A Design Strategy for Surface Nanostructures to Realize Sensitive Refractive-Index Optical Sensors Authors: Masanobu Iwanaga Published: 2023 ABSTRACT Refractive-index optical sensors have been extensively studied. Originally, they were surface plasmon resonance sensors using only a flat gold film. Currently, to develop practically useful label-free optical sensors, numerous proposals for refractive index sensors have been made using various nanostructures composed of metals and dielectrics. In this study, we explored a rational design strategy for sensors using surface nanostructures comprising metals or dielectrics. Optical responses, such as reflection and transmission, and resonant electromagnetic fields were computed using a numerical method of rigorous coupled-wave analysis combined with a scattering-matrix algorithm. As a result, good performance that almost reached the physical limit was achieved using a plasmonic surface lattice structure. Furthermore, to precisely trace the refractive-index change, a scheme using two physical quantities, resonant wavelength and reflection amplitude, was found to be valid for a 2D silicon metasurface. 1. Introduction Optical sensors are being developed for diverse targets, from gaseous molecules to large biomolecules such as antibodies and DNA. A significant feature of optical sensors is their direct contact with target molecules and the magnification of detection signals owing to their resonantly enhancing mechanisms. Considerable efforts have been devoted to the development of practical optical sensors for various applications. Refractive-index optical sensors are label-free sensors that enable evaluation of the refractive index of the medium with which the optical sensors are in contact. Originally, refractive-index sensors were validated using surface plasmon resonances (SPRs) in the visible range [1]. SPRs are typically observed using a gold thin film deposited on a glass prism at deep oblique p-polarized incidence. In addition to the refractive index of the surrounding medium, SPR sensors can evaluate the mass of molecules immobilized on the outermost surface of gold, which makes them useful for quantitatively analyzing biomolecule reactions involving immobilization and dissociation [2]. Thus, refractive-index sensors have the potential to function as mass analyzers for immobilized molecules. The requirement of an elaborate optical configuration that includes a total-reflection setup is a drawback of SPR sensors. Therefore, the development of refractive-index sensors that can operate in compact and simple optical configurations using surface nanostructures has been explored in the visible and near-infrared ranges [3,4,5,6,7,8,9,10,11,12,13]. From a similar motive, fiber-based refractive-index sensors were studied [14,15,16,17,18]; these sensors have long light propagating paths and high responsivity in the near-infrared range, while the sensing range is mostly limited to a narrow refractive-index range of 1.35–1.40. Despite the numerous proposals for refractive-index sensors, designing the high-performance sensors in a straightforward manner remains a challenge. Here, we first arranged the design flow and then studied several plasmonic and dielectric surface nanostructures that exhibited explicit responses as refractive-index sensors in the wavelength range of 500–1100 nm, where optical measurements are most cost-effective using silicon photon detectors. In this study, the physical limitations of the refractive-index sensors were clarified. Furthermore, one of the plasmonic sensors was found to exhibit a sensing performance that reached the physical limit. Other surface nanostructures stemming from ideas differing from the nearly ideal plasmonic sensor were also studied to examine their suitability as refractive-index sensors. A flow to design the optical sensors is shown in Figure 1. The goal is to realize high-performance practical refractive-index sensors. First, a basic structure, either periodic or non-periodic, is chosen. If periodic structures are chosen, the next step is whether or not to conduct designs based on dispersion: (i) if yes, feasible periodic structures realizing specific dispersions are designed, tested using a numerical simulation method, and evaluated in terms of performance as a refractive-index sensor; (ii) if no, nanostructures are designed based on interest and/or preference by each researcher. From this reason, Mie resonances in metallic nanoparticles [4,5], local resonances in metallic/dielectric nanostructures [3,6,7,8,9,10], and bound states in the continuum (BIC) [11,12,13] have often been studied. Artificial surfaces were designed by assembling the nanostructures on a substrate; subsequently, the surface structures were tested using the simulation method, and their performance was evaluated. These processes continued until an appropriate design was obtained. Various algorithms have been used to investigate artificial nanophotonic structures. Recently, inverse designs have gained popularity and have been used for metalenses [19,20,21], transmission waveguides [22,23], and metasurfaces [24,25]. Other search methods have been proposed. Nonempirical searches for nanostructures with optical functions have been used to obtain new structures [26]. Genetic algorithms have also been used to search for photonic structures [27,28,29,30,31,32]. However, to our knowledge, refractive-index optical sensors have not yet been designed using these search algorithms. This is partially because multifactors, including constituent materials, should be simultaneously optimized for refractive-index sensors, which complicates the search algorithm and makes its implementation difficult. 2. Methods and Material Parameters In this article, we address periodic nanostructures and employ the rigorous coupled-wave analysis (RCWA) [33], which incorporates a scattering-matrix algorithm [34] to stably compute arbitrary stacked structures. The results of the numerical method that was coded in Fortran 90 and executed using multiparallel implementation acceleration on supercomputers were consistent with experimental data in various cases [35]. The RCWA solves Fourier-transformed Maxwell equations for periodic structures and approximates the exact solutions using the inverse Fourier factorization and truncating the Fourier expansion [33]. To conduct realistic computations, the material parameter, permittivity, were taken from the literature [36,37], and the measured permittivity was used for Au and Si. Among the various metals, Au was chosen because of its high chemical stability. Si was selected because of its high permittivity. The permittivity of SiO2 was set to a representative value of 2.1316. 3. Results In this section, concrete designs are numerically studied. 1D and 2D surface-lattice structures are addressed in both plasmonic and all-dielectric cases. The physical limit of refractive-index sensing is explicitly described in Section 3.1. 3.1. 1D Plasmonic Surface Lattices The dispersions of light are illustrated on the (k,ℏω) planes in Figure 2a, where k and ω denote the wavenumber and angular frequency, respectively. On the left, the light cone (which is the dispersion in uniform media illustrated in the box below) for refractive index n=1.0 is drawn with black lines, while that for n>1 is shown with green lines for comparison. On the right, the 1D lattice structure of periodic length a is depicted in the box below, where we assume that the periodic structure (orange) is infinitely long perpendicular to the sheet of paper. The dispersions in Figure 2a show four lines located inside the light cones (dashed lines). This effect is often referred to as the folding of dispersion into the first Brillouin zone, that is, −π/a10,000 nm/RIU) was claimed in an early stage of the research [47]. In the Si surface lattices in Figure 4 and Figure 5, the resonant modes sensitive to the refractive index show quadratic responses that originate from photonic band structures. In the 2D Si lattice, the reflectance varies significantly under normal incidence, which makes refractive-index sensing feasible, similarly to the commercial SPR instrument. To our knowledge, such a large reflectance change has not been reported for Si surface nanostructures, including Si metasurfaces. Thus, the features of 2D Si lattices can contribute to practical refractive-index sensors in a compact optical setup equipped with only a Si photon detector. The Si metasurfaces reported to data [11,12,13] were designed to work at telecom wavelengths of approximately 1.55 µm. The RoL was lower than those of the plasmonic lattices in Table 1, probably because the BIC associated with the high-quality factor was focused on, and the RoL was not optimized. At present, the compatibility of high-quality-factor BIC and high RoL is unclear. We refer to a hollow-core fiber-based refractive-index sensor [16] listed in Table 1. A linear-response approximation at a very narrow range of refractive index of 1.43–1.44 can result in a large value of 19014 nm/RIU. However, the fiber-based sensors usually show quadratic refractometric responses, and in general, the linear approximation is invalid. We note that the fiber-based sensors work only at a small range of refractive index, typically 1.35–1.40 [14,15,16,17,18] because the operation principle is transmission through a long optical paths (∼mm) and interference. Thus, the fiber-based sensors are not comparable to the surface nanostructure sensors functioning at refractive index of 1.0–1.5. Refractive-index sensors are explored in THz ranges. An simulated result [46] is listed in Table 1. In THz ranges, refractive indices of materials are considerably different from those in the visible and near infrared ranges, and tend to be significantly larger. Thus, the sensors in THz ranges are not directly comparable to those studied in this article. In addition, it is to be noted that the detection range is narrow, limited to n=1.34–1.44 [46]. The refractive-index sensors studied here have features in wide-range sensing for the range n=1.0–1.5, large RoL, and dynamical change in wavelength and reflectance. These features are considered to make the application for biomolecule analysis feasible in a similar manner to the current commercial SPR sensors. One benefit of these sensors enables simple and compact optical setup, compared with the SPR sensors, and will lead a portable reflective-index and biomolecule-analyzing sensors for diverse targets. 5. Conclusions Designs for high-performance refractive-index sensors have been explored for plasmonic/all-dielectric surface-lattice platforms. The physical limit of the refractometric response was first revealed, and concrete designs were then studied from the viewpoints of spectroscopy and resonant modes. Consequently, we found that the 1D plasmonic surface lattice is a nearly ideal refractive-index sensor, while the 2D plasmonic lattice is a high-performance sensor independent of incident polarizations. In addition, we showed that the 2D Si surface lattice exhibited a reflectance change for the refractive index, which was similar to a commercial SPR setup. Howver, the Si lattice operated at the normal incidence, whereas the SPR sensors required deep oblique and polarized incidence. Thus, the designs presented here are useful for constructing compact, simple, and cost-effective refractive-index sensing setups, which have potential to serve as portable and handy sensors outside laboratories.