1. Introduction

Molecular fluorescence detection has various applications in many fields such as biosensors, early diagnosis, bioimaging, single photon source and so on [1,2]. However, the low signal to noise ratio (SNR) limits the detection efficiency of single molecule fluorescence. Thus, how to control the photoluminescence process, which can be divided into excitation and emission processes, becomes a main topic in this field. As we know, metallic nanostructures have novel characteristics, i.e. surface plasmon (SP) and localized surface plasmon (LSP) [3,4], which can be exploited to modify the photoluminescence process strongly [5–7]. For instance, metallic nano-apertures are able to confine electromagnetic field within the aperture due to LSP effect, and the resulted high intensity field can enhance the excitation efficiency. Also, the nano-apertures have an extra advantage of suppressing background [8–10]. In addition, plasmonic gratings have attracted many attentions due to their unique SP characteristic. The plasmonic gratings cannot only enhance specific SP mode but also modify its propagation direction. For example, plasmonic gratings are able to modify the light transmission from single nano-aperture [11]. Moreover, the hybrid nanostructure of plasmonic gratings and nano-aperture have been extensively investigated for single molecule fluorescence enhancement [12–16]. When the gratings coupling with the nano-aperture, it cannot only enhance the near-field excitation rate but also it can modify the emission rate and the far-field radiation pattern of the molecular fluorescence. It is also found that the resonance wavelength is dependent on their geometry parameters, such as the groove depth and grating period, which provide a way to optimize the optical response corresponding to specific fluorescence molecule [17,18]. While conventional plasmonic gratings response effectively only in a narrow spectral band. Therefore, there is a tradeoff between the excitation and emission processes when the Stokes shift of the photoluminescence process is relatively large. Although asymmetric dual-face grating antenna was proposed in theory to control the local excitation enhancement, the collection efficiency, and the quantum efficiency separately, the goals to optimize the fluorescence enhancement still remain an open challenge in experiment.

In this study, we demonstrated that the nano-apertures associated with stair-gratings have high surface enhancement factor and better beaming effect in comparison to the conventional ones. In contrast to the conventional gratings, we propose to excavate a rectangle part of corrugations and make the cross profile likes a stair-grating. The schematic of stair-gratings is shown in Fig. 1 as a hybrid of two gratings with different period or depth. Thus, a new periodic parameter is introduced into the plasmonic grating, which could increase the optical response both at the excitation and emission bands simultaneously. In experiment, we used fluorescence correlation spectroscopy (FCS) to analyze the fluorescence trace and the fluorescence count rate per molecule. The nano-apertures with stair-gratings truly presented higher enhancement effect. In addition, the emission angular patterns were measured with the back focal plane (BFP) imaging method, presenting a narrower directionality for the stair-gratings. By employing finite-difference time-domain (FDTD), the directional emission patterns and near-field enhancement were calculated. The numerical simulations are in good agreement with the experiments. The proposed stair gratings provide a flexible way to control the enhancement and beaming effect of the molecule fluorescence.

 

Fig. 1 Schematic of proposed stair-gratings. We excavate a rectangle part of the corrugations and make it like a stair. There are two new geometry parameters which can use to tune the optical response. (b) and (c) SEM cross profile of the common grating and stair-grating separately.

Download Full Size | PDF

2. Experimental methods

In the experiment, optical measurement methods are set up on an integrated microscopy system (NTEGRA Spectra, NT-MDT). The schematic of optical setup is shown in Fig. 2(a). We can measure white light dark-field scattering [19,20], photoluminescence(PL), FCS [12,21], and fluorescence radiation patterns [22,23] of the same single nano-aperture in situ. In all measurements, we used an oil-immersion objective lens (N.A. 1.49, 60 × , TIRF, Olympus). And the angular patterns of the fluorescence emission were obtained with the back focal plane (BFP) imaging method. A CW laser at wavelength of 632.8 nm was used as the excitation light with excitation power of ~60 μW. On the other hand, we used Alexa 647 in the solution with concentration of ~1μM. To prevent from molecular adsorption due to the local charges, we used phosphate aqueous buffered solution. To prepare the metallic nanostructures, there are four steps as shown in Fig. 2(c): (1) Deposit Au thin film on the cover glass by using magnetic sputtering; (2) Use focus ion beam (FIB) to penetrate through the Au film and etch into the glass to fabricate the corrugations of the gratings; (3) Deposit Au film again onto the samples in order to fill the corrugations; (4) Fabricate the central nano-aperture with FIB milling. By using the Pt deposition which is integrated in the FIB system, we can easily identify the fabricated nanostructure under the SEM. The SEM images of the common and stair gratings are shown in Figs. 1(b) and 1(c) separately. We find that the corrugations of the stair-gratings and the common gratings can be fabricated well as expected. In our experiments: the corrugation groove depth d = 200 nm, width a = 220 nm, the height of the excavated part d₂ = 100 nm, the width of the excavated part a₂ = 110, the central aperture diameter (same with bare aperture) D = 250 nm, there are 5 grooves for both the stair and common gratings, and Au film thickness H = 300 nm.

 

Fig. 2 (a) Schematic of optical experiment setup. (b) Optical confocal scanning image of a sample containing bare apertures, nano-apertures surrounded with common and stair grating. (c) Fabrication procedure of the nano-apertures with stair-gratings.

Download Full Size | PDF

3. Results and discussion

First of all, we use the white light dark-field scattering method to characterize the response of the nanostructures. The results are shown in Fig. 3(a). As we can see, in the range from 625 nm to 675 nm, both the stair-gratings (red line) and the common gratings (blue line) present a broad resonance. Obviously, the stair-gratings have higher scattering intensity which implies it has stronger optical response when comparing to the common gratings. Correspondingly, the Alexa 647 dye molecules used in the present experiment also emits in this range. Thus, spectral response of the nanostructures covers not only the excitation wavelength of 632.8 nm but also the emission spectral range of 640 ~700 nm. Figure 3(b) shows the fluorescence spectra of the molecules being confined within the nano-apertures. The results from three kinds of nanostructures are compared: bare nano-apertures, nano-apertures with common gratings, and nano-apertures with stair-gratings. As shown in Fig. 3(b), the nano-apertures with gratings modify the fluorescent spectral shape slightly when comparing to the fluorescence spectrum obtained from the dye molecules in free solution (data not shown). At wavelength ~700 nm, there is a “shoulder” for both the common and stair gratings. This spectral variation can be due to the strong interaction between the dye molecules and the nanostructures [10]. The PL spectrum of the dye molecules in free solution is also plotted for comparison.

 

Fig. 3 (a) Scattering spectra of nano-aperture with stair and common gratings. (b) Fluorescence spectra of the molecules within different nanostructures: stair-gratings, common gratings and bare aperture. PL spectrum in free solution is also plotted for comparison.

Download Full Size | PDF

In order to obtain normalized fluorescence count rate per molecule, it is necessary to obtain the average number of the dye molecules in the nano-apertures. In the following, we measured the fluorescence intensity from single nano-apertures with two avalanche photodiodes (SPCM-AQRH-16-FC, PerkinElmer) and obtained the FCS curves through cross-correlation to suppress after-pulsing (correlator.com, US). The results are shown in Fig. 4(a). The FCS curves of three kinds of nanostructures have little difference, which means that the average number of the molecules is almost the same because the different nanostructures actually have the same nano-aperture size. Nevertheless, we notice that the result of the bare aperture has a slightly higher G(0) which probably due to less background noise than the others. According to three dimensional Brownian diffusion model, we have the formula: $G\left(\tau \right)=1+\frac{1}{N}{\left(1-\frac{B}{F}\right)}^2\left[1+n_{T}exp\left(-\frac{\tau }{{\tau }_{bT}}\right)\right]{\left(1+\frac{\tau }{{\tau }_d}\right)}^{-1}{\left(1+s^2\tau /{\tau }_d\right)}^{-1/2}$ [8,10,21], where N is the total number of molecules, F the total signal, B the background noise, n_T the amplitude of the dark state population, b_T the dark state blinking time, τ_d the mean diffusion time, and s the ratio of transversal to axial dimensions of the analysis volume. We used this formula to fit the FCS curves and calculate the average number of molecules. After fitting the curves, we obtained the fluorescence count rate per molecule. As shown Fig. 4(c), we provide the averaged count rate and standard deviation in different structures, summarized from three stair-gratings, four common gratings and four bare apertures. In average, in comparing to the bare aperture, the stair-gratings can reach 2-fold fluorescence enhancement factor. And the stair-gratings perform better than common gratings. The stair-gratings have the highest count rate, i.e. the stair-gratings can still promote 20% of the count rate than the common ones. Both the count rate of the common gratings and the stair-gratings are much higher than that of the bare nano-apertures. It implies that the stair-gratings are able to significantly enhance the fluorescence.

 

Fig. 4 (a) Fluorescence correlation spectroscopy curves of three different nanostructures. (b) Normalized representative fluorescence intensity trace, and (c) Normalized fluorescence count rate per molecule for different structures.

Download Full Size | PDF

Additionally, it is necessary to discuss the molecular far-field radiation pattern modified by the different nanostructures, since it is another important factor to determine the collection efficiency during the surface enhanced fluorescence. Figure 5 shows the fluorescence far-field radiation patterns with the BFP imaging method, and all images with the same color bar. Figure 5(a) shows the results of the bare apertures and the distribution of the pattern is homogeneous. It implies that the molecular radiation within the bare aperture spread in a broad solid angle. The patterns of the common gratings are shown in Fig. 5(b). As we can see, the fluorescence signals concentrate more in the center, which means that the common gratings are able to boost the collection efficiency. Furthermore, as shown in Fig. 5(c), we find that the fluorescence signals are highly confined in the center when comparing to the former two nanostructures. These results imply that the stair-gratings can confine the molecular radiation in a narrower solid angles. We notice that the experimental results above are based on the use of the high N.A. objective (N.A. = 1.49) which can collect the fluorescence signal very efficiently. Hence, we can obtain a higher enhancement factor if we use an air objective to replace the oil immersion objective.

 

Fig. 5 Molecular radiation patterns from different structures (a) Bare nano-aperture (b) Nano-aperture with common gratings (c) Nano-aperture with stair-gratings.

Download Full Size | PDF

Furthermore, we employed the FDTD method to simulate the phenomena qualitatively [24]. In our simulations: the corrugations height d = 200 nm, width a = 220 nm, the height of the excavated part d₂ = 100 nm, the width of the excavated part a₂ = 110, the central aperture diameter (same with bare aperture) D = 250 nm, Au film thickness H = 300 nm. For the wavelength from 633 nm to 672 nm, we calculated the near-field enhancement factor within the nano-apertures and the far-field radiation patterns. It should be note that we only present the results for the dipole is parallel to the Au film, because such orientation is dominated over the perpendicular ones. For instance, the power radiative of parallel dipole is about two order of magnitudes higher than the perpendicular one. As shown in Fig. 6(b), the common gratings perform well at the wavelength of 633 nm, but the efficiency of the common gratings decrease for longer wavelength significantly. It is due to the width of the spectral response is narrower. In contrast, as shown in Fig. 6(c), the stair-gratings perform well at the wavelength of 633 nm. And the performance of the stair-gratings at wavelength of 650 ~670 nm are still good enough for good beaming effect although the excitation enhancement factor is lower slightly. These simulation results are in good agreement with the experiments, i.e. the stairs-gratings can confine the radiation better than the common gratings. It should be noted that the groove depth in the present study is pretty deep according previous studies [12,25]. It is still necessary to optimize the parameters of the stair grating, e.g. groove depth and geometry of excavated part so on, for better performance. Nevertheless, based on current experimental and numerical simulations, it is reliable to conclude that the stair grating can work better rather than the common grating.

 

Fig. 6 Simulated far-field radiation patterns at different wavelengths for (a) Bare nano-aperture (b) Nano-aperture with common gratings (c) Nano-aperture with stair-gratings, and near-field intensity enhancement indicated in each top-right corner correspondingly.

Download Full Size | PDF

4. Conclusions

In conclusion, we design a new type of gratings called as stair-gratings and combine them with nano-aperture for surface enhanced fluorescence detection. In comparison with the conventional ones, we find that the detected fluorescent intensity by the stair-gratings is higher than the common grating. And narrower directionality by the stair-gratings would enable the detection of molecular fluorescence with low N.A. objective. All these factors allow a higher SNR and higher detection efficiency of single molecule fluorescence. We also employed the FDTD method to simulate the near-field enhancement and far-field radiation patterns. The simulation results are in good agreement with the experimental results. Our research contributes to understanding to the plasmonic gratings for optimizing the surface enhancement process of photoluminescence process.