Plasmonic nanostructures offer a way to tailor how atomically thin semiconductors interact with light. For engineers and photonics developers, the appeal of transition-metal dichalcogenide (TMD) monolayers lies in their visible-to-near-infrared optical response, strong excitons and compatibility with nanoscale structures. Coupling these materials to metallic antennas, resonators and waveguides can increase light emission, influence its polarization and provide routes for on-chip optical signals.
Why TMD monolayers are useful optical materials
TMD monolayers consist of a transition-metal layer, such as molybdenum or tungsten, between two chalcogen layers, such as sulfur or selenium. Reducing the material to a single layer changes its electronic band structure; in some TMDs, thinning from a bilayer to a monolayer changes the bandgap from indirect to direct, strengthening absorption and emission.
When an electron and a hole bind together, they form an exciton. TMD monolayers have small exciton radii and relatively large binding energies and oscillator strengths. The resulting strong interaction with light is valuable for optoelectronics, but it does not by itself guarantee efficient useful output. Nonradiative losses can limit quantum yield, and emitted photons must still be collected and directed to a desired destination.
Two types of excitons are especially relevant. Mobile excitons extend through the monolayer and contribute to its characteristic optical response. Defect-bound excitons are localized and can produce narrow emission lines with single-photon characteristics, particularly at low temperatures. Local strain, including strain created when a monolayer conforms to a patterned surface, can help define where localized emission occurs.
How plasmonic nanostructures change emission
A metallic nanostructure can concentrate electromagnetic fields into a volume smaller than the wavelength of light. These plasmonic modes arise from collective electron oscillations. Localized surface plasmon resonances in nanoparticles can act as optical antennas, while surface plasmon polaritons can propagate along a metal–dielectric interface.
The structure around an emitter affects the optical modes available to it, and therefore can modify its radiative decay rate. This environment-dependent change is often described using the Purcell factor. Plasmonic designs combine small optical mode volumes with relatively broad resonances. Their broadband response can make spectral alignment less restrictive than in a high-quality-factor cavity, while near-field enhancement can strengthen interactions with nearby emitters.
Proximity is also a design constraint. Plasmonic fields decay rapidly away from the metal, so a TMD layer must be positioned close to the structure and its exciton dipoles must align appropriately with the field. At very short distances, energy can instead be lost nonradiatively to the metal. Consequently, a faster measured total decay rate does not necessarily mean a larger radiative output: radiative and nonradiative contributions need to be distinguished.
Enhancing TMD emission with plasmonic nanostructures
One implementation places metallic antennas directly on a TMD monolayer. Arrays of silver nanodisks and bowtie antennas coupled to MoS2 have shown increased photoluminescence. The observed enhancement can reflect several effects, including a stronger excitation field, a change in emission rate and improved collection or redirection of emitted light. Antenna geometry and resonance wavelength influence the result.
Self-aligned structures provide another approach. In a reported slot-antenna design, a narrow WSe2 strip was aligned with a silver cavity. The resulting photoluminescence enhancement combined increased excitation, improved collection and a higher radiative decay rate. This illustrates why brightness alone is not a complete measure of performance: the contribution of each mechanism matters when assessing a device.
Patterned substrates can also shape a monolayer and localize excitons through strain. WSe2 transferred over nanopillars or other patterned structures has produced localized emission centers, including emitters with single-photon characteristics. Hybrid metallic and dielectric structures have been used to modify emitter lifetimes. In a particularly confined nanocube-gap arrangement, measuring the same emitter before and after adding a mirror helped separate the change in lifetime from changes in quantum efficiency.
Valley polarization and control of emission
TMD monolayers have two inequivalent energy valleys, commonly labelled K and K′. Their contrasting optical selection rules allow circularly polarized light to address one valley preferentially. This valley degree of freedom is a potential information variable, but exciton valley polarization can be short-lived because of interactions that rapidly mix or depolarize exciton states.
Plasmonic structures can shape the local optical field and its polarization. In principle, metasurfaces can create a field that favors one valley, while chiral plasmonic responses can help enhance or direct valley-selective emission. These approaches aim not only to generate polarized light, but also to control where different valley-associated signals travel. Practical evaluation must account for the distinction between valley polarization in free carriers and polarization associated with optical excitations such as excitons and trions.
From emission enhancement to on-chip routing
An antenna that improves emission into free space does not automatically provide an integrated optical connection. For on-chip photonics, emitted energy must couple to a guided mode and propagate to another location. Experiments with WSe2 on silver nanowires have shown that defect-bound excitons can couple to propagating surface plasmons, with emission observed at the wire ends. Coupling depends on how the exciton dipole aligns with the plasmonic field.
Metal–insulator–metal slot waveguides offer a different field geometry. Their guided field can be concentrated in the gap, with an in-plane component that is better aligned with the dipole of a localized exciton. This illustrates a general integration principle: the field orientation and emitter position are as important as field strength when designing efficient emitter-to-waveguide coupling.
Practical implications for photonics design
For system designers, the results point to several considerations:
- Separate brightness mechanisms. Photoluminescence can rise because of excitation enhancement, radiative-rate modification or better collection. These contributions should not be treated as interchangeable.
- Manage metal losses. Close placement improves near-field coupling but can introduce additional nonradiative decay.
- Match dipoles to fields. Antenna and waveguide geometry should be selected with the emitter’s orientation and location in mind.
- Design for the intended output. Free-space collection, valley-selective emission and guided on-chip propagation require different optical configurations.
- Measure relevant quantities. Total lifetime measurements alone do not establish radiative enhancement; quantum efficiency and collection effects also matter.
Plasmonic nanostructures make atomically thin TMDs a flexible platform for controlling emission at the nanoscale. The practical opportunity is to combine localized excitons, engineered optical fields and carefully designed routes for collecting or guiding photons. Progress toward useful integrated devices will depend on balancing field confinement and emitter alignment against metallic losses, while measuring radiative performance and signal routing separately.