Fluorescence engineering in metamaterial-assisted super-resolution localization microscope

Abstract

Single-molecule localization microscopies have gained much attention for their efficient realization of a sub-diffraction-limit imaging with the resolution down to the 10-nm range. In contrast to conventional localization microscopes, which rely on particular fluorescent probes in specific conditions, metamaterial-assisted super-resolution microscopies can be implemented with any fluorescent dye under general conditions. Here, we present a systematic study of fluorescence engineering in metamaterial assisted localization microscopy by using cyclic group metasurfaces coated with a fluorescent film. Tailored variations are clearly demonstrated in both the photoluminescence intensity and the photobleaching lifetime of fluorophores based on the spatially varied Purcell effect near the metasurfaces. The enhanced emissions and blinking dynamics of the fluorophores on these metasurfaces lead to an increased signal-to-noise ratio, and therefore give rise to a super-resolution localization image with 0.9-nm localization accuracy. Our results are not only beneficial for super-resolution localization imaging but also push the control of light-matter interactions beyond the diffraction limit.

Keywords: Purcell effect; enhanced fluorescence; metamaterials; super-resolution imaging.

Figure 1:

Metamaterial assisted localization microscopy (MALM). (a) Cyclic group C₁ metasurface adopted for the MALM. Scanning electron microscope (SEM) image of the fabricated C₁ metasurface (top panel) shows the first few spiral rings with the radius of the Nth ring r = (pN – 300) nm, where the separation p = 600 nm, as designed (middle panel). Each spiral ring is composed of 8 arcs with widths from W ₁ = 45 nm to W ₈ = 150 nm with a 15-nm increment (bottom panel). Each arc functions as a nanoantenna. Scale bars: 1 μm. (b) Schematic illustration of the MALM setup showing its two main components: a fluorescence microscope and a metamaterial substrate coated with spontaneously blinking fluorophores. Here, the metamaterial substrate is the C₁ metasurface (a), on top of which a P3HT film is spin-coated and provides the spontaneously blinked emissions. The P3HT film was excited by a 488-nm CW laser, and the emission was collected by an objective lens (40 × /0.75 NA) and then recorded by a CCD camera after passing through a 650-nm long-pass filter. (c) Diffraction-limited fluorescence blinking images used for the MALM super-resolution image reconstruction. There are 300 frames of images recorded at a rate of 5 frames per second (5 fps). (d) Fluorescence engineering in the MALM imaging process. Owing to the Purcell effect near the arc nanoantennas, the surrounding fluorescence processes are altered: Enhanced PL intensity, accelerated spontaneous blinking, and improved photostability. (e) Example of the engineered fluorescence at two pixel positions. Compared to the case when the P3HT film is coated on top of glass (gray dots), the altered fluorescence dynamics in the MALM is clearly visible (red dots).

Figure 2:

Fluorescence engineering with MALM metasurface substrate. (a, c, e) PL intensity image of the P3HT-coated C₁ metasurface. Enhanced PL intensity is clearly visible (a). It directly shows that this intensity enhancement is determined by the width of the arc nanoantennas (c) and independent of their length (e). Note that the PL intensity of P3HT near the arc nanoantenna with a width of W ₈ = 150 nm (W ₁ = 45 nm) is about 4 (2) fold higher than that on the glass substrate. (b, d, f) Photobleaching lifetime image of the P3HT-coated C₁ metasurface. Improved photostability is clearly visible (b). It directly shows that this stability improvement is determined by the width of the arc nanoantennas (d) and independent of their length (f). Note that the photobleaching lifetime of P3HT near the arc nanoantenna with a width of W ₈ = 150 nm (W ₁ = 45 nm) is about 5 (2) fold longer than that on the glass substrate.

Figure 3:

Super-resolution image of MALM. (a, b) Diffraction-limited image (a) and the MALM super-resolution image (b) of the P3HT-coated C₁ metasurface. Scale bars: 25 μm. (c) Radial intensity distributions of the diffraction-limited image (black curve) and the MALM super-resolution image (green curve) along the lines indicated in (a) and (b), respectively. In addition to the sub-diffraction-limited resolution, the MALM super-resolution image also exhibits high contrast. (d, f) Magnified diffraction-limited images of two regions of interest. The spiral rings cannot be resolved. Scale bars: 5 μm. (e, g) Magnified MALM super-resolution images of the same regions of interest. The spiral rings are clearly visible. Scale bars: 5 μm.

Figure 4:

Localization performance of MALM. (a) Localization results overlapped with the corresponding SEM image. The error bars show the localization accuracy. Scale bar: 600 nm. (b) Localization results visualized as a scatter plot. The standard deviation σ _x shows the localization precision.

Figure 5:

MALM super-resolution images of P3HT-coated C₃ and C₄ metasurfaces. (a–f) Diffraction-limited images (a and d), magnified diffraction-limited images of a region of interest (b and e), and the MALM super-resolution images (c and f) of the C₃ and C₄ metasurfaces, respectively. Scale bars: 25 μm (a, d) and 5 μm (b, c, e, f).