Theory for coulomb scattering of trions in 2D materials. arXiv preprint arXiv:2207.02660 (2022).
42. Stepanov P Exciton-exciton interaction beyond the hydrogenic picture in a MoSe2 monolayer in the strong light-matter coupling regime Phys. Rev. Lett. 2021 126 167401 10.1103/PhysRevLett.126.167401 33961461
Stepanov, P. et al. Exciton-exciton interaction beyond the hydrogenic picture in a MoSe2 monolayer in the strong light-matter coupling regime. Phys. Rev. Lett. 126, 167401 (2021).33961461 10.1103/PhysRevLett.126.167401
43. Barachati F Interacting polariton fluids in a monolayer of tungsten disulfide Nat. Nanotechnol. 2018 13 906 10.1038/s41565-018-0219-7 30082925
Barachati, F. et al. Interacting polariton fluids in a monolayer of tungsten disulfide. Nat. Nanotechnol. 13, 906 (2018).30082925 10.1038/s41565-018-0219-7
44. Liu B Long-range propagation of exciton-polaritons in large-area 2D semiconductor monolayers ACS Nano 2023 17 14442 14448 10.1021/acsnano.3c03485 37489978
Liu, B. et al. Long-range propagation of exciton-polaritons in large-area 2D semiconductor monolayers. ACS Nano 17, 14442–14448 (2023).37489978 10.1021/acsnano.3c03485
45. Maggiolini E Strongly enhanced light–matter coupling of monolayer WS2 from a bound state in the continuum Nat. Mater. 2023 22 964 969 10.1038/s41563-023-01562-9 37217703
Maggiolini, E. et al. Strongly enhanced light–matter coupling of monolayer WS2 from a bound state in the continuum. Nat. Mater. 22, 964–969 (2023).37217703 10.1038/s41563-023-01562-9
46. Hudson, R. J. et al. A framework for multiexcitonic logic. Nat. Rev. Chem. 2 , 136–151 (2024).
47. Rupprecht C Micro-mechanical assembly and characterization of high-quality Fabry–Pérot microcavities for the integration of two-dimensional materials Appl. Phys. Lett. 2021 118 103103 10.1063/5.0034851
Rupprecht, C. et al. Micro-mechanical assembly and characterization of high-quality Fabry–Pérot microcavities for the integration of two-dimensional materials. Appl. Phys. Lett. 118, 103103 (2021).10.1063/5.0034851
