Hybrid Plasmonic Nanostructures for Enhanced Single-Molecule Detection Sensitivity

Abstract

Biosensing applications based on fluorescence detection often require single-molecule sensitivity in the presence of strong background signals. Plasmonic nanoantennas are particularly suitable for these tasks, as they can confine and enhance light in volumes far below the diffraction limit. The recently introduced antenna-in-box (AiB) platforms achieved high single-molecule detection sensitivity at high fluorophore concentrations by placing gold nanoantennas in a gold aperture. However, hybrid AiB platforms with alternative aperture materials such as aluminum promise superior performance by providing better background screening. Here, we report on the fabrication and optical characterization of hybrid AiBs made of gold and aluminum for enhanced single-molecule detection sensitivity. We computationally optimize the optical properties of AiBs by controlling their geometry and materials and find that hybrid nanostructures not only improve signal-to-background ratios but also provide additional excitation intensity and fluorescence enhancements. We further establish a two-step electron beam lithography process to fabricate hybrid material AiB arrays with high reproducibility and experimentally validate the higher excitation and emission enhancements of the hybrid nanostructures as compared to their gold counterpart. We foresee that biosensors based on hybrid AiBs will provide improved sensitivity beyond the capabilities of current nanophotonic sensors for a plethora of biosensing applications ranging from multicolor fluorescence detection to label-free vibrational spectroscopy.

Keywords: electron beam lithography; hybrid materials; optical nanoantennas; plasmonic biosensing; plasmonics.

Figure 1

Sketch of the increased single-molecule (SM) sensitivity at high concentrations by different types of nanostructures. (a) In the absence of nanostructures the diffraction-limited confocal beam not only excites the target molecule but also generates a significant fluorescence background from the surrounding excited molecules, prohibiting the detection of single molecules. (b) Nanoantennas enhance the light in subdiffraction volumes increasing the signal from the target molecule without reducing the fluorescence background. (c) Antenna-in-box nanostructures simultaneously reduce the fluorescence background by attenuating the excitation beam and increase the signal enhancement as compared to conventional nanoantennas. Employing hybrid material platforms can boost both effects.

Figure 2

FDTD simulations of (a, b) the spectral excitation intensity enhancements G_I and (c) fluorescence signal-to-background ratio (SBR). The top rows of (a) and (b) show the dependence of the excitation intensity enhancement on the aperture diameter d and the optimal diameter (d_opt) of maximal enhancement at λ_exc = 640 nm. The relative enhancements ΔG_I in the central rows indicate the regimes in which the AiB apertures provide additional gains or losses as compared to isolated BNAs. In the bottom rows, the spectral excitation intensity enhancement for the optimized AiBs is compared to the isolated Au-BNA. The results are shown for the Au-Au-AiB (a) and Au-Al-AiB (b) designs. (c) Signal-to-background ratios at different dye concentrations for the confocal reference (yellow), the Au-BNA (purple), the Au-Au-AiB (blue), and the Au-Al-AiB (red). The top row shows the results when the aperture diameter is d = d_opt, and the bottom row corresponds to d = 300 nm. G_F and G_SBR are displayed in parentheses and indicate the fluorescence and signal-to-background gain, respectively, of the corresponding configuration as compared to the confocal reference. The target dye is located in the hot-spot center and aligned with the excitation polarization for maximum enhancement and emits at λ_emi = 676 nm. Due to the similar SBR, the blue line is hidden behind the red line in the upper plot of (c).

Figure 3

Schematic of the EBL-based fabrication process (a) and SEM images of the Au-Au-AiB and hybrid Au-Al-AiB platforms (b, c). (a) The fabrication process consists of an automated two-step overlay EBL process with a positive- and negative-tone resist. In both steps, the resists are spin-coated on a glass coverslip, exposed in the EBL step, and developed before the metal deposition and lift-off. The only conceptual difference in step II is the use of a negative-tone resist and the automated overlay alignment using the markers fabricated in step I. (b) SEM images of individual Au-Au-AiB and hybrid Au-Al-AiBs for different aperture diameters d. The measured BNA length l and gap distance g are shown on the bottom left of (b). (c) SEM images of the 10 × 10 AiB arrays. The red arrows on the right image of (c) indicate the scanning marks from the alignment process. The SEM images are equally contrast adjusted.

Figure 4

Experimental and simulated spectral transmission cross sections of the Au-Au-AiBs (a–c) and Au-Al-AiBs (d–f). The spectral transmission cross sections σ̂_t are normalized to the aperture area and measured for the incoming light being polarized parallel (a, d) and perpendicular (b, e) to the BNA axis. In (c, f) the differences of the transmission cross sections for parallel and perpendicular polarization are shown. The experimental data show the mean transmission cross sections over 40–64 nominally equivalent nanostructures in the arrays. The simulations are using illumination by a plane wave at normal incidence, whereas a NA = 1.34 illumination was used for the experimental data.

Figure 5

Fluorescence enhancement analysis of Alexa Fluor 647 embedded in PMMA for the confocal reference and the different nanostructure types. (a) Exemplary time trace C(t) from a 60 s spot measurement on Au-Al-AiB deduced from TCSPC data with a 100 ms binning time. The raw time trace was baseline corrected before detecting the blinking or bleaching step height ΔC of a single molecule through an automated algorithm. The fluorescence enhancement G_F is computed through the normalization of ΔC as described in the main text. (b) Histogram of the fluorescence enhancement for the confocal reference measurements. G_F = 1 was chosen to be at the q = 0.95 quantile of the confocal distribution. (c) Fluorescence enhancement histogram for the BNAs showing 10–22 × fluorescence enhancement (q = 0.95–1). (d) Diameter-dependent histogram of G_F for Au-Au-AiBs with the maximum enhancement (red crosses) and 0.95 quantile (solid line) overlaid. (e) Same as (d) but for the hybrid Au-Al-AiBs. While the Au-Al-AiBs provide a fluorescence enhancement of 26–50×, the Au–Au-AiBs provide a fluorescence enhancement more similar to that of Au-BNAs of 10–26×.

Figure 6

Fluorescence decay analysis of Alexa Fluor 647 embedded in PMMA for the confocal reference and the different nanostructure types. (a) Exemplary measured decay curves (light gray) with the corresponding reconvolution fits (colored) and number of exponentials n_exp. (b) Decay times τ⁽ⁱ⁾ and relative weights w_rel⁽ⁱ⁾ of the ith component for different aperture diameters of the Au-Au-AiB platform. The solid red lines indicate the averages for the Au-Au-AiBs, while ⟨τ⁽ⁱ⁾⟩ and ⟨w_rel⁽ⁱ⁾⟩ indicate the average decay time and relative weights of the confocal reference (subscript c) and Au-BNA (subscript n). (c) shows the same as (b) but for the hybrid Au-Al-AiB platforms.