Home
Search
Collections
Journals
About
Contact us
My IOPscience
Ordered arrays of Au-nanobowls loaded with Ag-nanoparticles as effective SERS substrates for rapid detection of PCBs
This content has been downloaded from IOPscience. Please scroll down to see the full text. 2014 Nanotechnology 25 145605 (http://iopscience.iop.org/0957-4484/25/14/145605) View the table of contents for this issue, or go to the journal homepage for more
Download details: IP Address: 128.103.149.52 This content was downloaded on 01/07/2014 at 01:54
Please note that terms and conditions apply.
Nanotechnology Nanotechnology 25 (2014) 145605 (8pp)
doi:10.1088/0957-4484/25/14/145605
Ordered arrays of Au-nanobowls loaded with Ag-nanoparticles as effective SERS substrates for rapid detection of PCBs Bensong Chen1 , Guowen Meng1,2 , Fei Zhou1 , Qing Huang3 , Chuhong Zhu1 , Xiaoye Hu1 and Mingguang Kong1 1
Key Laboratory of Materials Physics, and Anhui Key Laboratory of Nanomaterials and Nanostructures, Institute of Solid State Physics, Chinese Academy of Sciences, Hefei, Anhui 230031, People’s Republic of China 2 University of Science and Technology of China, Hefei, Anhui 230026, People’s Republic of China 3 Key Laboratory of Ion Beam Bioengineering, Institute of Technical Biology and Agriculture Engineering, Hefei Institutes of Physical Sciences, Chinese Academy of Sciences, Hefei, Anhui 230031, People’s Republic of China E-mail: [email protected] Received 21 October 2013, revised 19 January 2014 Accepted for publication 4 February 2014 Published 14 March 2014
Abstract
Large-scale hexagonally close-packed arrays of Au-nanobowls (Au-NBs) with tens of Ag-nanoparticles (Ag-NPs) dispersed in each bowl (denoted as Ag-NPs@Au-NB arrays) are achieved and utilized as effective surface-enhanced Raman scattering (SERS) substrates. The field enhancement benefiting from the special particle-in-cavity geometrical structure as well as the high density of SERS hot spots located in the sub-10 nm gaps between adjacent Ag-NPs and at the particle–cavity junctions all together contribute to the high SERS activity of the Ag-NPs@Au-NB arrays; meanwhile the ordered morphological features of the Ag-NPs@Au-NB arrays guarantee uniformity and reproducibility of the SERS signals. By modifying the Ag-NPs@Au-NB arrays with mono-6-thio-β-cyclodextrin, the SERS detection sensitivity to 3,30 ,4,40 -tetrachlorobiphenyl (PCB-77, one congener of polychlorinated biphenyls (PCBs, kinds of persistent organic pollutants which represent a global environmental hazard)) can be further improved and a low concentration down to 5 × 10−7 M can still be examined, showing promising potential for application in rapid detection of trace-level PCBs in the environment. Keywords: Ag-NPs@Au-NB arrays, surface-enhanced Raman scattering, mono-6-thio-β-cyclodextrin, polychlorinated biphenyls S Online supplementary data available from stacks.iop.org/Nano/25/145605/mmedia (Some figures may appear in colour only in the online journal)
1. Introduction
SERS enhancement is to obtain highly active SERS hot spots (sub-10 nm gaps between noble metallic nanostructures) [6– 11]. In addition, uniformity and reproducibility of SERS signals are also critical for SERS applications, and are strongly dependent on the homogeneity of the structure and the morphology of the SERS substrate. Thus, it is highly desired to construct SERS substrates with not only sufficient
Surface-enhanced Raman scattering (SERS) is considered as one of the most powerful tools for detecting trace-level molecules, owing to its ultra-high detection sensitivity, realtime response and fingerprint effect [1–5]. Experimental and theoretical studies demonstrate that the key to acquiring strong 0957-4484/14/145605+08$33.00
1
c 2014 IOP Publishing Ltd
Printed in the UK
Nanotechnology 25 (2014) 145605
B Chen et al
highly active SERS hot spots, but also pre-controlled ordered structures. It has been demonstrated that the nanosphere-based approach [12] and the electron beam lithography approach [13, 14] are very effective to produce highly active and reproducible SERS substrates. However, these methods usually require complicated synthesis procedures and expensive equipment. To date, many other efforts have also been made to exploit various SERS substrates with appropriate nanostructured architectures, such as one-dimensional (1D) metallic nanoparticle chains [15], two-dimensional (2D) close-packed metallic nanoparticle arrays [16] and three-dimensional (3D) ordered composite micro- and nanoarchitectures [17–19], to achieve high SERS sensitivity and reproducible SERS signals. Among them, a kind of ‘particle-in-cavity’ structure has been theoretically predicted to produce extremely strong field enhancement resulting from cascaded focusing of large optical cross-sections into small gaps [20–22]. Recently, ordered arrays of Ag-nanoparticles in Au-nanobowls (AgNP-in-Au-NB) based on particle-in-cavity structure have been experimentally realized and show good signal reproducibility and excellent SERS performance, arising from their ordered structure and strong field enhancement of the highly active SERS hot spots mainly located at the particle–cavity junctions [23]. However, the fabrication process for the Au-nanobowls of the Ag-NP-in-Au-NB arrays is based on multiple steps, which include (1) self-assembly of close-packed monolayer polystyrene (PS) nanosphere arrays, (2) oxygen plasma etching to decrease the diameter of the PS nanospheres, (3) atomic layer deposition of a sacrificial ZnO layer on the PS nanospheres, (4) thermal annealing to remove the PS, (5) deposition of Au film on the ZnO, (6) etching of the ZnO by HCl solution, and (7) transformation into Au-nanobowl arrays. In addition, there is only one Ag-nanoparticle (Ag-NP) located on the bottom of each Au-nanobowl (Au-NB) in the Ag-NP-in-Au-NB arrays and highly sensitive SERS hot spots are merely formed at the particle–cavity junctions between the Ag-NPs and Au-NBs, leading to a relatively low density of SERS hot spots. Herein, we attempted to synthesize effective SERS substrates based on the particle-in-cavity structure with the following considerations. Firstly, since several convenient synthesis methods have been reported to construct ordered concave arrays [24, 25], the complicated multiple-step fabrication process can be replaced with a simple porous anodic aluminum oxide (AAO) template assisted approach to produce large-scale highly ordered nanobowl arrays. Secondly, by assembling tens of Ag-NPs rather than only one Ag-NP on the concave surface of each Au-NB, a higher density of SERS hot spots can be easily formed in the sub-10 nm gaps between neighboring Ag-NPs as well as at the particle–cavity junctions between the Ag-NPs and the Au-NB. With these two considerations, we fabricated large-scale hexagonal-packed arrays of AuNBs, with each bowl having tens of Ag-NPs with sub-10 nm gaps on its curvature surface (denoted as Ag-NPs@Au-NB), via simple successive sputtering of Au film and Ag-NPs on concave arrays on Al foil achieved by using an AAO template. Due to the high density of sensitive SERS hot spots located between the neighboring Ag-NPs and at the
Scheme 1. Schematic for the fabrication of Ag-NPs@Au-NB
arrays.
particle–cavity junctions, the morphological homogeneity, and the field enhancement benefiting from the particle-in-cavity structure, the Ag-NPs@Au-NB arrays can serve as effective SERS substrates with high SERS activity and reproducible signals. Finally, we tried to graft mon-6-thio-β-cyclodextrin (HS-β-CD) on the Ag-NPs@Au-NB arrays for SERS-based detection of polychlorinated biphenyls (PCBs), which are highly toxic persistent organic pollutants as defined in the Stockholm Convention and cause great threats to the global environment and human health [26]. By virtue of the HSβ-CD chemical modification, SERS detection of 3,30 ,4,40 tetrachlorobiphenyl (PCB-77, one congener of PCBs) at a concentration of 5 × 10−7 M has been achieved, demonstrating that the as-fabricated Ag-NPs@Au-NB arrays modified with HS-β-CD have great potential as effective SERS substrates for rapid detection of trace PCBs. 2. Experimental methods 2.1. Preparation of the Ag-NPs@Au-NB arrays
The synthesis procedure for the Ag-NPs@Au-NB arrays on Al foil (schematically shown in scheme 1) can be divided into the following steps: (1) Al foil is anodized in a twostep anodization [27] to obtain a highly ordered porous AAO membrane on the Al foil (anodized at 40 V in oxalic acid at 10 ◦ C, with a pore diameter of about 70–80 nm); (2) the topmost ordered porous AAO membrane is thoroughly removed in 6 wt% phosphoric and 1.8 wt% chromic acid solution (at 50 ◦ C, for 8 h), leaving large-area highly ordered hexagonal-packed hemispherical nanoconcave arrays on the top of the remaining Al foil; (3) a uniform thin Au layer (∼10 nm) is deposited onto the cavity surface of the concave arrays to form ordered Au-NB arrays on the Al foil; (4) Ag-NPs are ion-sputtered onto each of the Au-NBs to achieve largearea (up to cm2 ) ordered hexagonal-packed Ag-NPs@Au-NB arrays on the remaining Al foil. The sputtering of Au film on the concave arrays on the remaining Al foil was carried out in a homemade electron beam evaporation system (DZS-500, Shenyang Scientific Instrument Co., Ltd) with a film thickness monitor (FTM-V). The deposition of Ag-NPs was carried out in an ion-sputtering coater (Emitech, K550X) at a rate of 4 nm min−1 (adjusting the duration from 2 to 10 min) while keeping the sputtering stage stationary. 2
Nanotechnology 25 (2014) 145605
B Chen et al
The Raman spectra were recorded using a 532 nm laser and a 20× short focal length objective in a confocal microprobe Raman system (Renishaw, inVia). The laser spot focused on the sample surface was about 5 µm in diameter. The integration duration for SERS spectral recording was chosen as 10 and 30 s for R6G and PCB-77, while the laser power was 1 and 5 mW for R6G and PCB-77, respectively. 2.3. Modification of the Ag-NPs@Au-NB arrays with HS-β -CD
As-prepared small pieces of Ag-NPs@Au-NB arrays were immersed in HS-β-CD dissolved in N ,N -dimethylformamide (DMF) solution (0.5 ml, 0.1 mM) for 5 h for chemical modification. After the modification treatment was finished, the small pieces of SERS substrate were taken out, rinsed with DMF, and finally dried in high-purity nitrogen flow. Figure 1. Top-view SEM images of (a) concave arrays, (b) Au-NB arrays, (c) Ag-NPs@Au-NB arrays on the remaining Al foil. (d) Enlarged SEM image of the Ag-NPs@Au-NB arrays. The insets in (c) and (d) are oblique and close-up views of the Ag-NPs@Au-NB arrays, respectively.
3. Results and discussion 3.1. Characterization of the Ag-NPs@Au-NB arrays
Figure 1(a) is a representative top-view SEM image of the hemispherical concave arrays on the remaining Al foil after thorough removal of the topmost ordered porous AAO membrane, revealing the periodic hexagonal-packed arrangement of geometrical features which is an exact replica of the morphology of the AAO-nanopore bottoms. Figure 1(b) shows the top-view SEM image of the sputtered thin Au film on the concave arrays, demonstrating that the thin Au film is composed of close-packed tiny Au-nanoparticles and uniformly covers the entire surface of the concave arrays. Thus, it is feasible to treat the sputtered Au thin film as dense and continuous Au-NB arrays. The top-view SEM image of Ag-NPs@Au-NB arrays (figure 1(c)) reveals that a large number of Ag-NPs are evenly dispersed on the cavities and the edges of the apertures of the Au-NBs. The enlarged view (figure 1(d)) further displays that the average diameter of a nanobowl unit and the average inter-gap spacing of adjacent Ag-NPs are about 100 and 5 nm, respectively. The components of the Au-NB arrays and Ag-NPs@Au-NB arrays supported on Al foil are confirmed by energy-dispersive x-ray spectroscopy (EDS) measurements (figure S1, supporting information available at stacks.iop.org/Nano/25/145605/mmedia).
For a comparison of SERS detection sensitivity between the Ag-NPs@Au-NB arrays and Ag-NPs dispersed on flat polished Al foil with Au film (denoted as Ag-NPs@Au), the polished Al was prepared by electropolishing Al foil in a mixture of perchloric acid and absolute ethanol (at a ratio of 1:9 in volume) at 1 ◦ C for 3 min with a voltage of 20 V. 2.2. Characterizations
The morphology and components of the resultant samples were characterized by using scanning electron microscopy (SEM, Sirion 200, FEI, at 5 kV) with energy-dispersive x-ray spectroscopy (EDS, Oxford). For the SERS measurements, the as-prepared Ag-NPs@ Au-NB arrays were cut into small square pieces (5 mm × 5 mm) and immersed in Rhodamine 6G (R6G) aqueous solutions (0.25 ml) of different concentrations (from 10−6 to 10−10 M) for 20 min, then taken out and rinsed with deionized water, and finally dried with high-purity flowing nitrogen before Raman spectral examination. For the comparison of SERS activity between the AgNPs@Au-NB arrays and Ag-NPs@Au, 1 µl of 1 × 10−7 M R6G aqueous solution was respectively dropped and uniformly dispersed onto pieces of Ag-NPs@Au-NB array and AgNPs@Au with the same area (5 mm × 5 mm). Then, the samples were carefully dried in flowing nitrogen atmosphere under room temperature. Raman measurements for these substrates were carried out under identical experimental conditions (such as laser wavelength and power, microscope objective/lenses, integration duration, etc). To check the SERS sensitivity of the Ag-NPs@Au-NB arrays for PCB-77, PCB-77 was firstly dissolved in acetone to form solutions and then small pieces of SERS substrate were respectively immersed in PCB-77 solutions (0.25 ml) of different concentrations (10−4 , 10−5 , 5 × 10−6 M) for 5 h, taken out and cleaned with acetone, and finally dried with high-purity flowing nitrogen in a laboratory fume cupboard before SERS measurements.
3.2. Modulation and characterization of the SERS performance of the Ag-NPs@Au-NB arrays
Since both theoretical and experimental studies demonstrate that control of SERS hot spots within the sub-10 nm regime is essential to obtain high SERS activity [28, 29], we tailored the diameter and the inter-particle spacing of the Ag-NPs in the Ag-NPs@Au-NB arrays by adjusting the Ag-sputtering conditions (sputtering rate and duration) so as to acquire the optimal SERS activity. Figure 2(a) shows the SERS spectra of 10−7 M R6G aqueous solution dispersed on Ag-NPs@AuNB arrays with different Ag-sputtering durations from 2 to 10 min with a given sputtering rate. The SERS activity of the substrates improves with increase of the Ag-sputtering duration from 2 to 6 min, but gradually decreases with 3
Nanotechnology 25 (2014) 145605
B Chen et al
and diameter and gap distance distributions in figures S3(c) and (d), supporting information available at stacks.iop.org/ Nano/25/145605/mmedia), which is the optimal regime for forming highly active SERS hot spots. With further elongation of the Ag-sputtering duration, larger Ag-NPs agglomerate together and even form a continuous Ag-NPs film over the cavity surfaces of the Au-NB arrays (figure S2(c), supporting information available at stacks.iop.org/Nano/25/145605/mme dia), leading to narrower or even disappearance of the gaps between the neighboring Ag-NPs, so that the SERS activity deteriorates. To further investigate the SERS performance of the Ag-NPs@Au-NB arrays with 6 min Ag-sputtering, we tested their SERS performance using R6G aqueous solutions of different concentrations (from 10−6 to 10−10 M). As shown in figure 2(b), distinct bands can still be easily distinguished in the Raman spectrum even at the low concentration of 10−10 M, and the peaks at 611 and 774 cm−1 are assigned to the C–C–C in-plane and out-of-plane bending vibrations, respectively [29, 30]. The average SERS enhancement factor (EF) of the Ag-NPs@Au-NB arrays with 6 min Ag-sputtering was estimated by using 4-aminothiophenol (4-ATP) as test molecules and the EF value was estimated to be about 3.7 × 104 at the 1075 cm−1 band of 4-ATP (details are given in part S2, supporting information available at stacks.iop.org/ Nano/25/145605/mmedia). To quantitatively estimate the electromagnetic field distribution of our Ag-NPs@Au-NB arrays with optimized SERS performance (with 6 min Ag-sputtering), a 3D finitedifference-time-domain (FDTD) simulation was implemented to simulate the 3D spatial electric field profile of the AgNPs@Au-NB arrays (the mesh size in our simulation was 1 nm). A single Ag-NPs@Au-NB unit is illuminated from the top with a 532 nm linear polarized plane wave, as schematically depicted in figure 3(a). For simplicity, the cavity of the Au-NB is considered as a hemispherical shape with the average diameter and thickness being chosen as 100 and 10 nm, respectively. The Ag-NPs sputtered on the cavity of the Au-NB are treated as having spherical shape and the diameter of the Ag-NPs and the inter-gap spacing of the Ag-NPs are set as 17 nm and 5 nm, respectively. The Ag-NPs located at the edge of the aperture of the Au-NB are slightly larger and their diameter is chosen as 24 nm (see the enlarged SEM image in figure 1(d)). Figures 3(b) and (c) display the vertical and horizontal cross-sectional views of the aperture of the Au-NB from the FDTD simulated electric field profile of the Ag-NPs@Au-NB arrays, respectively. It can be seen that strong field enhancement mainly occurs in three kinds of positions: the sub-10 nm gaps located between neighboring Ag-NPs on the cavity surface of the Au-NB (position I, marked in figure 3(b)) and the edge of the aperture of the Au-NB (position II, marked in figure 3(c)) as well as the particle–cavity junction area between the Ag-NPs and the Au-NB (position III, marked in figure 3(b)). The maximum field enhancement values |E/E 0 | for the vertical and horizontal cross-sectional space are respectively 13.15 and 15.05, which can be attributed to the fact that the Ag-NPs located on the platform of the edge of the aperture of the Au-NB are larger than the Ag-NPs on the cavity of the Au-NB and can produce stronger field enhancement in the inter-particle spacing.
Figure 2. (a) SERS spectra of 10−7 M R6G adsorbed on the
Ag-NPs@Au-NB arrays with different Ag-sputtering durations (2, 4, 6, 8, and 10 min). (b) SERS spectra collected on the Ag-NPs@Au-NB arrays (with 6 min Ag-sputtering) exposed to different concentrations of R6G aqueous solution (10−6 , 10−7 , 10−8 , 10−9 and 10−10 M).
longer Ag-sputtering durations. The morphological evolution of the Ag-NPs@Au-NB arrays associated with the different Ag-sputtering durations can serve to explain this phenomenon. For the substrate with 2 min Ag-sputtering, only small Ag-NPs with an average diameter of ca. 10 nm and average gap distance of ca. 15 nm (figure S2(a), supporting information available at stacks.iop.org/Nano/25/145605/mmedia) are dispersed on the cavity surface and the edge of the aperture of the Au-NB arrays. The corresponding diameter and inter-particle gap distance distributions of the 2 min sputtered Ag-NPs are shown in figures S3(a) and (b) of the supporting information (available at stacks.iop.org/Nano/25/145605/mmedia). The relatively small diameter and large gap distance of the sputtered Ag-NPs are not beneficial to inducing strong field enhancement between adjacent Ag-NPs as well as the junctions of Ag-NPs and the Au-NB [23], resulting in poor SERS activity. By prolonging the Ag-sputtering duration to 6 min, the deposited Ag-NPs become larger (about 17 nm in average diameter) and large quantities of sub-10 nm inter-particle gaps and particle– cavity junctions are formed (SEM image in figure S2(b) 4
Nanotechnology 25 (2014) 145605
B Chen et al
Figure 3. 3D FDTD simulations of Ag-NPs@Au-NB arrays. (a) The 3D simulation model. (b) The vertical and (c) the horizontal cross-sectional view of the aperture of the Au-NB from the FDTD simulated electric field profile. The maximum field enhancement |E/E 0 | values for the vertical and horizontal cross-sectional space are 13.15 and 15.05, respectively.
p.org/Nano/25/145605/mmedia). Figures 4(b) and (c) show the representative morphology of Ag-NPs@Au-NB arrays and Ag-NPs@Au with their corresponding optimal SERS performance, respectively. When exposed to 10−7 M R6G aqueous solution under identical experimental conditions, the Ag-NPs@Au-NB arrays show higher SERS activity than that of Ag-NPs@Au under their corresponding optimized sputtering conditions (figure 4(d)). In addition, the average EF of the Ag-NPs@Au substrate (with optimal SERS property) was estimated to be about 2.5 × 104 at the 1075 cm−1 band of 4-ATP (details are shown in part S2, supporting information available at stacks.iop.org/Nano/25/145605/mmedia), and this value is smaller than that for the Ag-NP@Au-NB arrays (with optimized 6 min Ag-sputtering), demonstrating that the Ag-NP@Au exhibits weaker average field enhancement. To further understand the benefit of the particle-in-cavity geometrical structure of the Ag-NPs@Au-NB arrays to the field enhancement, we theoretically simulated and compared the field enhancement of Ag-NPs@Au-NB arrays and that of Ag-NPs@Au. The diameter of the Ag-NPs and the interparticle spacing used in the simulations for Ag-NPs@Au are all the same as those in the simulation of Ag-NPs@Au-NB arrays with 6 min Ag-sputtering. The simulated FDTD electric field profiles of the Ag-NPs@Au (detailed simulated results are shown in figure S8, supporting information available at stacks.iop.org/Nano/25/145605/mmedia) display that the SERS hot spots are mainly located in the inter-particle gaps and the junctions between Ag-NPs and the flat Au surface, showing weaker field enhancement at the SERS hot spots of Ag-NPs@Au compared to that of Ag-NPs@Au-NB arrays. It should be noted that there is a small deviation between the simulated and experimental comparisons of SERS activity between the Ag-NPs@Au-NB arrays and Ag-NPs@Au, which we believe comes from the assumptions and approximations made in the calculations. Nevertheless, the simulated results clearly illustrate that a higher field enhancement effect can be produced by assembling Ag-NPs on the Au cavity surface as compared to putting similar Ag-NPs on a flat Au surface with the same inter-particle spacing.
To test the uniformity and reproducibility of the AgNPs@Au-NB arrays as effective SERS substrates, SERS spectra from ten randomly selected positions across the whole substrate were recorded under identical experimental conditions, including the same laser power and integration time (figure S4, supporting information available at stacks.iop.org /Nano/25/145605/mmedia). The results show that the relative standard deviation (RSD) of the SERS intensity for the Raman peak at 611 cm−1 of 10−7 M R6G is about 8%, confirming the good uniformity of the measured SERS signals. The good reproducibility of the signals can also be demonstrated in the SERS measurement of R6G at different concentrations (shown in figure S5, supporting information available at stacks.io p.org/Nano/25/145605/mmedia). The excellent SERS signal uniformity and reproducibility of the Ag-NPs@Au-NB arrays derives from the fact that their morphology well replicates the morphological features of the native highly ordered porous AAO template. 3.3. Benefit of the particle-in-cavity geometrical structure of the Ag-NPs@Au-NB arrays for field enhancement
Theoretical and experimental studies reveal that metallic nanovoid [31–33] and particle-in-cavity architectures [20, 21, 34] can be used for plasmonic engineering and production of strong field enhancement. In order to demonstrate the benefit of the particle-in-cavity geometrical structure of the AgNPs@Au-NB arrays to field enhancement, we compared the SERS activity between Ag-NPs@Au-NB arrays and Ag-NPs sputtered on a flat polished Al sheet with a thin Au film (denoted as Ag-NPs@Au, with the same 10 nm Au film thickness as that of the Au-NBs in the Ag-NPs@Au-NB arrays), as shown in the schematic architectures in figure 4(a). In a similar way to how we optimized the SERS activity of the Ag-NPs@Au-NB arrays, we tailored and optimized the SERS performance of the Ag-NPs@Au by adjusting the Agsputtering duration (the details of the morphological evolution and variation of SERS performance of the Ag-NPs@Au with different Ag-sputtering durations are shown in figures S6 and S7 respectively, supporting information available at stacks.io 5
Nanotechnology 25 (2014) 145605
B Chen et al
Figure 4. (a) Schematic drawings of Ag-NPs@Au-NB arrays (substrate I) and Ag-NPs@Au (substrate II). Top-view SEM images of (b) substrate I and (c) substrate II with the corresponding optimized SERS performance. (d) The SERS spectra collected on substrate I (curve I) and substrate II (curve II) (with the corresponding optimized SERS performance) exposed to 10−7 M R6G aqueous solution.
of the main peak bands of PCB-77 versus concentration are shown in figure S9, supporting information (available at stacks.iop.org/Nano/25/145605/mmedia). The reason why the Ag-NPs@Au-NB arrays as SERS substrates exhibit higher SERS sensitivity to R6G than PCBs might originate from the fact that the bare surface of the metallic building blocks has a relatively poor ability to adsorb the hydrophobic PCB molecules. Surface modification of an SERS substrate is regarded as a useful solution to capture target molecules onto the hot spots of the SERS substrate and improve the SERS detection sensitivity. Since theoretical simulations have demonstrated that a PCB molecule can enter the cavity of β-cyclodextrin (β-CD) to form an inclusion complex with β-CD [40, 41] and strong chemical bonding can be formed between the mercapto functional group (HS-) of organic molecules and the metallic Ag surface, we chose HS-β-CD as a surface modifying agent and functionalized the as-prepared Ag-NPs@Au-NB arrays with HS-β-CD to capture more PCB-77 molecules onto the hot spots so as to improve the SERS sensitivity to PCB-77. Figure 5(b) displays the SERS spectra of Ag-NPs@Au-NB arrays modified with HS-β-CD for detection of PCB-77 solutions of 5 × 10−6 (curve IV) and 5 × 10−7 M (curve V), respectively. Comparing the measured SERS signals of curve IV with those of curve III in figure 5(a), it can be seen that the peak intensities of PCB-77 are increased nearly two-fold after HS-β-CD modification, verifying that modification of the Ag-NPs@Au-NB arrays with HS-β-CD can indeed enhance their ability to capture the target PCB molecules. The phenomenon of no pronounced bands of HS-β-CD emerging in the SERS spectra of Ag-NPs@Au-NB arrays modified with HS-β-CD for detection of PCB-77 may be due to the fact that the intrinsic Raman intensity of HS-β-CD is much weaker than that of PCB-77 (figure S11
The above comparisons of the measured SERS activity and the simulated field enhancement unambiguously confirm the benefit of the particle-in-cavity nanostructured Ag-NPs@Au-NB arrays, and the improvement of the SERS signals can be explained as follows. Firstly, the void-like AuNBs of the Ag-NPs@Au-NB arrays serve as light harvesters and focus the light in the cavity. The introduced Ag-NPs in the cavity of the Au-NB can strengthen the particle–cavity coupling effect when the nanoparticle mode is resonant with one of the cavity modes [20]. Secondly, in comparison to the junction between Ag-NPs and a flat Au surface, the junction of Ag-NPs and the negative curvature cavity of the Au-NB favors stronger excitation of field enhancement under a vertically incident laser. Finally, a higher density of SERS hot spots can be generated in the particle-in-cavity structure compared to that of particles dispersed on a flat surface under the same horizontal projected areas. 3.4. SERS sensitivity of the Ag-NPs@Au-NB arrays to PCBs
Finally, the as-prepared Ag-NPs@Au-NB arrays were examined as effective SERS substrates for the detection of PCB-77, one of 209 congeners of PCBs. Figure 5(a) displays the SERS spectra of PCB-77 at different concentrations of 10−4 (curve I), 10−5 (curve II) and 5 × 10−6 M (curve III) on the as-prepared bare Ag-NPs@Au-NB arrays. There are five distinct bands at 674 cm−1 (C–Cl stretching) [35, 36], 1030 cm−1 (ring breathing) [36, 37], 1244 cm−1 (C–H wagging) [35, 36], 1297 cm−1 (biphenyl C–C bridge stretching) [35, 36], and 1598 cm−1 (ring stretching) [36, 37] in the SERS spectra, which agree well with the normal Raman spectrum of PCB-77 [38, 39]. In the SERS spectra, it can be seen that the representative peaks still exist even when the concentration of measured PCB-77 solution decreases down to 5 × 10−6 M by using the bare AgNPs@Au-NB arrays as SERS substrates. The SERS intensities 6
Nanotechnology 25 (2014) 145605
B Chen et al
Ag-NPs and at the particle–cavity junctions as well as the particle-in-cavity geometrical structure. The morphological uniformity and homogeneity also ensure reproducibility of the SERS detection signals of the Ag-NPs@Au-NB arrays. Furthermore, by modifying the as-prepared Ag-NPs@AuNB arrays with HS-β-CD, the SERS substrate can capture hydrophobic molecules such as PCBs. The detected PCB-77 concentration can be as low as 5 × 10−7 M. Therefore, our work demonstrates that the as-fabricated Ag-NPs@Au-NB arrays modified with HS-β-CD have promising potential in SERS-based rapid detection of organic environmental pollutants such as PCBs. Acknowledgments
This work was supported by the National Key Basic Research Program of China (Grant No. 2013CB934304), the National Natural Science Foundation of China (Grant Nos 21107113 and 11274312), and the Postdoctoral Science Foundation of China (Grant No. 2012M510164). References [1] Fleischm M, Hendra P J and McQuilla A J 1974 Raman-spectra of pyridine adsorbed at a silver electrode Chem. Phys. Lett. 26 163–6 [2] Jeanmaire D L and Vanduyne R P 1977 Surface Raman spectroelectrochemistry: I. Heterocyclic, aromatic, and aliphatic-amines adsorbed on the anodized silver electrode J. Electroanal. Chem. 84 1–20 [3] Albrecht M G and Creighton J A 1977 Anomalously intense Raman-spectra of pyridine at a silver electrode J. Am. Chem. Soc. 99 5215–7 [4] Kneipp K, Wang Y, Kneipp H, Perelman L T, Itzkan I, Dasari R and Feld M S 1997 Single molecule detection using surface-enhanced Raman scattering (SERS) Phys. Rev. Lett. 78 1667–70 [5] Li J F et al 2010 Shell-isolated nanoparticle-enhanced Raman spectroscopy Nature 464 392–5 [6] Qin L D, Zou S L, Xue C, Atkinson A, Schatz G C and Mirkin C A 2006 Designing, fabricating, and imaging Raman hot spots Proc. Natl Acad. Sci. USA 103 13300–3 [7] Camden J P, Dieringer J A, Wang Y M, Masiello D J, Marks L D, Schatz G C and Van Duyne R P 2008 Probing the structure of single-molecule surface-enhanced Raman scattering hot spots J. Am. Chem. Soc. 130 12616–7 [8] Lee S J, Morrill A R and Moskovits M 2006 Hot spots in silver nanowire bundles for surface-enhanced Raman spectroscopy J. Am. Chem. Soc. 128 2200–1 [9] Braun G, Pavel I, Morrill A R, Seferos D S, Bazan G C, Reich N O and Moskovits M 2007 Chemically patterned microspheres for controlled nanoparticle assembly in the construction of SERS hot spots J. Am. Chem. Soc. 129 7760–1 [10] Li W Y, Camargo P H C, Lu X M and Xia Y N 2009 Dimers of silver nanospheres: facile synthesis and their use as hot spots for surface-enhanced Raman scattering Nano Lett. 9 485–90 [11] Camargo P H C, Rycenga M, Au L and Xia Y N 2009 Isolating and probing the hot spot formed between two silver nanocubes Angew. Chem. Int. Edn 48 2180–4
Figure 5. (a) SERS spectra of PCB-77 collected on bare Ag-NPs@Au-NB arrays (6 min Ag-sputtering) exposed to 10−4 (curve I), 10−5 (curve II) and 5 × 10−6 (curve III) M PCB-77, respectively. The inset is the schematic of PCB-77. (b) SERS spectra of 5 × 10−6 (curve IV) and 5 × 10−7 (curve V) M PCB-77 from HS-β-CD-modified Ag-NPs@Au-NB arrays. The inset displays the schematic diagram of the inclusion complex of PCB-77 with HS-β-CD located in the gap between two neighboring Ag-NPs.
in part S3, supporting information available at stacks.iop.org /Nano/25/145605/mmedia). When the PCB-77 concentration is decreased down to 5 × 10−7 M, the representative bands of PCB-77 in the SERS spectra (curve V in figure 5(b)) can still be distinguished, though the peaks intensities decrease sharply. Therefore, this experiment confirms that the surface modification with HS-β-CD can improve the sensitivity for detecting PCBs such as PCB-77. 4. Conclusion
In conclusion, large-scale ordered Ag-NPs@Au-NB arrays as effective SERS substrates have been achieved via a porous AAO template assisted approach with a physical deposition route. By tailoring the inter-particle gaps of neighboring Ag-NPs and the diameters of the Ag-NPs dispersed in the Au-NBs, high SERS activity of the Ag-NPs@Au-NB arrays has been acquired, which benefits from the high density of sub-10 nm SERS hot spots formed between adjacent 7
Nanotechnology 25 (2014) 145605
B Chen et al
[12] Chan G H, Zhao J, Hicks E M, Schatz G C and Van Duyne R P 2007 Plasmonic properties of copper nanoparticles fabricated by nanosphere lithography Nano Lett. 7 1947–52 [13] Margueritat J et al 2011 Influence of the number of nanoparticles on the enhancement properties of surface-enhanced Raman scattering active area: sensitivity versus repeatability ACS Nano 5 1630–8 [14] Billot L, De la Chapelle M L, Grimault A S, Vial A, Barchiesi D, Bijeon J L, Adam P M and Royer P 2006 Surface enhanced Raman scattering on gold nanowire arrays: evidence of strong multipolar surface plasmon resonance enhancement Chem. Phys. Lett. 224 303–7 [15] Yang Y, Shi J L, Tanaka T and Nogami M 2007 Self-assembled silver nanochains for surface-enhanced Raman scattering Langmuir 23 12042–7 [16] Wei A, Kim B, Sadtler B and Tripp S L 2001 Tunable surface-enhanced Raman scattering from large gold nanoparticle arrays ChemPhysChem 2 743–5 [17] Huang Z L, Meng G W, Huang Q, Yang Y J, Zhu C H and Tang C L 2010 Improved SERS performance from Au nanopillar arrays by abridging the pillar tip spacing by Ag sputtering Adv. Mater. 22 4136–9 [18] Tang H B, Meng G W, Huang Q, Zhang Z, Huang Z L and Zhu C H 2012 Arrays of cone-shaped ZnO nanorods decorated with Ag nanoparticles as 3D surface-enhanced Raman scattering substrates for rapid detection of trace polychlorinated biphenyls Adv. Funct. Mater. 22 218–24 [19] Liu G Q, Cai W P, Kong L C, Duan G T, Li Y, Wang J J, Zuo G M and Cheng Z X 2012 Standing Ag nanoplate-built hollow microsphere arrays: controllable structural parameters and strong SERS performances J. Mater. Chem. 22 3177–84 [20] Huang F M, Wilding D, Speed J D, Russell A E, Bartlett P N and Baumberg J J 2011 Dressing plasmons in particle-in-cavity architectures Nano Lett. 11 1221–6 [21] Ye J, Chen C, Lagae L, Maes G, Borghs G and Van Dorpe P 2010 Strong location dependent surface enhanced Raman scattering on individual gold semishell and nanobowl particles Phys. Chem. Chem. Phys. 12 11222–4 [22] Ye J, Lagae L, Maes G and Van Dorpe P 2011 A versatile method to fabricate particle-in-cavity plasmonic nanostructures J. Mater. Chem. 21 14394–7 [23] Li X L, Zhang Y Z, Shen Z X and Fan H J 2012 Highly ordered arrays of particle-in-bowl plasmonic nanostructures for surface-enhanced Raman scattering Small 8 2548–54 [24] Lee W and Lee J K 2002 Non-lithographic approach to the fabrication of polymeric nanostructures with a close-packed 2D hexagonal array Adv. Mater. 14 1187–90 [25] Liang F X, Cheng J W, Tsang C K, Cheng H and Li Y Y 2011 TiO2 -based nano-wells for fabricating nanocrystals J. Nanosci. Nanotechnol. 11 11059–63 [26] Gambaro A, Manodori L, Zangrando R, Cincinelli A, Capodaglio G and Cescon P 2005 Atmospheric PCB
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
8
concentrations at Terra Nova Bay, Antarctica Environ. Sci. Technol. 39 9406–11 Masuda H and Fukuda K 1995 Ordered matel nanohole arrays made by a 2-step replication of honeycomb structures of anodic alumina Science 268 1466–8 Wang H H, Liu C Y, Wu S B, Liu N W, Peng C Y, Chan T H, Hsu C F, Wang J K and Wang Y L 2006 Highly Raman-enhancing substrates based on silver nanoparticle arrays with tunable sub-10 nm gaps Adv. Mater. 18 491–5 Wang H, Levin C S and Halas N J 2005 Nanosphere arrays with controlled sub-10-nm gaps as surface-enhanced Raman spectroscopy substrates J. Am. Chem. Soc. 127 14992–3 Hildebrandt P and Stockburger M 1984 Surface-enhanced resonance Raman-spectroscopy of rhodamine-6G adsorbed on colloidal silver J. Phys. Chem. 88 5935–44 Cole R M, Baumberg J J, Garcia de Abajo F J, Mahajan S, Abdelsalam M and Bartlett P N 2007 Understanding plasmons in nanoscale voids Nano Lett. 7 2094–100 Kelf T A, Sugawara Y, Cole R M and Baumberg J J 2006 Localized and delocalized plasmons in metallic nanovoids Phys. Rev. B 74 245415 Ye J, Van Dorpe P, Van Roy W, Borghs G and Maes G 2009 Fabrication, characterization, and optical properties of gold nanobowl submonolayer structures Langmuir 25 1822–7 Ross B M and Lee L P 2009 Creating high density nanoantenna arrays via plasmon enhanced particle-cavity (PEP-C) architectures Opt. Express 17 6860–6 Bantz K C and Haynes C L 2009 Surface-enhanced Raman scattering detection and discrimination of polychlorinated biphenyls Vib. Spectrosc. 50 29–35 Socrates G 2001 Infrared and Raman Characteristic Group Frequencies: Tables and Charts 3rd edn (New York: Wiley) pp 158–67 Gut´es A, Carraro C and Maboudian R 2010 Silver dendrites from galvanic displacement on commercial aluminum foil as an effective SERS substrate J. Am. Chem. Soc. 132 1476–7 Zhu C H, Meng G W, Huang Q and Huang Z L 2012 Vertically aligned Ag nanoplate-assembled film as a sensitive and reproducible SERS substrate for the detection of PCB-77 J. Hazard. Mater. 211/212 389–95 Zhu C H, Meng G W, Huang Q, Li Z B, Huang Z L, Wang M L and Yuan J P 2012 Large-scale well-separated Ag nanosheet-assembled micro-hemispheres modified with HS-β-CD as effective SERS substrates for trace detection of PCBs J. Mater. Chem. 22 2271–8 Pan W X, Zhang D J and Zhan J H 2011 Theoretical investigation on the inclusion of TCDD with β-cyclodextrin by performing QM calculations and MD simulations J. Hazard. Mater. 192 1780–6 Liu P, Zhang D J and Zhan J H 2010 Investigation on the inclusions of PCB52 with cyclodextrins by performing DFT calculations and molecular dynamics simulations J. Phys. Chem. A 114 13122–8
Search
Collections
Journals
About
Contact us
My IOPscience
Ordered arrays of Au-nanobowls loaded with Ag-nanoparticles as effective SERS substrates for rapid detection of PCBs
This content has been downloaded from IOPscience. Please scroll down to see the full text. 2014 Nanotechnology 25 145605 (http://iopscience.iop.org/0957-4484/25/14/145605) View the table of contents for this issue, or go to the journal homepage for more
Download details: IP Address: 128.103.149.52 This content was downloaded on 01/07/2014 at 01:54
Please note that terms and conditions apply.
Nanotechnology Nanotechnology 25 (2014) 145605 (8pp)
doi:10.1088/0957-4484/25/14/145605
Ordered arrays of Au-nanobowls loaded with Ag-nanoparticles as effective SERS substrates for rapid detection of PCBs Bensong Chen1 , Guowen Meng1,2 , Fei Zhou1 , Qing Huang3 , Chuhong Zhu1 , Xiaoye Hu1 and Mingguang Kong1 1
Key Laboratory of Materials Physics, and Anhui Key Laboratory of Nanomaterials and Nanostructures, Institute of Solid State Physics, Chinese Academy of Sciences, Hefei, Anhui 230031, People’s Republic of China 2 University of Science and Technology of China, Hefei, Anhui 230026, People’s Republic of China 3 Key Laboratory of Ion Beam Bioengineering, Institute of Technical Biology and Agriculture Engineering, Hefei Institutes of Physical Sciences, Chinese Academy of Sciences, Hefei, Anhui 230031, People’s Republic of China E-mail: [email protected] Received 21 October 2013, revised 19 January 2014 Accepted for publication 4 February 2014 Published 14 March 2014
Abstract
Large-scale hexagonally close-packed arrays of Au-nanobowls (Au-NBs) with tens of Ag-nanoparticles (Ag-NPs) dispersed in each bowl (denoted as Ag-NPs@Au-NB arrays) are achieved and utilized as effective surface-enhanced Raman scattering (SERS) substrates. The field enhancement benefiting from the special particle-in-cavity geometrical structure as well as the high density of SERS hot spots located in the sub-10 nm gaps between adjacent Ag-NPs and at the particle–cavity junctions all together contribute to the high SERS activity of the Ag-NPs@Au-NB arrays; meanwhile the ordered morphological features of the Ag-NPs@Au-NB arrays guarantee uniformity and reproducibility of the SERS signals. By modifying the Ag-NPs@Au-NB arrays with mono-6-thio-β-cyclodextrin, the SERS detection sensitivity to 3,30 ,4,40 -tetrachlorobiphenyl (PCB-77, one congener of polychlorinated biphenyls (PCBs, kinds of persistent organic pollutants which represent a global environmental hazard)) can be further improved and a low concentration down to 5 × 10−7 M can still be examined, showing promising potential for application in rapid detection of trace-level PCBs in the environment. Keywords: Ag-NPs@Au-NB arrays, surface-enhanced Raman scattering, mono-6-thio-β-cyclodextrin, polychlorinated biphenyls S Online supplementary data available from stacks.iop.org/Nano/25/145605/mmedia (Some figures may appear in colour only in the online journal)
1. Introduction
SERS enhancement is to obtain highly active SERS hot spots (sub-10 nm gaps between noble metallic nanostructures) [6– 11]. In addition, uniformity and reproducibility of SERS signals are also critical for SERS applications, and are strongly dependent on the homogeneity of the structure and the morphology of the SERS substrate. Thus, it is highly desired to construct SERS substrates with not only sufficient
Surface-enhanced Raman scattering (SERS) is considered as one of the most powerful tools for detecting trace-level molecules, owing to its ultra-high detection sensitivity, realtime response and fingerprint effect [1–5]. Experimental and theoretical studies demonstrate that the key to acquiring strong 0957-4484/14/145605+08$33.00
1
c 2014 IOP Publishing Ltd
Printed in the UK
Nanotechnology 25 (2014) 145605
B Chen et al
highly active SERS hot spots, but also pre-controlled ordered structures. It has been demonstrated that the nanosphere-based approach [12] and the electron beam lithography approach [13, 14] are very effective to produce highly active and reproducible SERS substrates. However, these methods usually require complicated synthesis procedures and expensive equipment. To date, many other efforts have also been made to exploit various SERS substrates with appropriate nanostructured architectures, such as one-dimensional (1D) metallic nanoparticle chains [15], two-dimensional (2D) close-packed metallic nanoparticle arrays [16] and three-dimensional (3D) ordered composite micro- and nanoarchitectures [17–19], to achieve high SERS sensitivity and reproducible SERS signals. Among them, a kind of ‘particle-in-cavity’ structure has been theoretically predicted to produce extremely strong field enhancement resulting from cascaded focusing of large optical cross-sections into small gaps [20–22]. Recently, ordered arrays of Ag-nanoparticles in Au-nanobowls (AgNP-in-Au-NB) based on particle-in-cavity structure have been experimentally realized and show good signal reproducibility and excellent SERS performance, arising from their ordered structure and strong field enhancement of the highly active SERS hot spots mainly located at the particle–cavity junctions [23]. However, the fabrication process for the Au-nanobowls of the Ag-NP-in-Au-NB arrays is based on multiple steps, which include (1) self-assembly of close-packed monolayer polystyrene (PS) nanosphere arrays, (2) oxygen plasma etching to decrease the diameter of the PS nanospheres, (3) atomic layer deposition of a sacrificial ZnO layer on the PS nanospheres, (4) thermal annealing to remove the PS, (5) deposition of Au film on the ZnO, (6) etching of the ZnO by HCl solution, and (7) transformation into Au-nanobowl arrays. In addition, there is only one Ag-nanoparticle (Ag-NP) located on the bottom of each Au-nanobowl (Au-NB) in the Ag-NP-in-Au-NB arrays and highly sensitive SERS hot spots are merely formed at the particle–cavity junctions between the Ag-NPs and Au-NBs, leading to a relatively low density of SERS hot spots. Herein, we attempted to synthesize effective SERS substrates based on the particle-in-cavity structure with the following considerations. Firstly, since several convenient synthesis methods have been reported to construct ordered concave arrays [24, 25], the complicated multiple-step fabrication process can be replaced with a simple porous anodic aluminum oxide (AAO) template assisted approach to produce large-scale highly ordered nanobowl arrays. Secondly, by assembling tens of Ag-NPs rather than only one Ag-NP on the concave surface of each Au-NB, a higher density of SERS hot spots can be easily formed in the sub-10 nm gaps between neighboring Ag-NPs as well as at the particle–cavity junctions between the Ag-NPs and the Au-NB. With these two considerations, we fabricated large-scale hexagonal-packed arrays of AuNBs, with each bowl having tens of Ag-NPs with sub-10 nm gaps on its curvature surface (denoted as Ag-NPs@Au-NB), via simple successive sputtering of Au film and Ag-NPs on concave arrays on Al foil achieved by using an AAO template. Due to the high density of sensitive SERS hot spots located between the neighboring Ag-NPs and at the
Scheme 1. Schematic for the fabrication of Ag-NPs@Au-NB
arrays.
particle–cavity junctions, the morphological homogeneity, and the field enhancement benefiting from the particle-in-cavity structure, the Ag-NPs@Au-NB arrays can serve as effective SERS substrates with high SERS activity and reproducible signals. Finally, we tried to graft mon-6-thio-β-cyclodextrin (HS-β-CD) on the Ag-NPs@Au-NB arrays for SERS-based detection of polychlorinated biphenyls (PCBs), which are highly toxic persistent organic pollutants as defined in the Stockholm Convention and cause great threats to the global environment and human health [26]. By virtue of the HSβ-CD chemical modification, SERS detection of 3,30 ,4,40 tetrachlorobiphenyl (PCB-77, one congener of PCBs) at a concentration of 5 × 10−7 M has been achieved, demonstrating that the as-fabricated Ag-NPs@Au-NB arrays modified with HS-β-CD have great potential as effective SERS substrates for rapid detection of trace PCBs. 2. Experimental methods 2.1. Preparation of the Ag-NPs@Au-NB arrays
The synthesis procedure for the Ag-NPs@Au-NB arrays on Al foil (schematically shown in scheme 1) can be divided into the following steps: (1) Al foil is anodized in a twostep anodization [27] to obtain a highly ordered porous AAO membrane on the Al foil (anodized at 40 V in oxalic acid at 10 ◦ C, with a pore diameter of about 70–80 nm); (2) the topmost ordered porous AAO membrane is thoroughly removed in 6 wt% phosphoric and 1.8 wt% chromic acid solution (at 50 ◦ C, for 8 h), leaving large-area highly ordered hexagonal-packed hemispherical nanoconcave arrays on the top of the remaining Al foil; (3) a uniform thin Au layer (∼10 nm) is deposited onto the cavity surface of the concave arrays to form ordered Au-NB arrays on the Al foil; (4) Ag-NPs are ion-sputtered onto each of the Au-NBs to achieve largearea (up to cm2 ) ordered hexagonal-packed Ag-NPs@Au-NB arrays on the remaining Al foil. The sputtering of Au film on the concave arrays on the remaining Al foil was carried out in a homemade electron beam evaporation system (DZS-500, Shenyang Scientific Instrument Co., Ltd) with a film thickness monitor (FTM-V). The deposition of Ag-NPs was carried out in an ion-sputtering coater (Emitech, K550X) at a rate of 4 nm min−1 (adjusting the duration from 2 to 10 min) while keeping the sputtering stage stationary. 2
Nanotechnology 25 (2014) 145605
B Chen et al
The Raman spectra were recorded using a 532 nm laser and a 20× short focal length objective in a confocal microprobe Raman system (Renishaw, inVia). The laser spot focused on the sample surface was about 5 µm in diameter. The integration duration for SERS spectral recording was chosen as 10 and 30 s for R6G and PCB-77, while the laser power was 1 and 5 mW for R6G and PCB-77, respectively. 2.3. Modification of the Ag-NPs@Au-NB arrays with HS-β -CD
As-prepared small pieces of Ag-NPs@Au-NB arrays were immersed in HS-β-CD dissolved in N ,N -dimethylformamide (DMF) solution (0.5 ml, 0.1 mM) for 5 h for chemical modification. After the modification treatment was finished, the small pieces of SERS substrate were taken out, rinsed with DMF, and finally dried in high-purity nitrogen flow. Figure 1. Top-view SEM images of (a) concave arrays, (b) Au-NB arrays, (c) Ag-NPs@Au-NB arrays on the remaining Al foil. (d) Enlarged SEM image of the Ag-NPs@Au-NB arrays. The insets in (c) and (d) are oblique and close-up views of the Ag-NPs@Au-NB arrays, respectively.
3. Results and discussion 3.1. Characterization of the Ag-NPs@Au-NB arrays
Figure 1(a) is a representative top-view SEM image of the hemispherical concave arrays on the remaining Al foil after thorough removal of the topmost ordered porous AAO membrane, revealing the periodic hexagonal-packed arrangement of geometrical features which is an exact replica of the morphology of the AAO-nanopore bottoms. Figure 1(b) shows the top-view SEM image of the sputtered thin Au film on the concave arrays, demonstrating that the thin Au film is composed of close-packed tiny Au-nanoparticles and uniformly covers the entire surface of the concave arrays. Thus, it is feasible to treat the sputtered Au thin film as dense and continuous Au-NB arrays. The top-view SEM image of Ag-NPs@Au-NB arrays (figure 1(c)) reveals that a large number of Ag-NPs are evenly dispersed on the cavities and the edges of the apertures of the Au-NBs. The enlarged view (figure 1(d)) further displays that the average diameter of a nanobowl unit and the average inter-gap spacing of adjacent Ag-NPs are about 100 and 5 nm, respectively. The components of the Au-NB arrays and Ag-NPs@Au-NB arrays supported on Al foil are confirmed by energy-dispersive x-ray spectroscopy (EDS) measurements (figure S1, supporting information available at stacks.iop.org/Nano/25/145605/mmedia).
For a comparison of SERS detection sensitivity between the Ag-NPs@Au-NB arrays and Ag-NPs dispersed on flat polished Al foil with Au film (denoted as Ag-NPs@Au), the polished Al was prepared by electropolishing Al foil in a mixture of perchloric acid and absolute ethanol (at a ratio of 1:9 in volume) at 1 ◦ C for 3 min with a voltage of 20 V. 2.2. Characterizations
The morphology and components of the resultant samples were characterized by using scanning electron microscopy (SEM, Sirion 200, FEI, at 5 kV) with energy-dispersive x-ray spectroscopy (EDS, Oxford). For the SERS measurements, the as-prepared Ag-NPs@ Au-NB arrays were cut into small square pieces (5 mm × 5 mm) and immersed in Rhodamine 6G (R6G) aqueous solutions (0.25 ml) of different concentrations (from 10−6 to 10−10 M) for 20 min, then taken out and rinsed with deionized water, and finally dried with high-purity flowing nitrogen before Raman spectral examination. For the comparison of SERS activity between the AgNPs@Au-NB arrays and Ag-NPs@Au, 1 µl of 1 × 10−7 M R6G aqueous solution was respectively dropped and uniformly dispersed onto pieces of Ag-NPs@Au-NB array and AgNPs@Au with the same area (5 mm × 5 mm). Then, the samples were carefully dried in flowing nitrogen atmosphere under room temperature. Raman measurements for these substrates were carried out under identical experimental conditions (such as laser wavelength and power, microscope objective/lenses, integration duration, etc). To check the SERS sensitivity of the Ag-NPs@Au-NB arrays for PCB-77, PCB-77 was firstly dissolved in acetone to form solutions and then small pieces of SERS substrate were respectively immersed in PCB-77 solutions (0.25 ml) of different concentrations (10−4 , 10−5 , 5 × 10−6 M) for 5 h, taken out and cleaned with acetone, and finally dried with high-purity flowing nitrogen in a laboratory fume cupboard before SERS measurements.
3.2. Modulation and characterization of the SERS performance of the Ag-NPs@Au-NB arrays
Since both theoretical and experimental studies demonstrate that control of SERS hot spots within the sub-10 nm regime is essential to obtain high SERS activity [28, 29], we tailored the diameter and the inter-particle spacing of the Ag-NPs in the Ag-NPs@Au-NB arrays by adjusting the Ag-sputtering conditions (sputtering rate and duration) so as to acquire the optimal SERS activity. Figure 2(a) shows the SERS spectra of 10−7 M R6G aqueous solution dispersed on Ag-NPs@AuNB arrays with different Ag-sputtering durations from 2 to 10 min with a given sputtering rate. The SERS activity of the substrates improves with increase of the Ag-sputtering duration from 2 to 6 min, but gradually decreases with 3
Nanotechnology 25 (2014) 145605
B Chen et al
and diameter and gap distance distributions in figures S3(c) and (d), supporting information available at stacks.iop.org/ Nano/25/145605/mmedia), which is the optimal regime for forming highly active SERS hot spots. With further elongation of the Ag-sputtering duration, larger Ag-NPs agglomerate together and even form a continuous Ag-NPs film over the cavity surfaces of the Au-NB arrays (figure S2(c), supporting information available at stacks.iop.org/Nano/25/145605/mme dia), leading to narrower or even disappearance of the gaps between the neighboring Ag-NPs, so that the SERS activity deteriorates. To further investigate the SERS performance of the Ag-NPs@Au-NB arrays with 6 min Ag-sputtering, we tested their SERS performance using R6G aqueous solutions of different concentrations (from 10−6 to 10−10 M). As shown in figure 2(b), distinct bands can still be easily distinguished in the Raman spectrum even at the low concentration of 10−10 M, and the peaks at 611 and 774 cm−1 are assigned to the C–C–C in-plane and out-of-plane bending vibrations, respectively [29, 30]. The average SERS enhancement factor (EF) of the Ag-NPs@Au-NB arrays with 6 min Ag-sputtering was estimated by using 4-aminothiophenol (4-ATP) as test molecules and the EF value was estimated to be about 3.7 × 104 at the 1075 cm−1 band of 4-ATP (details are given in part S2, supporting information available at stacks.iop.org/ Nano/25/145605/mmedia). To quantitatively estimate the electromagnetic field distribution of our Ag-NPs@Au-NB arrays with optimized SERS performance (with 6 min Ag-sputtering), a 3D finitedifference-time-domain (FDTD) simulation was implemented to simulate the 3D spatial electric field profile of the AgNPs@Au-NB arrays (the mesh size in our simulation was 1 nm). A single Ag-NPs@Au-NB unit is illuminated from the top with a 532 nm linear polarized plane wave, as schematically depicted in figure 3(a). For simplicity, the cavity of the Au-NB is considered as a hemispherical shape with the average diameter and thickness being chosen as 100 and 10 nm, respectively. The Ag-NPs sputtered on the cavity of the Au-NB are treated as having spherical shape and the diameter of the Ag-NPs and the inter-gap spacing of the Ag-NPs are set as 17 nm and 5 nm, respectively. The Ag-NPs located at the edge of the aperture of the Au-NB are slightly larger and their diameter is chosen as 24 nm (see the enlarged SEM image in figure 1(d)). Figures 3(b) and (c) display the vertical and horizontal cross-sectional views of the aperture of the Au-NB from the FDTD simulated electric field profile of the Ag-NPs@Au-NB arrays, respectively. It can be seen that strong field enhancement mainly occurs in three kinds of positions: the sub-10 nm gaps located between neighboring Ag-NPs on the cavity surface of the Au-NB (position I, marked in figure 3(b)) and the edge of the aperture of the Au-NB (position II, marked in figure 3(c)) as well as the particle–cavity junction area between the Ag-NPs and the Au-NB (position III, marked in figure 3(b)). The maximum field enhancement values |E/E 0 | for the vertical and horizontal cross-sectional space are respectively 13.15 and 15.05, which can be attributed to the fact that the Ag-NPs located on the platform of the edge of the aperture of the Au-NB are larger than the Ag-NPs on the cavity of the Au-NB and can produce stronger field enhancement in the inter-particle spacing.
Figure 2. (a) SERS spectra of 10−7 M R6G adsorbed on the
Ag-NPs@Au-NB arrays with different Ag-sputtering durations (2, 4, 6, 8, and 10 min). (b) SERS spectra collected on the Ag-NPs@Au-NB arrays (with 6 min Ag-sputtering) exposed to different concentrations of R6G aqueous solution (10−6 , 10−7 , 10−8 , 10−9 and 10−10 M).
longer Ag-sputtering durations. The morphological evolution of the Ag-NPs@Au-NB arrays associated with the different Ag-sputtering durations can serve to explain this phenomenon. For the substrate with 2 min Ag-sputtering, only small Ag-NPs with an average diameter of ca. 10 nm and average gap distance of ca. 15 nm (figure S2(a), supporting information available at stacks.iop.org/Nano/25/145605/mmedia) are dispersed on the cavity surface and the edge of the aperture of the Au-NB arrays. The corresponding diameter and inter-particle gap distance distributions of the 2 min sputtered Ag-NPs are shown in figures S3(a) and (b) of the supporting information (available at stacks.iop.org/Nano/25/145605/mmedia). The relatively small diameter and large gap distance of the sputtered Ag-NPs are not beneficial to inducing strong field enhancement between adjacent Ag-NPs as well as the junctions of Ag-NPs and the Au-NB [23], resulting in poor SERS activity. By prolonging the Ag-sputtering duration to 6 min, the deposited Ag-NPs become larger (about 17 nm in average diameter) and large quantities of sub-10 nm inter-particle gaps and particle– cavity junctions are formed (SEM image in figure S2(b) 4
Nanotechnology 25 (2014) 145605
B Chen et al
Figure 3. 3D FDTD simulations of Ag-NPs@Au-NB arrays. (a) The 3D simulation model. (b) The vertical and (c) the horizontal cross-sectional view of the aperture of the Au-NB from the FDTD simulated electric field profile. The maximum field enhancement |E/E 0 | values for the vertical and horizontal cross-sectional space are 13.15 and 15.05, respectively.
p.org/Nano/25/145605/mmedia). Figures 4(b) and (c) show the representative morphology of Ag-NPs@Au-NB arrays and Ag-NPs@Au with their corresponding optimal SERS performance, respectively. When exposed to 10−7 M R6G aqueous solution under identical experimental conditions, the Ag-NPs@Au-NB arrays show higher SERS activity than that of Ag-NPs@Au under their corresponding optimized sputtering conditions (figure 4(d)). In addition, the average EF of the Ag-NPs@Au substrate (with optimal SERS property) was estimated to be about 2.5 × 104 at the 1075 cm−1 band of 4-ATP (details are shown in part S2, supporting information available at stacks.iop.org/Nano/25/145605/mmedia), and this value is smaller than that for the Ag-NP@Au-NB arrays (with optimized 6 min Ag-sputtering), demonstrating that the Ag-NP@Au exhibits weaker average field enhancement. To further understand the benefit of the particle-in-cavity geometrical structure of the Ag-NPs@Au-NB arrays to the field enhancement, we theoretically simulated and compared the field enhancement of Ag-NPs@Au-NB arrays and that of Ag-NPs@Au. The diameter of the Ag-NPs and the interparticle spacing used in the simulations for Ag-NPs@Au are all the same as those in the simulation of Ag-NPs@Au-NB arrays with 6 min Ag-sputtering. The simulated FDTD electric field profiles of the Ag-NPs@Au (detailed simulated results are shown in figure S8, supporting information available at stacks.iop.org/Nano/25/145605/mmedia) display that the SERS hot spots are mainly located in the inter-particle gaps and the junctions between Ag-NPs and the flat Au surface, showing weaker field enhancement at the SERS hot spots of Ag-NPs@Au compared to that of Ag-NPs@Au-NB arrays. It should be noted that there is a small deviation between the simulated and experimental comparisons of SERS activity between the Ag-NPs@Au-NB arrays and Ag-NPs@Au, which we believe comes from the assumptions and approximations made in the calculations. Nevertheless, the simulated results clearly illustrate that a higher field enhancement effect can be produced by assembling Ag-NPs on the Au cavity surface as compared to putting similar Ag-NPs on a flat Au surface with the same inter-particle spacing.
To test the uniformity and reproducibility of the AgNPs@Au-NB arrays as effective SERS substrates, SERS spectra from ten randomly selected positions across the whole substrate were recorded under identical experimental conditions, including the same laser power and integration time (figure S4, supporting information available at stacks.iop.org /Nano/25/145605/mmedia). The results show that the relative standard deviation (RSD) of the SERS intensity for the Raman peak at 611 cm−1 of 10−7 M R6G is about 8%, confirming the good uniformity of the measured SERS signals. The good reproducibility of the signals can also be demonstrated in the SERS measurement of R6G at different concentrations (shown in figure S5, supporting information available at stacks.io p.org/Nano/25/145605/mmedia). The excellent SERS signal uniformity and reproducibility of the Ag-NPs@Au-NB arrays derives from the fact that their morphology well replicates the morphological features of the native highly ordered porous AAO template. 3.3. Benefit of the particle-in-cavity geometrical structure of the Ag-NPs@Au-NB arrays for field enhancement
Theoretical and experimental studies reveal that metallic nanovoid [31–33] and particle-in-cavity architectures [20, 21, 34] can be used for plasmonic engineering and production of strong field enhancement. In order to demonstrate the benefit of the particle-in-cavity geometrical structure of the AgNPs@Au-NB arrays to field enhancement, we compared the SERS activity between Ag-NPs@Au-NB arrays and Ag-NPs sputtered on a flat polished Al sheet with a thin Au film (denoted as Ag-NPs@Au, with the same 10 nm Au film thickness as that of the Au-NBs in the Ag-NPs@Au-NB arrays), as shown in the schematic architectures in figure 4(a). In a similar way to how we optimized the SERS activity of the Ag-NPs@Au-NB arrays, we tailored and optimized the SERS performance of the Ag-NPs@Au by adjusting the Agsputtering duration (the details of the morphological evolution and variation of SERS performance of the Ag-NPs@Au with different Ag-sputtering durations are shown in figures S6 and S7 respectively, supporting information available at stacks.io 5
Nanotechnology 25 (2014) 145605
B Chen et al
Figure 4. (a) Schematic drawings of Ag-NPs@Au-NB arrays (substrate I) and Ag-NPs@Au (substrate II). Top-view SEM images of (b) substrate I and (c) substrate II with the corresponding optimized SERS performance. (d) The SERS spectra collected on substrate I (curve I) and substrate II (curve II) (with the corresponding optimized SERS performance) exposed to 10−7 M R6G aqueous solution.
of the main peak bands of PCB-77 versus concentration are shown in figure S9, supporting information (available at stacks.iop.org/Nano/25/145605/mmedia). The reason why the Ag-NPs@Au-NB arrays as SERS substrates exhibit higher SERS sensitivity to R6G than PCBs might originate from the fact that the bare surface of the metallic building blocks has a relatively poor ability to adsorb the hydrophobic PCB molecules. Surface modification of an SERS substrate is regarded as a useful solution to capture target molecules onto the hot spots of the SERS substrate and improve the SERS detection sensitivity. Since theoretical simulations have demonstrated that a PCB molecule can enter the cavity of β-cyclodextrin (β-CD) to form an inclusion complex with β-CD [40, 41] and strong chemical bonding can be formed between the mercapto functional group (HS-) of organic molecules and the metallic Ag surface, we chose HS-β-CD as a surface modifying agent and functionalized the as-prepared Ag-NPs@Au-NB arrays with HS-β-CD to capture more PCB-77 molecules onto the hot spots so as to improve the SERS sensitivity to PCB-77. Figure 5(b) displays the SERS spectra of Ag-NPs@Au-NB arrays modified with HS-β-CD for detection of PCB-77 solutions of 5 × 10−6 (curve IV) and 5 × 10−7 M (curve V), respectively. Comparing the measured SERS signals of curve IV with those of curve III in figure 5(a), it can be seen that the peak intensities of PCB-77 are increased nearly two-fold after HS-β-CD modification, verifying that modification of the Ag-NPs@Au-NB arrays with HS-β-CD can indeed enhance their ability to capture the target PCB molecules. The phenomenon of no pronounced bands of HS-β-CD emerging in the SERS spectra of Ag-NPs@Au-NB arrays modified with HS-β-CD for detection of PCB-77 may be due to the fact that the intrinsic Raman intensity of HS-β-CD is much weaker than that of PCB-77 (figure S11
The above comparisons of the measured SERS activity and the simulated field enhancement unambiguously confirm the benefit of the particle-in-cavity nanostructured Ag-NPs@Au-NB arrays, and the improvement of the SERS signals can be explained as follows. Firstly, the void-like AuNBs of the Ag-NPs@Au-NB arrays serve as light harvesters and focus the light in the cavity. The introduced Ag-NPs in the cavity of the Au-NB can strengthen the particle–cavity coupling effect when the nanoparticle mode is resonant with one of the cavity modes [20]. Secondly, in comparison to the junction between Ag-NPs and a flat Au surface, the junction of Ag-NPs and the negative curvature cavity of the Au-NB favors stronger excitation of field enhancement under a vertically incident laser. Finally, a higher density of SERS hot spots can be generated in the particle-in-cavity structure compared to that of particles dispersed on a flat surface under the same horizontal projected areas. 3.4. SERS sensitivity of the Ag-NPs@Au-NB arrays to PCBs
Finally, the as-prepared Ag-NPs@Au-NB arrays were examined as effective SERS substrates for the detection of PCB-77, one of 209 congeners of PCBs. Figure 5(a) displays the SERS spectra of PCB-77 at different concentrations of 10−4 (curve I), 10−5 (curve II) and 5 × 10−6 M (curve III) on the as-prepared bare Ag-NPs@Au-NB arrays. There are five distinct bands at 674 cm−1 (C–Cl stretching) [35, 36], 1030 cm−1 (ring breathing) [36, 37], 1244 cm−1 (C–H wagging) [35, 36], 1297 cm−1 (biphenyl C–C bridge stretching) [35, 36], and 1598 cm−1 (ring stretching) [36, 37] in the SERS spectra, which agree well with the normal Raman spectrum of PCB-77 [38, 39]. In the SERS spectra, it can be seen that the representative peaks still exist even when the concentration of measured PCB-77 solution decreases down to 5 × 10−6 M by using the bare AgNPs@Au-NB arrays as SERS substrates. The SERS intensities 6
Nanotechnology 25 (2014) 145605
B Chen et al
Ag-NPs and at the particle–cavity junctions as well as the particle-in-cavity geometrical structure. The morphological uniformity and homogeneity also ensure reproducibility of the SERS detection signals of the Ag-NPs@Au-NB arrays. Furthermore, by modifying the as-prepared Ag-NPs@AuNB arrays with HS-β-CD, the SERS substrate can capture hydrophobic molecules such as PCBs. The detected PCB-77 concentration can be as low as 5 × 10−7 M. Therefore, our work demonstrates that the as-fabricated Ag-NPs@Au-NB arrays modified with HS-β-CD have promising potential in SERS-based rapid detection of organic environmental pollutants such as PCBs. Acknowledgments
This work was supported by the National Key Basic Research Program of China (Grant No. 2013CB934304), the National Natural Science Foundation of China (Grant Nos 21107113 and 11274312), and the Postdoctoral Science Foundation of China (Grant No. 2012M510164). References [1] Fleischm M, Hendra P J and McQuilla A J 1974 Raman-spectra of pyridine adsorbed at a silver electrode Chem. Phys. Lett. 26 163–6 [2] Jeanmaire D L and Vanduyne R P 1977 Surface Raman spectroelectrochemistry: I. Heterocyclic, aromatic, and aliphatic-amines adsorbed on the anodized silver electrode J. Electroanal. Chem. 84 1–20 [3] Albrecht M G and Creighton J A 1977 Anomalously intense Raman-spectra of pyridine at a silver electrode J. Am. Chem. Soc. 99 5215–7 [4] Kneipp K, Wang Y, Kneipp H, Perelman L T, Itzkan I, Dasari R and Feld M S 1997 Single molecule detection using surface-enhanced Raman scattering (SERS) Phys. Rev. Lett. 78 1667–70 [5] Li J F et al 2010 Shell-isolated nanoparticle-enhanced Raman spectroscopy Nature 464 392–5 [6] Qin L D, Zou S L, Xue C, Atkinson A, Schatz G C and Mirkin C A 2006 Designing, fabricating, and imaging Raman hot spots Proc. Natl Acad. Sci. USA 103 13300–3 [7] Camden J P, Dieringer J A, Wang Y M, Masiello D J, Marks L D, Schatz G C and Van Duyne R P 2008 Probing the structure of single-molecule surface-enhanced Raman scattering hot spots J. Am. Chem. Soc. 130 12616–7 [8] Lee S J, Morrill A R and Moskovits M 2006 Hot spots in silver nanowire bundles for surface-enhanced Raman spectroscopy J. Am. Chem. Soc. 128 2200–1 [9] Braun G, Pavel I, Morrill A R, Seferos D S, Bazan G C, Reich N O and Moskovits M 2007 Chemically patterned microspheres for controlled nanoparticle assembly in the construction of SERS hot spots J. Am. Chem. Soc. 129 7760–1 [10] Li W Y, Camargo P H C, Lu X M and Xia Y N 2009 Dimers of silver nanospheres: facile synthesis and their use as hot spots for surface-enhanced Raman scattering Nano Lett. 9 485–90 [11] Camargo P H C, Rycenga M, Au L and Xia Y N 2009 Isolating and probing the hot spot formed between two silver nanocubes Angew. Chem. Int. Edn 48 2180–4
Figure 5. (a) SERS spectra of PCB-77 collected on bare Ag-NPs@Au-NB arrays (6 min Ag-sputtering) exposed to 10−4 (curve I), 10−5 (curve II) and 5 × 10−6 (curve III) M PCB-77, respectively. The inset is the schematic of PCB-77. (b) SERS spectra of 5 × 10−6 (curve IV) and 5 × 10−7 (curve V) M PCB-77 from HS-β-CD-modified Ag-NPs@Au-NB arrays. The inset displays the schematic diagram of the inclusion complex of PCB-77 with HS-β-CD located in the gap between two neighboring Ag-NPs.
in part S3, supporting information available at stacks.iop.org /Nano/25/145605/mmedia). When the PCB-77 concentration is decreased down to 5 × 10−7 M, the representative bands of PCB-77 in the SERS spectra (curve V in figure 5(b)) can still be distinguished, though the peaks intensities decrease sharply. Therefore, this experiment confirms that the surface modification with HS-β-CD can improve the sensitivity for detecting PCBs such as PCB-77. 4. Conclusion
In conclusion, large-scale ordered Ag-NPs@Au-NB arrays as effective SERS substrates have been achieved via a porous AAO template assisted approach with a physical deposition route. By tailoring the inter-particle gaps of neighboring Ag-NPs and the diameters of the Ag-NPs dispersed in the Au-NBs, high SERS activity of the Ag-NPs@Au-NB arrays has been acquired, which benefits from the high density of sub-10 nm SERS hot spots formed between adjacent 7
Nanotechnology 25 (2014) 145605
B Chen et al
[12] Chan G H, Zhao J, Hicks E M, Schatz G C and Van Duyne R P 2007 Plasmonic properties of copper nanoparticles fabricated by nanosphere lithography Nano Lett. 7 1947–52 [13] Margueritat J et al 2011 Influence of the number of nanoparticles on the enhancement properties of surface-enhanced Raman scattering active area: sensitivity versus repeatability ACS Nano 5 1630–8 [14] Billot L, De la Chapelle M L, Grimault A S, Vial A, Barchiesi D, Bijeon J L, Adam P M and Royer P 2006 Surface enhanced Raman scattering on gold nanowire arrays: evidence of strong multipolar surface plasmon resonance enhancement Chem. Phys. Lett. 224 303–7 [15] Yang Y, Shi J L, Tanaka T and Nogami M 2007 Self-assembled silver nanochains for surface-enhanced Raman scattering Langmuir 23 12042–7 [16] Wei A, Kim B, Sadtler B and Tripp S L 2001 Tunable surface-enhanced Raman scattering from large gold nanoparticle arrays ChemPhysChem 2 743–5 [17] Huang Z L, Meng G W, Huang Q, Yang Y J, Zhu C H and Tang C L 2010 Improved SERS performance from Au nanopillar arrays by abridging the pillar tip spacing by Ag sputtering Adv. Mater. 22 4136–9 [18] Tang H B, Meng G W, Huang Q, Zhang Z, Huang Z L and Zhu C H 2012 Arrays of cone-shaped ZnO nanorods decorated with Ag nanoparticles as 3D surface-enhanced Raman scattering substrates for rapid detection of trace polychlorinated biphenyls Adv. Funct. Mater. 22 218–24 [19] Liu G Q, Cai W P, Kong L C, Duan G T, Li Y, Wang J J, Zuo G M and Cheng Z X 2012 Standing Ag nanoplate-built hollow microsphere arrays: controllable structural parameters and strong SERS performances J. Mater. Chem. 22 3177–84 [20] Huang F M, Wilding D, Speed J D, Russell A E, Bartlett P N and Baumberg J J 2011 Dressing plasmons in particle-in-cavity architectures Nano Lett. 11 1221–6 [21] Ye J, Chen C, Lagae L, Maes G, Borghs G and Van Dorpe P 2010 Strong location dependent surface enhanced Raman scattering on individual gold semishell and nanobowl particles Phys. Chem. Chem. Phys. 12 11222–4 [22] Ye J, Lagae L, Maes G and Van Dorpe P 2011 A versatile method to fabricate particle-in-cavity plasmonic nanostructures J. Mater. Chem. 21 14394–7 [23] Li X L, Zhang Y Z, Shen Z X and Fan H J 2012 Highly ordered arrays of particle-in-bowl plasmonic nanostructures for surface-enhanced Raman scattering Small 8 2548–54 [24] Lee W and Lee J K 2002 Non-lithographic approach to the fabrication of polymeric nanostructures with a close-packed 2D hexagonal array Adv. Mater. 14 1187–90 [25] Liang F X, Cheng J W, Tsang C K, Cheng H and Li Y Y 2011 TiO2 -based nano-wells for fabricating nanocrystals J. Nanosci. Nanotechnol. 11 11059–63 [26] Gambaro A, Manodori L, Zangrando R, Cincinelli A, Capodaglio G and Cescon P 2005 Atmospheric PCB
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
8
concentrations at Terra Nova Bay, Antarctica Environ. Sci. Technol. 39 9406–11 Masuda H and Fukuda K 1995 Ordered matel nanohole arrays made by a 2-step replication of honeycomb structures of anodic alumina Science 268 1466–8 Wang H H, Liu C Y, Wu S B, Liu N W, Peng C Y, Chan T H, Hsu C F, Wang J K and Wang Y L 2006 Highly Raman-enhancing substrates based on silver nanoparticle arrays with tunable sub-10 nm gaps Adv. Mater. 18 491–5 Wang H, Levin C S and Halas N J 2005 Nanosphere arrays with controlled sub-10-nm gaps as surface-enhanced Raman spectroscopy substrates J. Am. Chem. Soc. 127 14992–3 Hildebrandt P and Stockburger M 1984 Surface-enhanced resonance Raman-spectroscopy of rhodamine-6G adsorbed on colloidal silver J. Phys. Chem. 88 5935–44 Cole R M, Baumberg J J, Garcia de Abajo F J, Mahajan S, Abdelsalam M and Bartlett P N 2007 Understanding plasmons in nanoscale voids Nano Lett. 7 2094–100 Kelf T A, Sugawara Y, Cole R M and Baumberg J J 2006 Localized and delocalized plasmons in metallic nanovoids Phys. Rev. B 74 245415 Ye J, Van Dorpe P, Van Roy W, Borghs G and Maes G 2009 Fabrication, characterization, and optical properties of gold nanobowl submonolayer structures Langmuir 25 1822–7 Ross B M and Lee L P 2009 Creating high density nanoantenna arrays via plasmon enhanced particle-cavity (PEP-C) architectures Opt. Express 17 6860–6 Bantz K C and Haynes C L 2009 Surface-enhanced Raman scattering detection and discrimination of polychlorinated biphenyls Vib. Spectrosc. 50 29–35 Socrates G 2001 Infrared and Raman Characteristic Group Frequencies: Tables and Charts 3rd edn (New York: Wiley) pp 158–67 Gut´es A, Carraro C and Maboudian R 2010 Silver dendrites from galvanic displacement on commercial aluminum foil as an effective SERS substrate J. Am. Chem. Soc. 132 1476–7 Zhu C H, Meng G W, Huang Q and Huang Z L 2012 Vertically aligned Ag nanoplate-assembled film as a sensitive and reproducible SERS substrate for the detection of PCB-77 J. Hazard. Mater. 211/212 389–95 Zhu C H, Meng G W, Huang Q, Li Z B, Huang Z L, Wang M L and Yuan J P 2012 Large-scale well-separated Ag nanosheet-assembled micro-hemispheres modified with HS-β-CD as effective SERS substrates for trace detection of PCBs J. Mater. Chem. 22 2271–8 Pan W X, Zhang D J and Zhan J H 2011 Theoretical investigation on the inclusion of TCDD with β-cyclodextrin by performing QM calculations and MD simulations J. Hazard. Mater. 192 1780–6 Liu P, Zhang D J and Zhan J H 2010 Investigation on the inclusions of PCB52 with cyclodextrins by performing DFT calculations and molecular dynamics simulations J. Phys. Chem. A 114 13122–8
