Optimisation and Precision Tuning of Localised Surface Plasmon Resonance in AuFON Systems

Technical Summary: Optimisation and Precision Tuning of Localised Surface Plasmon Resonance in AuFON Systems

Problem Statement
Metal film on nanosphere (MFON) plasmonic systems are widely utilized for molecular detection applications, including biosensing and surface-enhanced Raman spectroscopy (SERS). While these substrates offer high signal amplification, a significant limitation in their practical application is the frequent lack of optical optimisation. Often, the incident radiation used in experiments is not tuned to resonate with the localised surface plasmons (LSPR) of the specific nanostructure geometry. This mismatch limits the efficiency of optical amplification. Consequently, there is a need for a comprehensive characterisation to precisely identify the energy positions where plasmons resonate and to understand how these resonances depend on nanostructure dimensions and optical conditions.

Methodology
The authors employed a combined approach of experimental fabrication, optical characterisation, and numerical simulation to study gold-film on nanosphere (AuFON) systems.

• Fabrication: AuFON substrates were created by self-assembling polystyrene nanospheres (500–800 nm diameter) onto a 100 nm gold-coated glass plate. A subsequent gold layer (~140 nm thick) was deposited via physical vapour deposition, creating a metal film conforming to the nanosphere array.
• Experimental Characterisation: Reflectivity measurements were conducted using a variable-angle spectroscopic ellipsometer. The setup utilized a 100 µm spot size to probe the plasmonic modes. Scanning Electron Microscopy (SEM) was used to verify the morphological ordering of the nanospheres.
• Numerical Simulations: Finite element method (FEM) simulations were performed using COMSOL Multiphysics. The model consisted of a unit cell representing half a gold nanosphere on a flat gold surface. The simulations solved Maxwell's equations in 3D, varying the nanosphere diameter (500–900 nm), the angle of incidence ($\theta$), the azimuthal angle ($\phi$), and the polarisation (TM and TE).

Key Results
The study identified two distinct localised surface plasmon resonance modes (LSPR1 and LSPR2) in the 600–900 nm wavelength range. The characterisation yielded the following specific findings:

• Resonance Identification: For a 600 nm diameter sample, resonances were observed at 664 nm (LSPR1) and 830 nm (LSPR2) for TM polarisation, and at 660 nm and 808 nm for TE polarisation. For 500 nm samples, resonances occurred at shorter wavelengths (approx. 574–591 nm for LSPR1 and 700–740 nm for LSPR2).
• Polarisation Dependence: The resonance energy of both modes showed a significant dependence on light polarisation. Differences of approximately 21 nm (LSPR1) and 30 nm (LSPR2) were observed between TM and TE polarisations. TM polarisation generally exhibited higher excitation efficiency, particularly under oblique incidence.
• Geometric Dependence: The resonance energy is highly sensitive to the nanosphere diameter. As the diameter increased from 500 to 900 nm, both modes exhibited a red shift (dispersion towards longer wavelengths). The LSPR2 mode remained more stable against diameter increases compared to LSPR1, which weakened for diameters exceeding 800 nm.
• Angular Dependence:
  • Azimuthal Angle ($\phi$): Neither experimental nor simulation results showed significant changes in resonance energy when varying the azimuthal angle ($\phi$) from 0° to 90°. This confirms the high symmetry of the honeycomb-ordered nanostructures.
  • Incidence Angle ($\theta$): For TE polarisation, resonance energy showed low dependence on the incidence angle. However, for TM polarisation, the LSPR2 mode exhibited a gradual red shift with increasing $\theta$, attributed to the coupling of localised modes with propagating Bragg modes due to the periodic arrangement of the nanospheres.
• Field Distribution: Spatial distribution analysis revealed that the electric field is concentrated primarily at the junctions between nanospheres and their upper surfaces. The LSPR2 mode consistently demonstrated higher electric field intensification than LSPR1, indicating a superior capacity for light concentration.

Key Contributions
The paper provides a detailed characterisation of AuFON systems that bridges the gap between theoretical simulation and experimental reality. The primary contributions include:

1. Precise Tuning Criteria: Establishing clear criteria for identifying resonance peaks based on nanosphere size and incident light conditions.
2. Mode Differentiation: Distinguishing between two specific plasmonic modes (LSPR1 and LSPR2) and characterising their distinct responses to polarisation and geometry.
3. Optimisation Strategy: Demonstrating that the LSPR2 mode is the most efficient for amplification applications due to its stability and higher field concentration, and showing that this mode can be precisely tuned by adjusting the nanosphere diameter and polarisation.
4. Validation of Symmetry: Confirming that the azimuthal angle does not significantly affect resonance, simplifying the practical alignment requirements for these substrates.

Significance and Claims
The authors claim that this work addresses a critical bottleneck in the application of MFON systems: the lack of proper optical optimisation. By providing a methodology to precisely identify and tune the spectral position of plasmonic modes, the study enables more efficient amplification of optical signals. The authors state that these results facilitate the design of more efficient and reproducible substrates tailored for specific applications, particularly molecular detection via SERS. The work is presented as a step toward advancing technologies in molecular diagnostics and environmental monitoring by ensuring that the incident radiation is optimally matched to the nanostructure's resonance.