Fluorescence enhancement in an over-etched gold zero-mode waveguide

Meiyan Wu; Wenzhao Liu; Jinyong Hu; Zhensheng Zhong; Thitima Rujiralai; Lidan Zhou; Xinlun Cai; Jie Ma

doi:10.1364/OE.27.019002

1. Introduction

Single molecule techniques can reveal individual molecule behaviors in biomolecular processes that are not easily accessible by ensemble measurements. Among them, single molecule fluorescence (SMF) method is one of the most widely used techniques. However, conventional SMF technique often requires much diluted samples and the allowed concentration for detecting the fluorescence from only a single molecule is generally at the level of nanomolar (nM) or below due to the large excitation volume limited by the optical diffraction. Yet most biologically relevant or physiological concentration requires to detect molecules in the micro- to milli-molar ranges [1], especially for monitoring the less strong biomolecular interactions with K_d >1 μM.

Zero-mode waveguides (ZMWs) are subwavelength apertures in a thin metal film patterned on a glass coverslip. Because the sizes of these apertures are smaller than the cut-off diameter of the fundamental excitation mode of the light which is allowed to propagate through the aperture, the excitation light is restricted at the bottom of the aperture, making it possible to detect one molecule at a time at high concentrations [2]. Until now, ZMWs have been employed in many areas, including single molecule DNA sequencing [3], real-time imaging of protein-protein interactions [4], spectroelectrochemical investigation of redox reactions [5], direct observation of translation processes [6] as well as protein-cytoskeletal filaments interactions [7].

However, the intensities of single molecule fluorescence in ZMWs are usually weak and vary significantly. The latter has also been shown to be closely related with the randomly distributed distances of the immobilized fluorescent molecules to the metallic walls where quenching often occurs [8–10]. In order to improve the sensitivity of ZMWs, the optimization of the single molecule fluorescence signals in ZMWs is essential. Different methods have been explored to either enhance the fluorescence intensity or improve the collection efficiency. For example, some studies utilized special nanostructures such as double nanohole apertures [11], bowtie nano-apertures [12], antenna-in-box [13,14], heterogeneous optical slot antennas [15,16] to enhance the local electric field and hence increase the single molecule fluorescence signals. Others tried to improve the collection efficiency by using finite-sized hexagonal arrays of nanoapertures [17], dense array of nanoholes [18], single nanoaperture surrounded by concentric surface corrugations [19–21] or parabola-shaped microreflector beneath the ZMWs [22]. Nonetheless, most of these methods require more complicated fabrication procedures than that used in making the conventional ZMWs (i.e. round nanoapertures) and therefore set up some technical limitations.

In addition to these advanced methods, there is actually an easy way to enhance both the fluorescence intensity and collection efficiency to finally improve the signal to noise ratio (SNR) of ZMWs, simply by over-etching the bottom of the ZMWs. Previously, Tanii et al has simulated the fluorescence intensity enhancement and SNR improvement in an Al-ZMW at different over-etching depths [23]. However, in their calculations, they only considered the contributions from the enhanced local excitation field and collection efficiency. The contribution from the modification of quantum yield of the fluorescent molecules was not taken into account. Later, Jiao et al reported the modification of the UV fluorescence lifetime of molecules within Al-ZMWs from both simulation and experiments [24]. The fluorescence enhancement in the UV range for an over-etched ZMW was also calculated. However, they did not measure the fluorescent enhancement and make comparison with the simulation.

Besides, although Al is extensively used as the metal film for ZMWs thanks to its short skin depth and high reflectivity, the chemical nature of Al is not stable. In contrast, Au is more stable in chemical properties and more amenable to surface modification. In addition, Au has stronger plasmonic effects in visible light region, which is especially valuable for the fluorescence enhancement in biological studies. It has been shown that Au-ZMWs with r = 100 nm can enhance the fluorescence in both the green and red regions while Al-ZMWs with r = 100 nm can enhance the green fluorescence but not the red [25]. Therefore, in this paper, using Cy3 and Cy5 as fluorescent probes, we investigate how over-etching affects the fluorescence enhancement in an Au-ZMW at various wavelengths both by simulation and FCS measurements. These results are expected to serve as an important guide for designing an optimized Au-ZMW used for single molecule detection in the future.

2. Numerical simulation of fluorescence enhancement

2.1 Theory

We use the fluorescence count rate per molecule (CRM) under the steady-state condition to characterize the fluorescence intensity, which is given by [26,27]

C R M = η ϕ \frac{σ I_{e}}{1 + I_{e} / I_{s}},

where

η

is the collection efficiency of the instrument,

ϕ

is the quantum yield of the fluorophore,

σ

is the adsorption cross-section, I_e and I_s denote the excitation and saturation intensity, respectively. When the fluorescence intensity is well below the saturated condition

(I_{e} \leq I_{s})

, the CRM can be simplified as [24]

C R M = η ϕ σ I_{e} .

In the free solution case, the CRM can be written as

C R M = η^{0} ϕ^{0} σ I_{e}^{0} .

The fluorescence decay rate of the molecule is given by

k_{t o t}^{0} = k_{r}^{0} + k_{n r}^{0},

where

k_{r}^{0}

and

k_{n r}^{0}

are the fluorescence radiative and non-radiative decay rates with (⁰) indicating the situation in the absence of metallic structure (i.e. the free solution condition). The quantum yield can thus be expressed as

ϕ^{0} = \frac{k_{r}^{0}}{k_{t o t}^{0}} = \frac{k_{r}^{0}}{k_{r}^{0} + k_{n r}^{0}} .

When the fluorescent molecule is placed in the ZMW, due to the coupling between the molecule and ZMW, an additional non-radiative rate, k^’_abs, is introduced to account for the Ohmic loss of the metallic structure and non-radiative energy transfer to the free electrons in the metal surface [28]. Here, we used (’) to indicate the situation in presence of metallic structure. The new decay rate then changes to ${k^{'}}_{t o t} = {k^{'}}_{r} + {k^{'}}_{n r} + {k^{'}}_{a b s}$ and the quantum yield is modified to be

ϕ^{'} = \frac{{k^{'}}_{r}}{{k^{'}}_{r} + {k^{'}}_{n r} + {k^{'}}_{a b s}} .

Next, we assumed that the ZMW does not influence the intrinsic non-radiative decay rate $k_{n r}^{0}$ , i.e. ${k^{'}}_{n r} = k_{n r}^{0}$ . The enhancement of quantum yield by ZMW is then given by

f_{ϕ} = \frac{ϕ^{'}}{ϕ^{0}} = \frac{{k^{'}}_{r}}{k_{r}^{0}} \frac{k_{r}^{0} + k_{n r}^{0}}{{k^{'}}_{r} + k_{n r}^{0} + {k^{'}}_{a b s}} = \frac{{k^{'}}_{r}}{k_{r}^{0}} \frac{τ^{'}}{τ^{0}} = \frac{Γ}{F_{p}} .

where

τ^{'}

and

τ^{0}

are the lifetimes of the fluorophores in ZMW and free solution conditions, respectively.

Γ = \frac{{k^{'}}_{r}}{k_{r}^{0}}

is the radiative enhancement.

F_{p}

is the Purcell factor, which characterizes how the spontaneous emission rate of the fluorescent molecule is modified by its environment [29,30] and is given by

F_{p} = \frac{τ^{0}}{τ^{'}} = \frac{{k^{'}}_{t o t}}{k_{t o t}^{0}} = \frac{{k^{'}}_{r} + k_{n r}^{0} + {k^{'}}_{a b s}}{k_{r}^{0} + k_{n r}^{0}} .

Notice that $k_{n r}^{0}$ in Eq. (8) is usually difficult to calculate directly. However, with Eq. (5) we can get

k_{n r}^{0} = \frac{1 - ϕ^{0}}{ϕ^{0}} k_{r}^{0} .

Substituting Eq. (9) into Eq. (8) to eliminate $k_{n r}^{0}$ , the expression of the Purcell factor now changes to

F_{p} = \frac{{k^{'}}_{r} + \frac{1 - ϕ^{0}}{ϕ^{0}} k_{r}^{0} + {k^{'}}_{a b s}}{k_{r}^{0} + \frac{1 - ϕ^{0}}{ϕ^{0}} k_{r}^{0}} = \frac{{k^{'}}_{r} + {k^{'}}_{a b s}}{k_{r}^{0}} ϕ^{0} + 1 - ϕ^{0} .

We further define the parameter $F_{p 0} = \frac{{k^{'}}_{r} + {k^{'}}_{a b s}}{k_{r}^{0}}$ , which represents the Purcell factor of an ideal emitter (i.e. $k_{r}^{0} ≫ k_{n r}^{0}$ thus $ϕ^{0} = 1$ ). After that, Eq. (10) can be rewritten as

F_{p} = 1 + ϕ^{0} (F_{p 0} - 1) .

The enhancement of quantum yield in Eq. (7) can then be rewritten as

f_{ϕ} = \frac{ϕ^{'}}{ϕ^{0}} = \frac{{k^{'}}_{r}}{k_{r}^{0}} \frac{1}{1 + ϕ^{0} (F_{p 0} - 1)} = \frac{Γ}{F_{p}} .

Finally, the fluorescence intensity enhancement, FE, can be obtained by

F E = \frac{C R M^{'}}{C R M^{0}} = \frac{η^{'}}{η^{0}} \frac{ϕ^{'}}{ϕ^{0}} \frac{σ {I^{'}}_{e}}{σ I_{e}^{0}} = f_{η} f_{ϕ} f_{I} .

Where

f_{I}

and

f_{η}

are the excitation intensity enhancement and light collection efficiency enhancement, respectively. In our simulation, we assumed that the intrinsic light collection efficiency of the instrument is not changed by the metallic structure, i.e.

η^{'} = η^{0}

and

f_{η} = 1

, the FE can then be simplified as

F E = f_{I} f_{ϕ} .

Notice that the fluorescence signal enhancement consists of two contributions: one is from the enhancement of the local excitation intensity and the other is from the gain in the quantum yield of the fluorophores.

2.2 Simulation model

In the model, we considered a cylindrical Au-ZMW with the metal layer (100 nm in thickness) over-etched into the glass substrate to different depths. The whole device was assumed to be covered by water, which has a refractive index of 1.33. The spacing between two neighboring nanoapertures in the ZMW arrays is 2 μm in both X and Y direction. Three-dimensional finite-difference time-domain (3D-FDTD) method was then employed for the simulation.

To calculate the excitation enhancement of ZMWs with different radius, we assumed that a normally incident plane wave (1V/m) with x polarization illuminates the ZMW structure from the substrate side along z direction. The total calculated volume is 2 μm × 2 μm × 2 μm and the grid size is 2 nm. The Perfectly Matched Layers (PML) were introduced at the + z and –z boundaries to avoid influence from the reflection of electromagnetic waves from them. Considering the symmetry of the ZMW structure, the same boundary conditions were used for both x and y directions to shorten the simulation time. The average enhancement of the local excitation intensity, f_I, was calculated by integrating the total excitation intensity within a defined volume (i.e. a cylindrical disk of 10 nm height with radius equal to the ZMW radius) in the ZMW and then dividing it by the integrated intensity within the same volume in absence of the ZMW (i.e. the free solution case), as shown in Fig. 1.

Fig. 1 The schematics of simulation models for (a) an over-etched ZMW and (b) free solution. Notice that in free solution case, there are no metallic structures and the coverslip also has no undercuts.

Download Full Size | PDF

To calculate the emission enhancement, an electric dipole was placed at different vertical position, z, along the central axis of the ZMW. The PML boundary conditions were applied to all directions. Again, for simplification, the dipole was assumed to be x polarized in view of the symmetry of the nanoaperture as well as the fact that the z polarized dipole makes an insignificant contribution to far-field emission [24]. The modification of quantum yield is attributed to the radiative enhancement and Purcell factor according to Eq. (7). Because every rate is proportional to its corresponding transmission power [31], the radiative enhancement can be calculated as the ratio of the transmission powers that radiate into the far field in the presence ( ${P^{'}}_{r}$ ) and absence ( $P_{r}^{0}$ ) of ZMW, i.e.

Γ = \frac{{k^{'}}_{r}}{k_{r}^{0}} = \frac{{P^{'}}_{r}}{P_{r}^{0}} .

However, considering in the experiment, only the transmission powers that radiate downwards into the substrate can be detected, we thus define

Γ_{e f f} = \frac{{P^{'}}_{r z -}}{P_{r z -}^{0}}

as the effective radiative enhancement. Here

{P^{'}}_{r z -}

and

P_{r z -}^{0}

are the transmission powers that radiate downwards into the substrate.

To calculate the Purcell factor $F_{p}$ , we first calculate $F_{p 0}$ using the following equation,

F_{p 0} = \frac{{k^{'}}_{r} + {k^{'}}_{a b s}}{k_{r}^{0}} = \frac{{P^{'}}_{r} + {P^{'}}_{a b s}}{P_{r}^{0}},

where

{P^{'}}_{a b s}

is the power loss due to the absorption in the metallic structure.

{P^{'}}_{r} + {P^{'}}_{a b s}

is the total radiated power from the emitter in the ZMW. Then, for fluorescent molecules with a known native quantum yield

ϕ^{0}

,

F_{p}

can be obtained using Eq. (11). We further define

f_{ϕ e f f} = \frac{Γ_{e f f}}{F_{P}}

as the effective quantum yield enhancement. Finally, with the obtained

f_{I}

and

f_{ϕ e f f}

, we can get fluorescence signal enhancement, FE as

F E = f_{I} f_{ϕ e f f} .

2.3 Simulation results

Figure 2(a) and 2(b) show the vertical cross-section view of the calculated excitation intensity (in log scale) and its excitation intensity along the central z-axis in an Au-ZMW with the glass substrate over-etched by 80 nm under 532 nm and 637 nm excitations, respectively. Here we define z = 0 nm as the position where the interface between the glass substrate and metal film is. The undercut region (z<0 nm) and the evanescent region (z≥0 nm) can be clearly distinguished from their excitation intensities. It can be seen that the excitation fields are mainly confined within the ZMW for both excitation conditions. It is also worth mentioning that, in our simulations, for ZMWs with and without undercut, the excitation intensities in the evanescent region are almost identical, suggesting that the undercut doesn’t perturb the field confinement capability of ZMWs in the evanescent region. In addition, since in many applications such as single molecule sequencing, the molecules of interest are often anchored to the bottom of nanoapertures, we thus choose to compare the excitation intensity distribution along x direction at the position of 10 nm above the bottom of nanoapertures (i.e. at z = 10 nm for ZMW without undercut and at z = −70 nm for ZMW with 80 nm undercut), as shown in Fig. 2(c) and 2(d). In contrast to ZMWs without undercut, the radial uniformity of the excitation intensities in the over-etched ZMWs is greatly improved for both 532 nm and 637 nm excitation conditions, which could be even more clearly visualized in the transverse cross-section view of the excitation intensity distribution as shown in the insets. Besides that, the excitation intensity in the center region of nanoapertures with undercut (z = −70 nm) is also slightly enhanced in comparison with that of without undercut (z = 10 nm), as indicated by the blue dash line in Fig. 2(a) and 2(b). Therefore, for this case, over-etching not only enhances the excitation intensity in the center of nanoapertures, but also significantly improves its transverse homogeneity, making the excitation intensity almost independent of x and y-position in the nanoaperture.

Fig. 2 The excitation intensity distribution in an Au-ZMW (r = 40 nm, undercut 80 nm) (a) A vertical cross-section view of the excitation intensity (log scale) in an over-etched ZMW under 532 nm excitation and its excitation intensity along the central z-axis. (b) A vertical cross-section view of the excitation intensity (log scale) of an over-etched ZMW at 637 nm excitation and its excitation intensity along the central z-axis. (c) The normalized excitation intensity distribution along x for ZMWs with (green solid line) and without (black solid line) undercut at a vertical position of 10 nm above the bottom of ZMWs under 532 nm excitation. The insets show the transverse cross-section view of the excitation intensities with and without undercut at 532 nm excitation. (d) The normalized excitation intensity distribution along x for ZMWs with (red solid line) and without (black solid line) undercut at a vertical position of 10 nm above the bottom of ZMWs under 637 nm excitation. The insets show the transverse cross-section view of the excitation intensities with and without undercut at 637 nm excitation.

Download Full Size | PDF

Figures 3 and 4 show the simulated results of the excitation intensity, effective quantum yield and fluorescence enhancement for Cy3 and Cy5 fluorophores, respectively, in Au-ZMWs with 230 nm undercut and various radial sizes. Here, again, z = 0 nm indicates the interface position of the glass substrate and metal film. In the simulation, we assumed that the dye molecule has a negligible size and can be taken as an electric dipole which emits at 568 nm when excited at 532 nm for Cy3 or emits at 670 nm when excited at 637 nm for Cy5.

Fig. 3 Calculated enhancement factors for Cy3-ssDNA (Φ₀ = 0.15 [35,36]) versus z-position in an Au-ZMW with 230 nm undercut and various radial size. (a) Excitation intensity enhancement. (b) Effective quantum yield enhancement and (c) Fluorescence enhancement.

Download Full Size | PDF

Fig. 4 Calculated enhancement factors for Cy5-ssDNA (Φ₀ = 0.27 [37]) versus z-position in an Au-ZMW with 230 nm undercut and various radial size. (a) Excitation intensity enhancement. (b) Effective quantum yield enhancement and (c) Fluorescence enhancement.

Download Full Size | PDF

The excitation enhancement for Cy3 at different vertical positions within an Au-ZMW with 230 nm under-cut is shown in Fig. 3(a). The overall profile of the excitation intensity is due to the formation of a standing wave at the glass side as the incident excitation light and the light reflected by the metal layer are superposed with each other [23]. The small enhancement peak near the gold-glass interface (i.e. z = 0 nm) comes from the localized surface plasmon resonance (LSPR) effect, which decreases as the nanoaperture’s radius increases. The maximal peak of the excitation enhancement also reduces and its position moves toward the metal side when the aperture size increases, a result related to the propagating mode cutoff of ZMWs [32], i.e. for the aperture with larger radius, the excitation light will become less confined and penetrate more into the nanoaperture, thereby making the standing wave peak finally shift towards the metal side. Figure 3(b) displays the effective quantum yield enhancement of Cy3 as a function of vertical position in the Au-ZMW. Similar to the excitation field, the superposition of the radiative field of the fluorescent molecule with its reflection from the metal layer also forms a standing wave. The interaction of the fluorescent molecule with this superposed field will modify its radiative rates, which eventually leads to the peak at z~80 nm in the effective quantum yield enhancement. As the radius of the Au-ZMW becomes smaller, the larger enhancement in the effective quantum yield is achieved in most of the undercut regions. It is noticed that for ZMWs with r = 40 nm, near the glass-gold interface, the effective quantum yield decreases slightly. Our simulation shows that both the effective radiative rate enhancement and Purcell factor in Eq. (7) increase near the interface owing to the stronger LSPR. However, the change of Purcell factor is slightly larger than that of the effective radiative rate enhancement, which finally leads to a small drop in the enhancement of the effective quantum yield near the glass-gold interface. The fluorescence enhancement, FE, is calculated as the product of the excitation enhancement and the effective quantum yield enhancement, as shown in Fig. 3(c). The ZMW with a smaller diameter turns out to have a higher FE and the optimal over-etching depth is about 75 nm for an Au-ZMW with r<100 nm. As the radius of the ZMW increases, the FE peak slightly shifts towards the metal side.

Similar simulation was also carried out for Cy5 with the excitation at 637 nm as shown in Fig. 4. In contrast to the results for Cy3, the enhancement profile here exhibits a wider distribution and the peak on the “shoulder” at z = 0 nm is more striking owing to the stronger LSPR effect for gold at this wavelength. The resulting fluorescence enhancement FE is thus higher for Cy5 than for Cy3, as shown in Fig. 4(c). The optimal over-etching depth is about 90 nm in this case. It is worth noticing that the peak (which is mainly due to LSPR) at z = 0 nm has similar magnitudes to the peak at about z = −90 nm (which is mainly due to the undercut and standing wave formation) for ZMWs with r≥60 nm. For these cases, the advantage of an over-etched ZMW over that without undercut might not be so evident. However, considering the improved homogeneity shown in Fig. 2, ZMWs with undercut shall still be more preferred. Moreover, for ZMWs with even smaller radius such as r = 40 nm, the loss due to the absorption of the excitation light by the metal wall increases significantly near the glass-gold interface, leading to a drop in local excitation intensity, as shown in Fig. 4(a). Quenching due to the metal wall will also become more significant for ZMWs with r = 40 nm, making the effective quantum yield enhancement drop sharply near the interface, see Fig. 4(b). The peak of FE at z = 0 nm thus greatly diminishes as shown in Fig. 4(c). This result is actually consistent with previous observations [33,34]. It has been shown that the fluorescence enhancement in a conventional Au-ZWM without undercut dropped quickly when the radius became smaller than 60 nm under 633 nm excitations. The quenching thus set a limitation on further improving the performance of a conventional ZMW without undercut simply by reducing its radial size. By contrast, the peak of FE at z = −90 nm in our simulation still continues to increase as the radius of the ZMW decreases even to 40 nm. The reason is due to the fact that the boundaries in the over-etched region are no more metallic and the fluorescence quenching is thus greatly relieved. Meanwhile, as the aperture size reduces, more light is reflected and thus the excitation intensity at z = −90 nm also increases. These results thus demonstrate that the over-etched ZMW is clearly more advantageous than that without undercut.

3. The fabrication of over-etched ZMWs

3.1 Fabrication

Fabrication of the over-etched ZMW arrays was carried out using 150 ± 20 μm thickness borosilicate coverslip as the underlying substrate. Before the fabrication, the coverslip was sonicated sequentially in a series of solutions of detergent, acetone and ethanol. Later it was cleaned in piranha solution (H₂SO₄:H₂O₂ = 3:1) and rinsed with copious amount of deionized water. After that, it was dried using N₂ gas flow and then the oven. Then 2 nm thick Ti as an adhesion layer and 100 nm thick Au films were deposited sequentially on the cleaned coverslip using a high vacuum electron beam evaporation system (DE400, Wavetest). Next, a dual source focused ion beam (FIB) instrument (Auriga Ga⁺, Zeiss) which is also integrated with a scanning electron microscope (SEM) was used for milling and characterization of ZMW arrays. Finally, ZMW arrays with different over-etching depths were obtained by carefully tuning the parameters such as the milling dose and dwell time. The detailed milling parameters for different over-etching depths are summarized in Table 1. To determine the over-etching depth, after the ZMW array was made, we used FIB tool to cut one row of ZMWs to show the vertical section of the nanoapertures. The over-etching depth thus can be directly measured in the SEM mode which has already automatically corrected the angle effect due to the tilt of the sample. One example is given in Fig. 5. The contrast between the gold and coverslip is sharp, making the boundary between them easy to recognize. The bottom of the nanoaperture is less visible but still can be recognized. The depth of the undercut can then be determined by measuring the distance between these two boundaries.

Table 1. FIB milling parameters used for ZMW array fabrication

View Table | View all tables in this article

Fig. 5 The SEM image of over-etched ZMWs.

Download Full Size | PDF

3.2 Surface functionalization

Before adding the sample, the clean ZMW arrays were incubated with 1 mg/ml of BSA for 30 min at room temperature and rinsed thoroughly with 10 mM Tris-HCl (pH 8.0) buffer. Next, 1% Tween 20 was added and incubated for 1 h. The purpose of adding BSA and Tween 20 is to prevent the non-specific binding of the fluorescent labelled DNA molecules to the surface of ZMWs. Finally, the ZMW arrays were rinsed again thoroughly with 10 mM Tris-HCl (pH 8.0) buffer before being used for experiments.

3.3 Sample preparation

Two single stranded DNA samples were used in the experiments. One has the sequence of 5′-TCGCCG-(A)₃₁-CGGCGATTTTCTTCACAAACCAGTCCAAACTATCACAAACTTA-3′ and is labelled with Cy5 at its 5′ end. The other has the sequence of 5′-CGTGCGCTCTCCTG TTCCGACCTATAGTGAGTCGTATTA-3′ and is labelled with Cy3 at its 5′ end. These DNA templates were chemically synthesized and then purified by HPLC (Shanghai Sangon Biotech Co., Ltd., China). Before the experiment, stocks were diluted with a 10 mM Tris-HCl (pH 8.0), 50 mM NaCl buffer into various concentrations as needed.

4. Experimental results

4.1 Experimental setup

Fluorescence correlation spectroscopy (FCS) is a common way to characterize the effective excitation volume and fluorescence intensity enhancement of plasmonic nano-structures [2,11,38]. A schematic of our FCS experimental setup is shown in Fig. 6 which includes a home-built confocal FCS system and a commercial inverted microscope (Ti-E 2000, Nikon Instruments). The excitation module in the FCS system consists of a green solid-state laser (532 nm, MW-SL-532, Changchun Laser Optoelectronics Technology Co., Ltd., China) and a red semiconductor laser (637 nm, MW-SGX-635, Changchun Laser Optoelectronics Technology Co., Ltd., China). In the experiments the laser powers were set at 200 μW for 532 nm laser and 20 μW for 637 nm laser, respectively, to ensure that the fluorescence signal is strong enough for detection and at the same time the excitation is not saturated. These two laser beams were combined by using a dichroic mirror (ZT532rdc, Chroma) and then expanded by a telescope to ~15 mm in diameter in order to fully use the large numeric aperture of the objective. A 100x oil-immersion objective (PlanApo TIRF/1.49 NA, Nikon) is used to tightly focus the two laser beams into a spot. The precise alignment of the ZMW with the laser spot was obtained by using a 3D nano-piezo stage (NS152028, Sanying Motioncontrol Instruments Ltd., China). The emissions from the fluorophores (Cy3 or Cy5) were spectrally filtered using a 532/640 dual-band cube (TRF59907, Chroma) and then focused by the tube lens of microscope onto an optical fiber (M42L02, NA 0.22, Thorlabs) which also serves as a confocal pinhole to reject out-of-focus light due to its small inner core diameter (~50 μm). The fluorescence light from the fiber output end is then collimated and later split into two channels by a dichroic mirror (ZT561rdc, Chroma). The fluorescence lights in these two channels were further filtered by two emission filters (T550lpxr, Chroma; T647lpxr, Chroma) and finally focused onto two APDs (SPCM-AQRH-15, Excelitas), respectively. The signals from APDs were then acquired using a 32-bit counter-board (PCIe-6612, National Instruments) and the recorded data were processed to calculate FCS using a home-written LabVIEW program. The typical duration for each measured trace was 20 s and a binning time of 1 μs was used. All experiments were performed at room temperature (23 ± 1 °C).

Fig. 6 Schematic of the experimental setup for FCS measurement with an over-etched ZMW. DM: dichroic mirror; ND: neutral density filter; BPF: band-pass filter; APD: avalanche photodiode.

Download Full Size | PDF

4.2 Results and discussion

Before using Cy3 and Cy5 labelled DNA to characterize the fluorescence enhancement per molecule and effective excitation volume of ZMWs, the CRMs of these two DNA molecules in free solution have to be determined in advance from FCS experiments, which further requires us to calibrate the geometric parameters (such as the beam waist and axial half-height) of the optical field in the laser focus in free solution.

Our laser beam can be taken as a Gaussian beam around the focus region. To determine the radius at the beam waist, r₀, as well as the axial half-height z₀, we used Rhodamine 6G (λ_exc = 532 nm) as the calibration dye, which has a known diffusion coefficient D = 4.14 × 10⁻⁶ cm²/s in water [39]. The concentration C, diffusion time τ_d and effective excitation volume V thus can be obtained from a nonlinear fitting of the measured FCS data with Eq. (20) using the Levenberg-Marquardt algorithm,

G (τ) = 1 + \frac{1}{C N_{A} (τ + τ_{d})} \frac{1}{\sqrt{{(\frac{V}{τ_{d}})}^{2} + {(4 π D)}^{3} τ}},

where τ is the correlation time, N_A is the Avogadro constant. With the measured τ_d (~25.45 μs) and V (~0.205 fL), finally, we got

r_{0} = \sqrt{4 D τ_{d}}

= 205 nm and

z_{0} = V / (π^{3 / 2} r_{0}^{2})

= 882 nm, which are comparable to the radius of a diffraction-limited spot (~218 nm) and depth of focus (~728 nm) estimated from theory [40]. The difference between the concentrations from fitting and expectation is within 2.7%, further validating our measurements.

With the calibrated r₀ and z₀, the autocorrelation function can be rewritten as

G (τ) = 1 + \frac{1}{N} \frac{1}{(1 + τ / τ_{d}) \sqrt{1 + τ / (s^{2} τ_{d})}},

where N is the average number of fluorescent molecules in the excitation volume and s = z₀ /r₀ = 4.3.

Equation (21) can be further improved by taking into account the background noise,

G (τ) = 1 + \frac{1}{N} {(1 - \frac{〈 B 〉}{〈 F 〉})}^{2} \frac{1}{(1 + τ / τ_{d}) \sqrt{1 + τ / (s^{2} τ_{d})}},

where <F> is the total fluorescence signal, <B> is the background noise.

Note that the blinking of Cy5 is significant due to its triplet states, which also need to be taken into account in the autocorrelation function calculation. So for Cy5, Eq. (22) is finally modified to be [41]

G (τ) = 1 + \frac{1}{N} {(1 - \frac{〈 B 〉}{〈 F 〉})}^{2} \frac{1}{(1 + τ / τ_{d}) \sqrt{1 + τ / (s^{2} τ_{d})}} (1 + \frac{T}{1 - T} \exp (- τ / τ_{t r i p l e t})),

where τ_triplet is the characteristic correlation time associated with the triplet states and T is the fraction of the triplets.

Fitting the measured FCS data for Cy3-ssDNA (39nt) and Cy5-ssDNA (80nt) with Eqs. (22) and (23), respectively, we can finally get the CRMs from Eq. (24)

C R M = (〈 F 〉 - 〈 B 〉) / N,

where N is the average number of molecules in the excitation volume obtained from fitting. The results of the measured CRMs are shown in Tables 2 and 3 (see those numbers below “Coverslip”). The diffusion coefficients of these two kinds of molecules can also be determined from fitting to be 1.23 × 10⁻⁶ cm²/s and 1.06 × 10⁻⁶ cm²/s in 10 mM Tris-HCl (pH 8.0), 50 mM NaCl buffer.

Table 2. The measured CRMs for Cy3-ssDNA

View Table | View all tables in this article

Table 3. The measured CRMs for Cy5-ssDNA

View Table | View all tables in this article

Next, to get the CRMs in ZMWs with different undercuts, 100 nM Cy3-ssDNA or 50 nM Cy5-ssDNA was added into the ZMW chip. The measured results of CRM (background noise corrected) via FCS experiments are also shown in Tables 2 and 3. With these data, we then investigated how the over-etching depth affects the fluorescence enhancement.

Firstly, we used Cy3-ssDNA as the probing molecule. As shown in Fig. 7(a), the measured autocorrelation curves for Cy3-ssDNA are normalized for easy comparison. When the undercut is less than 50 nm, the measured FCS curves and diffusion time τ_d in ZMWs are close to that in free solution. It is worth mentioning that the excitation volume in ZMWs is generally much smaller than in a focused laser spot in free solution and hence should exhibit a much shorter diffusing time. However, because of the confining effect of ZMWs, i.e. molecules in a ZMW can be reflected back to the excitation volume after colliding with the wall, the time for molecules to diffuse away from the excitation volume in ZMWs is greatly elongated which finally results in an apparently large τ_d. As the over-etching depth increases, the autocorrelation curve also exhibits a larger diffusion time as a result of the enlarged nanoaperture size and excitation volume. To compare the experiment with the simulation, both the measured and simulated FEs are shown in Fig. 7(b) for ZMWs with different over-etching depths. It is worth noting that in the measurement fluorophores which diffuse through the nanoaperture will experience drastically different conditions based on proximity to the metal structure. Those in the metal region will emit at one level and when it diffuses to the over-etched region would emit at another. Therefore, the measured FEs (i.e. CRM enhancements) are actually an average over the whole volume with heterogeneous excitations. To make comparison fair, the calculated FEs are also an average over the whole volume. As can be seen, only reasonable agreement is achieved here, i.e. both the measured and simulated FEs exhibit an overall increasing trend with increasing the over-etching depth. The measured FEs, however, have slightly larger values than the simulated ones, and also do not have an obvious peak occurring at ~120 nm undercut condition. These differences possibly come from the fact that in the experiments a focused excitation light with a confocal detection scheme was adopted while in the simulation, we assumed a simple plane wave for the incident light.

Fig. 7 FCS measurements of Cy3-ssDNA in Au-ZMWs (r = 80 nm) with different over-etching depths at 532 nm excitation. (a) The measured auto-correlation curves (normalized) of Cy3-ssDNA in free solution and ZMWs with various undercuts. (b) The measured (green solid line) and simulated (blue dash line) FEs in ZMWs with different over-etching depths. The error bars in the experimental data are the standard errors of the mean (SE).

Download Full Size | PDF

We also measured FCS curves and FEs using Cy5-ssDNA molecules and the results are shown in Fig. 8. Compared to the results presented in Fig. 7(a), here, the auto-correlation curves in the ZMWs with the undercut less than 50 nm exhibit a shorter diffusion time compared to that in free solution. The measured FEs for ZMWs with different over-etching depths again exhibit only reasonable agreement with the simulation, with the measured FEs slightly larger than those from simulation. In addition, although the simulation predicts that the maximal FE occurs at ~150 nm undercut condition, the largest FE in the experiment was found to occur at 100 nm undercut condition. Similar as in Fig. 7, we attribute such observed discrepancies to the different illumination and detection schemes used in the simulation and experiments.

Fig. 8 FCS measurements of Cy5-ssDNA in Au-ZMWs (r = 80 nm) with different over-etching depths at 637 nm excitation. (a) The measured auto-correlation curves (normalized) of Cy5-ssDNA in free solution and ZMWs with various undercuts. (b) The measured (red solid line) and simulated (blue dash line) FEs in ZMWs with different over-etching depths. The error bars in the experimental data are the standard errors of the mean (SE).

Download Full Size | PDF

The effective excitation volumes (V_eff) of Au-ZMWs can be obtained from FCS measurements using Eq. (25) [2,16,38].

V_{e f f} = \frac{N / N_{A}}{C},

where N is the average number of molecules in the excitation volume measured by FCS. N_A is the Avogadro constant, C is the molar concentration of fluorophores which is known. The measured V_eff as a function of over-etching depth are shown in Fig. 9. As expected, a deeper undercut results in a larger excitation volume. It was noted that the effective excitation volume determined using Cy3-ssDNA at 532 nm excitation is slightly larger than that measured with Cy5-ssDNA at 637 nm excitation. This observation is actually consistent with the prediction from the propagating mode cutoff conditions of ZMWs, i.e. light of the longer wavelength will fulfill the cut-off condition more easily and is thus more confined.

Fig. 9 The effective excitation volumes V_eff measured by FCS for Cy3-ssDNA and Cy5-ssDNA in Au-ZMWs (r = 80 nm) with different over-etching depths.

Download Full Size | PDF

In the end, we used FCS to study the dependence of the FE and V_eff on the radius of ZMW with Cy5-ssDNA as the probing molecule. The undercut was chosen to be 100 nm as FE was previously found to be the largest under this condition in the experiment as shown in Fig. 8(b). The final results are shown in Fig. 10(a). Clearly, reducing the radius of ZMW leads to an even larger FE and smaller V_eff. V_eff is also more sensitive to the change of the radius than the over-etching depth. Further, single molecule fluorescence detection at the concentration up to ~1 μM was demonstrated with Cy5-ssDNA molecules using the over-etched ZMWs (r = 65 nm, 100 nm undercut), as shown in Fig. 10(b).

Fig. 10 The measured effective excitation volumes, fluorescence enhancements per molecule and auto-correlation curve in Au-ZMWs with 100 nm undercut. (a) The measured V_eff and FE in ZMWs with different radius and 100 nm undercut. (b) The measured auto-correlation curve of Cy5-ssDNA and its fitting in an Au-ZMW with r = 65 nm and 100 nm undercut. The concentration of Cy5-ssDNA used in the experiment is 1 μM. The fitting yielded that the average number of fluorescent molecules (N) detected in the excitation volume is ~1.1.

Download Full Size | PDF

5. Conclusion

We have studied the influence of over-etching on the fluorescence enhancement in Au-ZMWs at different excitation wavelengths (i.e. 532 and 637 nm) both theoretically and experimentally in details. Our simulation demonstrates that over-etching can greatly improve the homogeneity of the excitation field in an Au-ZMW and also enhance the excitation intensities for both wavelengths. The quantum yield can also be improved with the undercut and eventually the fluorescence signal can be essentially enhanced. The maximum fluorescence enhancement increases as reducing the radius of the ZMW and its value can be ~14 for Cy3 and ~24 for Cy5 for an over-etched Au-ZMW with r = 40 nm.

We further fabricated Au-ZMW arrays using a FIB instrument and characterized the fluorescence enhancement per molecule and the effective excitation volume for ZMWs with different radius and over-etching depth via FCS measurements. Our experimental results indicate that, in average, the fluorescence in a 100 nm over-etched ZMW with r = 80 nm can be 3 times enhanced for a diffusing Cy3 molecule and 7 times enhanced for a diffusing Cy5 molecule. The simulation, however, only showed reasonable agreement with our experimental observations.

Finally, although it has been shown that over-etching can enhance the fluorescence signal both in our simulation and experiments, the undercut will also lead to an enlarged excitation volume and thus lower the upper-limit of the concentration that can be used for single molecule fluorescence detection. Fortunately, this can be solved by reducing the radius of ZMWs, as shown in our measurements. Moreover, as the radius of ZMWs decreases, the fluorescence signal can also be further enhanced even when r is very small. This is in sharp contrast with a conventional ZMW without undercut, in which reducing radius eventually would cause the molecules too close to the metal wall and hence result in strong quenching in the fluorescence. Our results therefore indicate that combining over-etching and reducing radius of the ZMW can be a simple and efficient way to improve the performance of an Au-ZMW in single molecule fluorescence detection.

Funding

National Natural Science Foundation of China (NSFC-11674403), the Science and Technology Planning Project of Guangdong Province, China (2018A050506034), the Fundamental Research Funds for the Central Universities (18lgzd16) and the Open Fund of the State Key Laboratory of Optoelectronic Materials and Technologies, Sun Yat-sen University.

Acknowledgments

We thank members of the J.M. laboratory for helpful discussions. We also wish to thank the engineers of State Key Laboratory of Optoelectronic Materials and Technologies at Sun Yat-sen University for the technical support.

References

1. J. M. Moran-Mirabal and H. G. Craighead, “Zero-mode waveguides: sub-wavelength nanostructures for single molecule studies at high concentrations,” Methods 46(1), 11–17 (2008). [CrossRef] [PubMed]

2. M. J. Levene, J. Korlach, S. W. Turner, M. Foquet, H. G. Craighead, and W. W. Webb, “Zero-mode waveguides for single-molecule analysis at high concentrations,” Science 299(5607), 682–686 (2003). [CrossRef] [PubMed]

3. J. Eid, A. Fehr, J. Gray, K. Luong, J. Lyle, G. Otto, P. Peluso, D. Rank, P. Baybayan, B. Bettman, A. Bibillo, K. Bjornson, B. Chaudhuri, F. Christians, R. Cicero, S. Clark, R. Dalal, A. Dewinter, J. Dixon, M. Foquet, A. Gaertner, P. Hardenbol, C. Heiner, K. Hester, D. Holden, G. Kearns, X. Kong, R. Kuse, Y. Lacroix, S. Lin, P. Lundquist, C. Ma, P. Marks, M. Maxham, D. Murphy, I. Park, T. Pham, M. Phillips, J. Roy, R. Sebra, G. Shen, J. Sorenson, A. Tomaney, K. Travers, M. Trulson, J. Vieceli, J. Wegener, D. Wu, A. Yang, D. Zaccarin, P. Zhao, F. Zhong, J. Korlach, and S. Turner, “Real-time DNA sequencing from single polymerase molecules,” Science 323(5910), 133–138 (2009). [CrossRef] [PubMed]

4. T. Miyake, T. Tanii, H. Sonobe, R. Akahori, N. Shimamoto, T. Ueno, T. Funatsu, and I. Ohdomari, “Real-time imaging of single-molecule fluorescence with a zero-mode waveguide for the analysis of protein-protein interaction,” Anal. Chem. 80(15), 6018–6022 (2008). [CrossRef] [PubMed]

5. L. P. Zaino 3rd, D. A. Grismer, D. Han, G. M. Crouch, and P. W. Bohn, “Single occupancy spectroelectrochemistry of freely diffusing flavin mononucleotide in zero-dimensional nanophotonic structures,” Faraday Discuss. 184, 101–115 (2015). [CrossRef] [PubMed]

6. A. Tsai, J. D. Puglisi, and S. Uemura, “Probing the translation dynamics of ribosomes using zero-mode waveguides,” Prog. Mol. Biol. Transl. Sci. 139, 1–43 (2016). [CrossRef] [PubMed]

7. K. Fujimoto, Y. Morita, R. Iino, M. Tomishige, H. Shintaku, H. Kotera, and R. Yokokawa, “Simultaneous observation of kinesin-driven microtubule motility and binding of adenosine triphosphate using linear zero-mode waveguides,” ACS Nano 12(12), 11975–11985 (2018). [CrossRef] [PubMed]

8. S. F. Heucke, F. Baumann, G. P. Acuna, P. M. Severin, S. W. Stahl, M. Strackharn, I. H. Stein, P. Altpeter, P. Tinnefeld, and H. E. Gaub, “Placing individual molecules in the center of nanoapertures,” Nano Lett. 14(2), 391–395 (2014). [CrossRef] [PubMed]

9. E. Pibiri, P. Holzmeister, B. Lalkens, G. P. Acuna, and P. Tinnefeld, “Single-molecule positioning in zeromode waveguides by DNA origami nanoadapters,” Nano Lett. 14(6), 3499–3503 (2014). [CrossRef] [PubMed]

10. P. Holzmeister, E. Pibiri, J. J. Schmied, T. Sen, G. P. Acuna, and P. Tinnefeld, “Quantum yield and excitation rate of single molecules close to metallic nanostructures,” Nat. Commun. 5(1), 5356 (2014). [CrossRef] [PubMed]

11. R. Regmi, A. A. Al Balushi, H. Rigneault, R. Gordon, and J. Wenger, “Nanoscale volume confinement and fluorescence enhancement with double nanohole aperture,” Sci. Rep. 5(1), 15852 (2015). [CrossRef] [PubMed]

12. G. Lu, W. Li, T. Zhang, S. Yue, J. Liu, L. Hou, Z. Li, and Q. Gong, “Plasmonic-enhanced molecular fluorescence within isolated bowtie nano-apertures,” ACS Nano 6(2), 1438–1448 (2012). [CrossRef] [PubMed]

13. D. Punj, M. Mivelle, S. B. Moparthi, T. S. van Zanten, H. Rigneault, N. F. van Hulst, M. F. García-Parajó, and J. Wenger, “A plasmonic ‘antenna-in-box’ platform for enhanced single-molecule analysis at micromolar concentrations,” Nat. Nanotechnol. 8(7), 512–516 (2013). [CrossRef] [PubMed]

14. V. Flauraud, R. Regmi, P. M. Winkler, D. T. Alexander, H. Rigneault, N. F. van Hulst, M. F. García-Parajo, J. Wenger, and J. Brugger, “In-plane plasmonic antenna arrays with surface nanogaps for giant fluorescence enhancement,” Nano Lett. 17(3), 1703–1710 (2017). [CrossRef] [PubMed]

15. C. Zhao, Y. Liu, J. Yang, and J. Zhang, “Single-molecule detection and radiation control in solutions at high concentrations via a heterogeneous optical slot antenna,” Nanoscale 6(15), 9103–9109 (2014). [CrossRef] [PubMed]

16. P. Ponzellini, X. Zambrana-Puyalto, N. Maccaferri, L. Lanzanò, F. De Angelis, and D. Garoli, “Plasmonic zero mode waveguide for highly confined and enhanced fluorescence emission,” Nanoscale 10(36), 17362–17369 (2018). [CrossRef] [PubMed]

17. L. Langguth, D. Punj, J. Wenger, and A. F. Koenderink, “Plasmonic band structure controls single-molecule fluorescence,” ACS Nano 7(10), 8840–8848 (2013). [CrossRef] [PubMed]

18. Y. Wang, L. Wu, T. I. Wong, M. Bauch, Q. Zhang, J. Zhang, X. Liu, X. Zhou, P. Bai, J. Dostalek, and B. Liedberg, “Directional fluorescence emission co-enhanced by localized and propagating surface plasmons for biosensing,” Nanoscale 8(15), 8008–8016 (2016). [CrossRef] [PubMed]

19. H. Aouani, O. Mahboub, N. Bonod, E. Devaux, E. Popov, H. Rigneault, T. W. Ebbesen, and J. Wenger, “Bright unidirectional fluorescence emission of molecules in a nanoaperture with plasmonic corrugations,” Nano Lett. 11(2), 637–644 (2011). [CrossRef] [PubMed]

20. H. Aouani, O. Mahboub, E. Devaux, H. Rigneault, T. W. Ebbesen, and J. Wenger, “Plasmonic antennas for directional sorting of fluorescence emission,” Nano Lett. 11(6), 2400–2406 (2011). [CrossRef] [PubMed]

21. H. Aouani, O. Mahboub, E. Devaux, H. Rigneault, T. W. Ebbesen, and J. Wenger, “Large molecular fluorescence enhancement by a nanoaperture with plasmonic corrugations,” Opt. Express 19(14), 13056–13062 (2011). [CrossRef] [PubMed]

22. J. Chen, R. V. Dalal, A. N. Petrov, A. Tsai, S. E. O’Leary, K. Chapin, J. Cheng, M. Ewan, P. L. Hsiung, P. Lundquist, S. W. Turner, D. R. Hsu, and J. D. Puglisi, “High-throughput platform for real-time monitoring of biological processes by multicolor single-molecule fluorescence,” Proc. Natl. Acad. Sci. U.S.A. 111(2), 664–669 (2014). [CrossRef] [PubMed]

23. T. Tanii, R. Akahori, S. Higano, K. Okubo, H. Yamamoto, T. Ueno, and T. Funatsu, “Improving zero-mode waveguide structure for enhancing signal-to-noise ratio of real-time single-molecule fluorescence imaging: a computational study,” Phys. Rev. E Stat. Nonlin. Soft Matter Phys. 88(1), 012727 (2013). [CrossRef] [PubMed]

24. X. Jiao, E. M. Peterson, J. M. Harris, and S. Blair, “UV fluorescence lifetime modification by aluminum nanoapertures,” ACS Photonics 1(12), 1270–1277 (2014). [CrossRef]

25. W. E. Martin, B. R. Srijanto, C. P. Collier, T. Vosch, and C. I. Richards, “A Comparison of Single-Molecule Emission in Aluminum and Gold Zero-Mode Waveguides,” J. Phys. Chem. A 120(34), 6719–6727 (2016). [CrossRef] [PubMed]

26. J. Wenger, B. Cluzel, J. Dintinger, N. Bonod, A. L. Fehrembach, E. Popov, P. F. Lenne, T. W. Ebbesen, and H. Rigneault, “Radiative and nonradiative photokinetics alteration inside a single metallic nanometric aperture,” J. Phys. Chem. C 111(30), 11469–11474 (2007). [CrossRef]

27. J. Wenger, H. Aouani, D. Gérard, S. Blair, T. W. Ebbesen, and H. Rigneault, “Enhanced fluorescence from metal nanoapertures: physical characterizations and biophotonic applications,” Proc. SPIE 7577(1), 75770J (2010). [CrossRef]

28. A. Kinkhabwala, Z. Yu, S. Fan, Y. Avlasevich, K. Mullen, and W. E. Moerner, “Large single-molecule fluorescence enhancements produced by a bowtie nanoantenna,” Nat. Photonics 3(11), 654–657 (2009). [CrossRef]

29. E. M. Purcell, H. C. Torrey, and R. V. Pound, “Resonance absorption by nuclear magnetic moments in a solid,” Phys. Rev. 69(1-2), 37–38 (1946). [CrossRef]

30. A. E. Krasnok, A. P. Slobozhanyuk, C. R. Simovski, S. A. Tretyakov, A. N. Poddubny, A. E. Miroshnichenko, Y. S. Kivshar, and P. A. Belov, “An antenna model for the Purcell effect,” Sci. Rep. 5(1), 12956 (2015). [CrossRef] [PubMed]

31. L. Novotny and B. Hecht, Principles of nano-optics (Cambridge University, 2012), 2nd ed.

32. E. Popov, M. Nevière, J. Wenger, P. F. Lenne, H. Rigneault, P. Chaumet, N. Bonod, J. Dintinger, and T. Ebbesen, “Field enhancement in single subwavelength apertures,” J. Opt. Soc. Am. A 23(9), 2342–2348 (2006). [CrossRef] [PubMed]

33. D. Gérard, J. Wenger, N. Bonod, E. Popov, H. Rigneault, F. Mahdavi, S. Blair, J. Dintinger, and T. W. Ebbesen, “Nanoaperture-enhanced fluorescence: Towards higher detection rates with plasmonic metals,” Phys. Rev. B Condens. Matter Mater. Phys. 77(4), 045413 (2008). [CrossRef]

34. J. Wenger, D. Gérard, J. Dintinger, O. Mahboub, N. Bonod, E. Popov, T. W. Ebbesen, and H. Rigneault, “Emission and excitation contributions to enhanced single molecule fluorescence by gold nanometric apertures,” Opt. Express 16(5), 3008–3020 (2008). [CrossRef] [PubMed]

35. J. Spiriti, J. K. Binder, M. Levitus, and A. van der Vaart, “Cy3-DNA stacking interactions strongly depend on the identity of the terminal basepair,” Biophys. J. 100(4), 1049–1057 (2011). [CrossRef] [PubMed]

36. T. Ha and P. Tinnefeld, “Photophysics of fluorescent probes for single-molecule biophysics and super-resolution imaging,” Annu. Rev. Phys. Chem. 63(1), 595–617 (2012). [CrossRef] [PubMed]

37. R. B. Mujumdar, L. A. Ernst, S. R. Mujumdar, C. J. Lewis, and A. S. Waggoner, “Cyanine dye labeling reagents: sulfoindocyanine succinimidyl esters,” Bioconjug. Chem. 4(2), 105–111 (1993). [CrossRef] [PubMed]

38. H. Rigneault, J. Capoulade, J. Dintinger, J. Wenger, N. Bonod, E. Popov, T. W. Ebbesen, and P. F. Lenne, “Enhancement of single-molecule fluorescence detection in subwavelength apertures,” Phys. Rev. Lett. 95(11), 117401 (2005). [CrossRef] [PubMed]

39. C. B. Müller, A. Loman, V. Pacheco, F. Koberling, D. Willbold, W. Richtering, and J. Enderlein, “Precise measurement of diffusion by multi-color dual-focus fluorescence correlation spectroscopy,” Europhys. Lett. 83(4), 46001 (2008). [CrossRef]

40. J. B. Pawley and B. R. Masters, Handbook of biological confocal microscopy (Springer Science+Business Media, LLC, 2006), 3rd ed.

41. J. Widengren, U. Mets, and R. Rigler, “Fluorescence correlation spectroscopy of triplet states in solution: a theoretical and experimental study,” J. Phys. Chem. 99(36), 13368–13379 (1995). [CrossRef]

Cy3-ssDNA	Coverslip	ZMWs
Cy3-ssDNA	Coverslip	undercut 0 nm	undercut 50 nm	undercut 100 nm	undercut 150 nm	undercut 230 nm
CRM (kHz)	34.931	60.046	79.361	105.251	114.402	140.001
SE	0.146	1.616	2.115	2.118	6.184	2.768

Cy5-ssDNA	Coverslip	ZMWs
Cy5-ssDNA	Coverslip	undercut 0 nm	undercut 50 nm	undercut 100 nm	undercut 150 nm	undercut 230 nm
CRM (kHz)	11.634	70.7	85.868	88.923	80.567	76.126
SE	0.068	0.701	1.293	2.639	1.218	1.581

Cy3-ssDNA	Coverslip	ZMWs
Cy3-ssDNA	Coverslip	undercut 0 nm	undercut 50 nm	undercut 100 nm	undercut 150 nm	undercut 230 nm
CRM (kHz)	34.931	60.046	79.361	105.251	114.402	140.001
SE	0.146	1.616	2.115	2.118	6.184	2.768

Cy5-ssDNA	Coverslip	ZMWs
Cy5-ssDNA	Coverslip	undercut 0 nm	undercut 50 nm	undercut 100 nm	undercut 150 nm	undercut 230 nm
CRM (kHz)	11.634	70.7	85.868	88.923	80.567	76.126
SE	0.068	0.701	1.293	2.639	1.218	1.581

Fluorescence enhancement in an over-etched gold zero-mode waveguide

Abstract

1. Introduction

2. Numerical simulation of fluorescence enhancement

2.1 Theory

2.2 Simulation model

2.3 Simulation results

3. The fabrication of over-etched ZMWs

3.1 Fabrication

3.2 Surface functionalization

3.3 Sample preparation

4. Experimental results

4.1 Experimental setup

4.2 Results and discussion

5. Conclusion

Funding

Acknowledgments

References

Cited By

Figures (10)

Tables (3)

Equations (25)

Optics Express

Ion Beam Current	Dose (nC/μm²)	Dwell times (μs)	Over-etching depth (nm)
30kV·20pA	0.05	50	0
	0.07	50	50
	0.2	50	100
	0.5	100	150
	1	50	230

Ion Beam Current	Dose (nC/μm²)	Dwell times (μs)	Over-etching depth (nm)
30kV·20pA	0.05	50	0
	0.07	50	50
	0.2	50	100
	0.5	100	150
	1	50	230