Nonlinear excitation power dependence of surface enhanced fluorescence from a nanostructured Ag film

Kun-Yu Tai; Ti-Li Lin; Hung-Chih Kan

doi:10.1364/OE.21.031293

1. Introduction

Surface enhanced fluorescence (SEF) [1] on metallic nanostructure occurs when an excitation light illuminates fluorescent molecules placed in close proximity of, but not in direct contact with, metal surfaces. One of the mechanisms that contributes to SEF is the localized enhancement of electric field due to plasmonic response of metallic nanostructures to the incident light, which also is essential for other interesting phenomena such as surface enhanced Raman scattering (SERS) [2,3]. Since the rate of molecular excitation increases with local E-field intensity |E|² of the incident light, substrates that can generate high field enhancement seemingly are suitable for both SERS and SEF applications [4–11]. The other mechanism is the modification of the emission rate of excited fluorescent molecules due to modification of photonic model density by the metal surfaces [12–14]. Similar to SERS [15,16], direct multiplication of enhancement factors from individual mechanisms to account for the total enhancement factor has been used for single molecule SEF [17]. This simple approach implies that the fluorescent intensity increases linearly with the intensity of excitation light. However, since excited fluorescent molecules have non-zero lifetime for relaxation, the total decay rate, which includes the fluorescent emission rate, is finite for fluorescence. It is thus clear that at high enough excitation power the finite decay rate of excited fluorescent molecules would eventually interfere with the excitation rate. Therefore, deviation from linear dependence of the fluorescent intensity on the excitation power can be expected in general. Furthermore, since most of the SEF enhancement factors were defined operationally as the ratio of the fluorescent intensity from regions of interest on the sample such as those patterned with nanostructures to a reference fluorescent intensity such as that from unpatterned area, the nonlinear dependence of fluorescence on excitation power leads to an excitation power-dependent SEF enhancement factor. For realistic applications of SEF on a substrate where molecules of interest may be present at both patterned and un-patterned area, the ratio of the signal intensities from these two types of area may be more important than the absolute intensities. Therefore, the power dependence of the SEF contrast on nanostructured sample is also crucial for practical applications. However, to the best of our knowledge there has not been a systematic investigation on the excitation power dependence of the SEF contrast on nanostructured metal surfaces reported in the literature.

In this report, we first present a simple theoretical model based on basic physical properties of SEF which predicts nonlinear dependence of the fluorescent contrast between the patterned area and the flat area of a nanostructured metallic substrate. We then report on experimental characterization of the excitation power dependence of SEF from Ag films patterned with arrays of squares pits, whose size is varied from 75nm to 400nm. We demonstrate SEF on this substrate for a given excitation power. We also carry out finite difference time domain method (FDTD) calculations to simulate the interaction between the incident light and pit-array structures, and compare the results with the experiment. We then vary the excitation power over a range of two decades and measure the fluorescent intensities from the patterned area and that from the flat region of the Ag film. We show that while the fluorescent intensities from both areas increase with excitation power the contrast between the patterned area and the flat area indeed decreases with excitation power. We compare the observed trend with the predictions of our simple model of SEF. We find good agreement between the experiment and the model. This allows us to extract the enhancement factor of the fluorescent emission rate and the ratio of the enhancement factor for the excitation rate of fluorescent molecules to that of the total decay rate of excited fluorescent molecules for individual pit arrays.

2. Model of SEF on patterned substrate

We first present a simple theoretical model for the excitation power dependence of fluorescent emission from molecules coated on metal substrate patterned with nanostructures. As shown in Fig. 1(a) fluorescent molecules are coated on both the flat and patterned areas of the substrate. A spacer layer is assumed to define a constant distance between the metal and the molecules to prevent direct contact. The Jablonski diagram in Fig. 1(b) indicates that fluorescent molecules on flat area in ground state $S_{0}$ could absorb an incident photon at rate $k_{a}$ and transition to state $S_{1}$ . Excited molecules in state $S_{1}$ may emit a photon at rate $k_{e}$ and returnto $S_{0}$ , or relaxation processes can be non-radiative at rate $k_{n}$ , due to interaction between the excited molecule with the environment including collision with surrounding molecules in the solution, or interactions between the excited molecule and the metal substrate such as excitation of surface plasmon on the metal surface and excitation of electron-hole pairs in the metal [13,18]. Excited molecules can also transition to triplet state $T_{1}$ through intersystem crossing process at rate $k_{S T}$ . From $T_{1}$ , the molecule relaxes back to $S_{0}$ at rate $k_{T S}$ over a longer period of time compared to the lifetime at $S_{1}$ . This simple model presumes photo-bleaching is insignificant for the experiment under consideration. Figure 1(c) shows the Jablonski diagram for fluorescence of molecules under the influence of the metallic nanostructures. In our simple model, we assume that the main effects are first that the absorption rate $k_{a}$ is enhanced by a factor $e_{1}$ due to field enhancement of the incident light resulting from the plasmonic response of the metallic nanostructure, and secondly that the fluorescent emission rate $k_{e}$ and the non-radiative relaxation rate $k_{n}$ are enhanced by a factor $e_{2}$ and $e_{n}$ ,respectively, due to a possible modification of the electromagnetic interaction between the excited molecules in $S_{1}$ state and the metal surface due to the presence of nano-structures. With $N_{S 1}$ denoting the population of the molecules in state $S_{1}$ , the fluorescent intensities emitted from molecules coated on flat and patterned areas are $k_{e} \cdot N_{S 1}$ and $e_{2} \cdot k_{e} \cdot N_{S 1}$ , respectively. The intrinsic absorption rate $k_{a}$ is proportional to the incident light intensity, i.e. $k_{a} = σ \cdot I_{e x}$ , where $σ$ is a proportionality constant and $I_{e x}$ is the incident light intensity. Our model makes the following simplifying assumption: firstly, the intersystem crossing from $S_{1}$ to $T_{1}$ is negligible, i.e. $k_{S T} < < (k_{e} + k_{n})$ . Therefore, the transition to triplet states and the following relaxation process are ignored in the model. Secondly, the preferred direction of photon emission of individual excited molecules may depend on their relative positions and orientation in the nanostructure on the metal surface [19]. Here we consider an experiment that measures the ensemble average of the fluorescent emission from large number of molecules collected over a wide range of emission angle with, for example, an objective lens. Therefore, the collection efficiency for the fluorescent signal from the unpatterned surface and that from the patterned surface are considered roughly the same in this model. We also consider the case where the molecules are sufficiently close to the metal such that no waveguide mode of emission light propagation can be supported by the spacer layer, which may leads to highly directional fluorescent emission. Let $I_{F, S}$ and $I_{F, P}$ denote the fluorescent intensities from the flat area and the patterned area, respectively, and $C_{F}$ the fluorescent contrast, defined as $I_{F, P} / I_{F, S}$ , and their dependence on $I_{e x}$ in this model can be expressed with the following equations.

I_{F, S} = \frac{σ \cdot I_{e x} \cdot k_{e}}{σ \cdot I_{e x} + k_{e} + k_{n}}

I_{F, P} = \frac{e_{1} \cdot σ \cdot I_{e x} \cdot e_{2} \cdot k_{e}}{e_{1} \cdot σ \cdot I_{e x} + e_{2} \cdot k_{e} + e_{n} \cdot k_{n}}

C_{F} \equiv \frac{I_{F, P}}{I_{F, S}} = \frac{e_{1} \cdot e_{2} \cdot σ \cdot I_{e x} + e_{1} \cdot e_{2} \cdot (k_{e} + k_{n})}{e_{1} \cdot σ \cdot I_{e x} + (e_{2} \cdot k_{e} + e_{n} \cdot k_{n})}

Equations (1) and (2) predict that

I_{F, S}

and

I_{F, P}

increase with

I_{e x}

monotonically, and saturate as

I_{e x} \to \infty

. This trend is shown in Figs. 1(d)-(f). To see the general trend of

C_{F}

, we rewrite Eq. (3) below

C_{F} = e_{2} \cdot \frac{σ \cdot I_{e x} + ϕ_{S}}{σ \cdot I_{e x} + ϕ_{P} / e_{1}}

,where

ϕ_{S} = k_{e} + k_{n}

and

ϕ_{P} = e_{2} \cdot k_{e} + e_{n} \cdot k_{n}

are the total relaxation rates of the excited molecules on the flat metal surface and the patterned metal surface, respectively. Note that

C_{F} \to e_{2}

as

I_{e x} \to \infty

, and

C_{F} \to e_{1} \cdot e_{2} \cdot ϕ_{S} / ϕ_{P}

as

I_{e x} \to 0

. An important question, which is related to the relative contributions of the enhancement of the absorption e₁ and the emission e₂, is how does C_F vary with incident intensity? By defining

η \equiv ϕ_{P} / ϕ_{S}

as the enhancement factor of the total decay rate of excited molecules due to the presence of nanostructure, we show in Fig. 1(d) the case of

e_{1} > η

. In this case the excitation rate enhancement factor e₁ due to local field enhancement is larger than the enhancement factor of the total relaxation rate on the patterned area, and

C_{F}

decreases monotonically with

I_{e x}

. Figure 1(e) shows the opposite case, i.e.

e_{1} < η

. The fluorescent and

C_{F}

increases monotonically with

I_{e x}

. When

e_{1} = η

, this leads to

C_{F} (I_{e x}) = e_{2}

, i.e. the fluorescent contrast becomes independent of

I_{e x}

. Our simple model predicts first that the enhancement in fluorescent emission rate produces a overall scaling factor of the fluorescent contrast

C_{F}

, and secondly the relative strength between the excitation rate enhancement

e_{1}

and the overall decay rate enhancement factor

η

determines the trend of the excitation power dependence of

C_{F}

.

Fig. 1 Schematics of a simple model of surface enhanced fluorescence (SEF) of molecules coated on nanostructured metal substrate. (a) Cross-sectional view of the metal substrate with area patterned with nanostructures (corrugation on the right). The dotted line represents fluorescent molecules coated on top of the substrate with a spacer layer (not shown) in between. (b) and (c) are the Jablonski diagrams for fluorescence of molecules on the flat area and patterned area, respectively. (d)-(f): the fluorescent intensity emitted from molecules on the patterned area I_F,P (thin solid curve) and that from the flat area I_F,S (dashed curve), and the fluorescent contrast C_F (thick solid curve) plotted as a function of the excitation light intensity I_ex, for the case of (d) enhancement of excitation rate larger than enhancement of total decay rate, (e) the opposite of (d), and (f) the case of constant C_F.

Download Full Size | PDF

Below we present an experimental characterization of the excitation power dependence of the surface enhanced fluorescence from an Ag film deposited on the PMMA film which had been patterned with square arrays of nanometer size square pits. We find that the fluorescent intensities measured from Cy3 molecules deposited on both the flat area and the patterned area increase monotonically with the power of the excitation laser. However, the fluorescent contrast between the two areas decreases with excitation power. By fitting the excitation power dependence of the contrast with a function equivalent to Eq. (3), we find that the trend shown in Fig. 1(d) agrees qualitatively with the experiment. Further analysis of the experimental results according to our theoretical model allows us to extract e₂ and the ratio of emission enhancement factor e₁ to the ratio of the relaxation rates with and without the pattern η for most of the pit arrays on the sample.

3. Experimental methods

The nanostructured Ag film sample was fabricated with electron beam lithography (EBL). Figure 2(a) shows the schematics of the sample design. We first spin-coated a layer of Poly(methyl methacrylate) (PMMA) film with thickness of ~180 nm on top of a indium-tin oxide (ITO) coated glass substrate, and patterned arrays of square pits on the PMMA film with EBL. The size of the pits ranges from 75nm to 400nm. A 70nm thick Ag film is then deposited on top of the sample with thermal evaporation in vacuum. Field-emission scanning electron microscope (FESEM) images show that the Ag film on top of the PMMA film is not connected to the Ag film in the pit. Therefore, the Ag nanostructure consists of an array of square holes with square Ag pads inside the holes. Figure 2(b) shows the atomic force microscope (AFM) image of the 400nm size pit array and the line profile scanned across the pit center. The height difference between the Ag film at the surface and the Ag pads is 180 nm. Figures 2(c)-(d) show a selection of SEM images of the pit arrays on the Ag film. Thelarger pits are square. As the size decreases significant rounding of the pit occurs due to the proximity effect in EBL.

Fig. 2 (a) Schematic of the sample design in a cross-sectional view. (b) AFM image of Ag film structure scanned in the area of 400nm size pit array (upper part) and a height profile across the pit along the dashed line indicated above (lower part). (c)-(f) SEM images of the square pit array structures on the Ag film. The nominal ratio of the pit size to center-to-center spacing between neighboring pits is indicated in each panel.

Download Full Size | PDF

For the fluorescence measurement, the Ag film was first coated with a spacer layer which consists of BSA-biotin molecules. Then a layer of cyanine dye molecule Cy3 tagged Streptavidin biomolecules was deposited onto the spacer layer, which bond covalently to the BSA-biotin molecules [4]. In this arrangement the Cy3 tags are about 8 nm away from the Ag surfaces on average. For fluorescence imaging, the sample was maintained moisturized with a phosphate solution (PH~7.9) under a glass cap slide. The fluorescent imaging of the sample was carried out with a near-field scanning microscope system (Nanonics model Multiview 2000) operated in scanning optical imaging mode. Figure 3 shows the schematics of the measurement setup. The incident laser beam of 532 nm wavelength light from a CW laser was directed into an upright optical microscope module through one of the eye piece port with optical fiber and focused on the sample with a 10x objective lens (0.25 NA). The fluorescent light emitted from the Cy3 molecules on the Ag film was collected by the same objective lens and directed toward a photo-multiplier tube (PMT). In front of the PMT a 532 nm notch filter (NF) was used to block the excitation light from reaching the detector and another 570nm ± 10 nm bandpass filter (BF) was used to filter out light whose wavelength is not relevant to thefluorescence of Cy3 (emission maximum at 565nm). We rastered the sample during the imaging process. Prior to coupling to the optical fiber, the intensity of the incident beam emitted from the 532nm wavelength laser was adjusted with a polarizer and a half-wavelength plate. The fiber does not preserve the polarization of the incident light; therefore the incident light is non-polarized. The total power of the incident light at the sample position was measured with a model NOVA II power meter (OPHIR OPTRONICS). Since the spot size of laser beam focused by a 10x objective is fixed, we use the total power measured at the sample position as an indication of the excitation intensity for the results shown below.

Fig. 3 Schematic of fluorescent measurement setup: Solid arrows indicate the optical path for the incident light and dashed arrows indicate the optical path for collection of the fluorescent light.

Download Full Size | PDF

4. Results and discussion

4.1 Enhancement of fluorescence from patterned Ag film

Figure 4(a) shows a fluorescence image of the arrays of nano-pits on the sample. The fluorescent intensity appears brightest at the 175 nm size pit array. This is shown quantitatively by the intensity line profiles in Fig. 3(b). The slightly higher intensities seen in the flat area in between the bright square patterned area may result from imperfect focusing of the excitation beam. We determined the fluorescent contrast $C_{F}$ between the patterned area and a region in flat Ag film outside of the cluster of patterned area using the following equation.

C_{F} = \frac{I_{F, P} - I_{b k g}}{I_{F, S} - I_{b k g}}

where

I_{F, P}

and

I_{F, S}

are the average fluorescent intensities measured from the patterned area and that from the flat area, respectively, and

I_{b g k}

is the intensity collected without the incident light illuminating the sample. This was measured at the dark band at the top of image in Fig. 4(a). The calculated

C_{F}

of each pattern is plotted as a function of pit size in Fig. 4(c). Maximum contrast occurs to pit size around 175 nm, and the contrast decreases as the pit size moves away from 175 nm.

Fig. 4 (a) Fluorescent image of Cy3 tagged-molecules coated Ag nano-pit array. The power of the focused excitation beam is 5.8μW. Each square consists of an array of square pits. The pit sizes for the top row from left to right are 400nm, 300nm, and 250nm, for the middle row: 200nm, 175nm 150nm, 125nm, and for the bottom row: 100nm 75nm. The overall size of each array is 7.5 μm. (b) Intensity line profiles scanned across pit arrays in (a). The top, middle, and bottom curves correspond to scans from left to right across arrays in the top, middle, bottom row, respectively. (c) The fluorescent contrast of the pit array (filled triangles) and calculated incident light E-field enhancement of the pit structure (open triangles) plotted as a function of pit size. The dashed line represents the results of including the field strength near the vertical side walls of the pits. (d) Calculated |E|² distribution of the incident light (532nm wavelength in vacuum) on pit array. Panels in the top row show the |E|² distribution on an incident plane across pit center of a unit cell of the pit array. The incident orientation and the E-field polarization are indicated in the right-most panel.|E|² of the incident light is set to unity. The size of the pit is labeled at the bottom of each panel. The pit edges are designed with finite curvature according to FESEM images of the pit structure. Corresponding panels in the bottom row show |E|² distribution at positions 8 nm above the Ag surfaces shown in the top row.

Download Full Size | PDF

Anticipating that localized incident field enhancement should occur to the nano-pit array, we carried out numerical calculations using the finite difference time domain (FDTD) method [20] to simulate the interaction between the incident light and the pit array structures. Figure 4(d) shows the distribution of the electric field strength |E|² in a unit cell of each pit array in a cross-sectional view and at positions 8 nm above silver surfaces, where Cy3 molecules were presumably located. It is clear that localized field enhancement occurs for all structures. For the largest pit structures, i.e. the 400 nm size pit, the field enhancement takes place at the Ag pad inside the pit, the corners of the pit and top Ag surfaces in between the pits. As the pit size decreases, the field enhancement first localized to the pit edges and expands outwards on the top Ag surfaces. For the 175 nm pit size, high field enhancement occurs over almost all the top Ag surfaces. As the pit size further decreases, the location of field enhancement first retracts and then shifts down to the Ag pad inside the pit. To compare with the experiment, we calculated the average field enhancement factor FE of each pit structure by dividing the average of the |E|² distribution shown in the panels in the bottom row in Fig. 4(d) by the average |E|² at positions 8 nm from the Ag surface in the simulation for the case of unpatterned substrate. The result is summarized in Fig. 4(c). A comparison shows that the simulation agrees qualitatively with the experiment. Both indicate that the 175 nm pit array produces the highest fluorescent contrast and field enhancement. Note that the surface plasmon wavelength (λ_spp) at the interface of water and Ag is ~370 nm for light with a 532 nm wavelength in vacuum. This is indeed quite close to the 350nm periodicity of the 175nm size pit array. Therefore the high field enhancement could be attributed to a plasmonic standing wave phenomenon occurring in this pit array. There is discrepancy between the simulation and the experiment for pit size between 100 nm and 150 nm. We refine the comparison by adding the contribution of the enhanced E-field near the side walls of the top Ag film, and the results is shown in Fig. 4(c). The agreement between the experiment and the simulation improves for arrays of smaller size pits but the peak wavelength i.e. maximum field enhancement shifts toward longer wavelengths in the simulation in this case. The partial agreement between the simulation and the experiment suggests that field enhancement is necessary but not sufficient to account for the overall trend of SEF contrast observed here. However, it is clear that the field enhancement contributes significantly to the observed enhanced fluorescence from the patterned area.

4.2 Incident power dependence of fluorescent contrast

We next explore the major issue motivating this paper i.e. the excitation power dependence of the fluorescent contrast of the sample. Figure 5(a) shows the fluorescent intensity measured from the patterned area plotted as a function of the pit size with various excitation powers. The fluorescent intensities increase monotonically with excitation power, which agrees with the prediction of the simple model we presented above and illustrated in Fig. 1. However, Fig. 5(b) shows that the fluorescent contrast decreases monotonically with excitation power for all pit sizes. Figure 5(c) shows specifically the fluorescent intensities measured from the 175 nm size pit array and that from the flat area on the Ag film over a wider range of excitation power as well as the fluorescent contrast between them. The excitation power dependence of the fluorescent intensities and the contrast is similar to that in Fig. 1(d). We thus fitted the fluorescent contrast data in Fig. 5(c) with a functional form consistent with Eq. (4), which we rewrite in a simpler from:

C_{F} (I) = \frac{b \cdot I + c}{a \cdot I + 1}

where C_F(I) is the fluorescent contrast, I the excitation power, and a, b, and c are taken to be fitting parameters. The result of the best fit is the solid curved plotted in Fig. 5 (c). The agreement between the model and data is good. It indicates that compared to the case of Cy3-tagged molecules on flat area the enhancement of the excitation rate of Cy3 tags coated on 175nm pit array area is stronger than the enhancement of the total decay rate of excited Cy3 tags. Comparing Eqs. (4) and (6), it is clear that e₂ = b/a, and

e_{1} / η = c / e_{2}

.We performed the same fitting for data from pit arrays with pit size ranging from 100nm to 250nm, where the measured fluorescent contrast is less noisy for a reasonable fit, and calculated

e_{2}

and

e_{1} / η

from the fitting parameters in Eq. (6). Figures 5(d) and 5(e) show the estimated values of

e_{2}

and

e_{1} / η

as a function of pit size, respectively. It is clear that the fluorescent emission ratewas enhanced for all pit arrays with size from 100 nm to 250 nm, with slightly higher enhancement for smaller sizes. The fact that the ratio

e_{1} / η

>1 occurs for all of the sizes in Fig. 5(e) indicate that the pit arrays enhances the excitation rate of the fluorescent molecule more than the total decay rate of excited molecules. Since the excitation rate is proportional to the field strength |E|² of the incident light, we used the values of the numerically calculated near-field enhancement factor shown in Fig. 4(c) for e₁ and estimated

η

for arrays of the same pit size range. The results are shown in Fig. 5(f). It is clear that the total decay rates of excited Cy3 molecules on nano-pit array patterned Ag film were not substantially modified compared to the case of flat Ag film. This is qualitatively consistent with previous results [21], except for the case of 100nm size pit array. For high fluorescent contrast region such as the 175 nm and the 150 nm size pit arrays, a shallow minimum in

η

occurs. Since the fluorescent emission rate was increased roughly by a factor between 2 and 5 as shown in Fig. 5(d), this implies that the fluorescent emission rate was enhanced at the cost of the non-radiative decay rate. Plausible explanation would be the nanostructures coupled the surface plasmon waves excited by the excited fluorescent molecules back to photons, or suppression of the excitation of lossy waves [13,18]. The case of 100 nm size pit array is exceptional to the general trend described above. Our estimation indicates that the total decay rate was significantly decreased, roughly by a factor of 10, and the fluorescent emission rate is enhanced by a factor of ~4. For the small spacing (~8nm) between the Cy3 and Ag surfaces in our experiment, the total decay rate is dominated by the excitation of lossy waves [13,18], which dissipate the energy through excitation of electron-hole pairs in the metal substrates. This implies that for 100 nm size pit array, the excitation of lossy wave could be significantly suppressed, and that the interaction between excited Cy3 molecules and pit array structure was significantly modified by the pit array. This may explain the discrepancy between the size dependence of the fluorescent contrast and the calculated field enhancement at smaller size pits shown in Fig. 4(c). Further experimental or theoretical investigations on the nature of this strong interaction are beyond the scope of this report.

Fig. 5 Excitation power dependence of (a) the fluorescent intensity and (b) the fluorescent contrast measured from the patterned area of the Ag film substrate. Plot with small, middle size, and large triangles represent the data measured with the excitation power equals to 2.0 μW, 5.8 μW, and 10.8 μW, respectively. (c) The Excitation power dependence of the fluorescent intensity of the 175 nm size pit array (filled circles), the fluorescent intensity of flat Ag film (filled squares), and the fluorescent contrast between the two (open diamonds) plotted as a function of excitation power. The solid curve represents a best fit to the fluorescent contrast data with the functional form of Eq. (6). (d), (e) and (f) Estimated fluorescent emission factor e₂, the ratio e₁/η and η as a function of the pit size, respectively. In (f) open circles (squares) represent results using the calculated incident field enhancement factor in Fig. 4(c) without (with) the contribution of the sidewalls. The error bars in panel (d), (e) and (f) result from propagation of errors of fitting parameters in Eq. (6).

Download Full Size | PDF

5. Conclusion

In conclusion, we observed that the fluorescent intensity emitted from Cy3-tagged molecules on a nano-pit array patterned Ag film increases monotonically with the incident power of the 532 nm wavelength excitation light. However, the fluorescent contrast between the patterned area and the surrounding unpatterned Ag film decreases with the excitation power. We show that this effect can be understood with a simple model in which the relative size of the enhancement of absorption and total decay rate is considered. The comparison between our measured variation of contrast with incident power and the predictions of this model shows that the pit array structure enhances the fluorescent emission rate by a factor between 2 and 5, with larger enhancement for smaller pit size, and produces a stronger effect on enhancing the excitation rate than that on the total decay rates of the excited Cy3 tags.

Acknowledgment

This work is supported by National Science Council, Taiwan, ROC under the grant number NSC 102-2112-M-194-004.

References and links

1. E. Fort and S. Gresillon, “Surface enhanced fluorescence,” J. Phys. D Appl. Phys. 41(1), 013001 (2008). [CrossRef]

2. C. L. Haynes, A. D. McFarland, and R. P. Van Duyne, “Surface-enhanced Raman spectroscopy,” Anal. Chem. 77(17), 338A–346A (2005). [CrossRef]

3. M. Moskovits, “Surface-enhanced spectroscopy,” Rev. Mod. Phys. 57(3), 783–826 (1985). [CrossRef]

4. S.-H. Guo, S.-J. Tsai, H.-C. Kan, D.-H. Tsai, M. R. Zachariah, and R. J. Phaneuf, “The Effect of an Active Substrate on Nanoparticle-Enhanced Fluorescence,” Adv. Mater. 20(8), 1424–1428 (2008). [CrossRef]

5. S.-H. Guo, J. J. Heetderks, H.-C. Kan, and R. J. Phaneuf, “Enhanced fluorescence and near-field intensity for Ag nanowire/nanocolumn arrays: evidence for the role of surface plasmon standing waves,” Opt. Express 16(22), 18417–18425 (2008). [CrossRef] [PubMed]

6. G. Zirubuabts and W. L. Barnes, “Fluorescence enhancement through modified dye molecule absorption associated with localized surface plasmon resonance of metallic dimmers,” New J. Phys. 10(10), 105002 (2008). [CrossRef]

7. E. M. Hicks, O. Lyandres, W. P. Hall, S. Zou, M. R. Glucksburg, and R. P. Van Duyne, “Plasmonic Properties of Anchored Nanoparticles Fabricated by Reactive Ion Etching and Nanosphere Lithography,” J. Phys. Chem. 111, 4116–4124 (2007).

8. S. Zou and G. C. Schatz, “Silver nanoparticle array structures that produce giant enhancements in electromagnetic fields,” Chem. Phys. Lett. 403(1-3), 62–67 (2005). [CrossRef]

9. F. Jäckel, A. A. Kinkhabwala, and W. E. Moerner, “Gold bowtie nanoantennas for surface-enhanced Raman scattering under controlled electrochemical potential,” Chem. Phys. Lett. 446(4-6), 339–343 (2007). [CrossRef]

10. A. Gopinath, S. V. Boriskina, B. M. Reinhard, and L. Dal Negro, “Deterministic aperiodic arrays of metal nanoparticles for surface-enhanced Raman scattering (SERS),” Opt. Express 17(5), 3741–3753 (2009). [CrossRef] [PubMed]

11. Q. Yu, P. Guan, D. Qin, G. Golden, and P. M. Wallace, “Inverted Size-Dependence of Surface-Enhanced Raman Scattering on Gold Nanohole and Nanodisk Arrays,” Nano Lett. 8(7), 1923–1928 (2008). [CrossRef] [PubMed]

12. E. M. Purcell, “Spontaneous emission probabilities at radio frequencies,” Phys. Rev. 69, 681 (1946).

13. W. L. Barnes, “Fluorescence near interfaces: the role of photonic mode density,” J. Mod. Opt. 45(4), 661–699 (1998). [CrossRef]

14. J. R. Lakowicz, “Radiative decay engineering: biophysical and biomedical applications,” Anal. Biochem. 298(1), 1–24 (2001). [CrossRef] [PubMed]

15. S. K. Saikin, Y. Chu, D. Rappoport, K. B. Crozier, and A. A. Aspuru-Guzik, “Separation of electromagnetic and chemical contributions to surface-enhanced Raman spectra on nanoengineered plasmonic substrates,” J. Phys. Chem. Lett. 1(18), 2740–2746 (2010). [CrossRef]

16. A. T. Zayak, Y. S. Hu, H. Choo, J. Bokor, S. Cabrini, P. J. Schuck, and J. B. Neaton, “Chemical Raman Enhancement of Organic Adsorbates on Metal Surfaces,” Phys. Rev. Lett. 106(8), 083003 (2011). [CrossRef] [PubMed]

17. 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]

18. G. W. Ford and W. H. Weber, “Electromagnetic Interactions of Molecules with Metal Surfaces,” Phys. Rep. 4, 197–287 (1984).

19. H. Gersen, M. F. García-Parajó, L. Novotny, J. A. Veerman, L. Kuipers, and N. F. van Hulst, “Influencing the Angular Emission of a Single Molecule,” Phys. Rev. Lett. 85(25), 5312–5315 (2000). [CrossRef] [PubMed]

20. T. Pistor, “Generalizing the TEMPEST FDTD Electro-magnetic Simulation Program,” UCB/ERL M97/52, EECS Department, UC Berkeley (1997).

21. R. M. Amos and W. L. Barnes, “Modification of spontaneous emission lifttimes in the presence of corrugated metallic surfaces,” Phys. Rev. B 59(11), 7708–7714 (1999). [CrossRef]

Nonlinear excitation power dependence of surface enhanced fluorescence from a nanostructured Ag film

Abstract

1. Introduction

2. Model of SEF on patterned substrate

3. Experimental methods

4. Results and discussion

4.1 Enhancement of fluorescence from patterned Ag film

4.2 Incident power dependence of fluorescent contrast

5. Conclusion

Acknowledgment

References and links

Cited By

Figures (5)

Equations (6)

Optics Express