Fig. 1. Structural characterization of nanoantennas and MoS2. (a) Schematic diagram of bowtie nanoantenna structure from the side top view, side sectional view, and top view. (b) SEM image of nanoantennas, with a spacing of 2 μm between adjacent nanoantennas. (c) Microscopic image of a nanoantennas array covered with a single layer of MoS2. (d) Raman shifts of MoS2 monolayer and MoS2 monolayer+nanoantenna where the difference between E2g1 and A1g is 19.0 and 20.5  cm−1, respectively. (e) Non-radiative recombination dominates in monolayer MoS2, with low quantum yield. Resonant nanoantennas enhance the competitiveness of radiative recombination, and make the radiation angle smaller, which is more conducive to top surface emission.

Fig. 2. Relationship between nanoantenna dimensions and optical properties. (a) Relationship between nanoantenna length and resonance wavelength. Other parameters remain unchanged with a=b=60  nm. (b) Relationship between nanoantenna width and resonance wavelength. Other parameters remain unchanged with a=b, c=300  nm. (c) Charge density distribution and electric field distribution of λeff antenna and λeff/2 antenna under 665 nm resonance condition. Simulation considers the enhancement of radiation sources by Purcell effect. The simulation size is cλeff=300  nm, cλeff/2=150  nm, and d=50  nm. (d) Relationship between gap spacing and resonance intensity for λeff antenna. Other parameters remain unchanged with a=b=60  nm, c=300  nm. (e) Simulation calculation results of Purcell factors for λeff and λeff/2 nanoantennas. (f) Radiation patterns of light sources on nanoantennas. (g) Relationship between sharpness of the inner side of antenna and resonance intensity. Other parameters remain unchanged with b=60  nm, c=300  nm, d=50  nm.

Fig. 3. Enhancement and modulation of PL in MoS2 by nanoantennas. (a) Integrated PL mapping of MoS2 on nanoantennas, integration range 1.80–1.85 eV. (b) Influence of different nanoantennas on PL. The specific dimensions are: bowtie antenna a=30  nm, b=60  nm, c=300  nm, d=50  nm, rectangular rod antenna a=b=60  nm, c=300  nm, d=50  nm, λeff/2 antenna a=b=60  nm, c=300  nm. (c) Time resolved PL test results of different nanoantennas. The size of the nanoantenna is the same as (b). (d) Variations in MoS2 PL enhancement for six types of nanoantennas depicted in the black box of (a), all with b=60  nm, c=300  nm. (e) Normalized scattering spectra of nanoantennas of different sizes and typical PL spectra of monolayer MoS2. All structures are rectangular antennas with dimensions of a=b and d=50  nm. (f) Influence of the antenna length and outer width on the PL peak. PL intensity is normalized for comparison. The size of the nanoantenna is the same as (e).

Fig. 4. Polarization characteristics of nanoantennas. (a), (b) Side view (top right), top view (center), and front view (bottom) of the spatial distribution of (E/E0)2 for the bowtie nanoantenna calculated using the FDTD method. The curve at the bottom shows the simulation results for (E/E0)2 along the x direction at y=0  μm and z=0.02  μm. To furnish more detailed information, logarithmic coordinates were used for the scale of the electric field map. The wavelength of the simulated light source is 665 nm. In (a), the incident polarization direction is parallel to the long axis. In (b), the incident polarization direction is perpendicular to the long axis. (c), (d) Experimentally measured relationship between PL intensity and light source polarization angle: (c) bowtie antennas, (d) cross antennas. (e) Enhancement of PL under unpolarized light excitation. (f) Raman spectra of MoS2 on different nanoantennas, with the size of the nanoantenna consistent with Fig. 3(b).