Facile fabrication of superhydrophobic hybrid nanotip and nanopore arrays as surface-enhanced Raman spectroscopy substrates

Applied Surface Science 443 (2018) 138–144

Contents lists available at ScienceDirect

Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc

Full Length Article

Facile fabrication of superhydrophobic hybrid nanotip and nanopore arrays as surface-enhanced Raman spectroscopy substrates Yuxin Li a, Juan Li a,⇑, Tiankun Wang a, Zhongyue Zhang a, Yu Bai a, Changchun Hao a, Chenchen Feng a, Yingjun Ma a,b, Runguang Sun a,⇑ a b

School of Physics and Information Technology, Shaanxi Normal University, Xi’an 710119, PR China School of Science, Ningxia Medical University, Yinchuan 750004, PR China

a r t i c l e

i n f o

Article history: Received 9 December 2017 Revised 7 February 2018 Accepted 24 February 2018 Available online 27 February 2018 Keywords: Electrochemical anodization Porous anodic alumina Hybrid nanotip and nanopore arrays Surface-enhanced Raman spectroscopy Superhydrophobic

a b s t r a c t We demonstrate the fabrication of superhydrophobic hybrid nanotip and nanopore arrays (NTNPAs) that can act as sensitive surface-enhanced Raman spectroscopy (SERS) substrates. The large-area substrates were fabricated by following a facile, low-cost process consisting of the one-step voltage-variation anodization of Al foil, followed by Ag nanoparticle deposition and fluorosilane (FS) modification. Uniformly distributed, large-area (5  5 cm2) NTNPAs can be obtained rapidly by anodizing Al foil for 1560 s followed by Ag deposition for 400 s, which showed good SERS reproducibility as using1 lM Rhodamine 6G (R6G) as analyte. SERS performances of superhydrophobic NTNPAs with different FS modification and Ag nanoparticle deposition orders were also studied. The nanosamples with FS modification followed by Ag nanoparticle deposition (FS-Ag) showed better SERS sensitivity than the nanosamples with Ag nanoparticle deposition followed by FS modification (Ag-FS). The detection limit of a directly dried R6G droplet can reach 10 8 M on the FS-Ag nanosamples. The results can help create practical high sensitive SERS substrates, which can be used in developing advanced bio- and chemical sensors. Ó 2018 Elsevier B.V. All rights reserved.

1. Introduction Surface-enhanced Raman spectroscopy (SERS) is a powerful spectroscopic analytical technique that is widely utilized in developing advanced bio- and chemical sensors to realize images of living cells [1,2], in vitro and in vivo diagnosis [3], food analysis [4], trace molecule detection [5] and so on. SERS is dominantly dependent upon the intensive electromagnetic field in the vicinity of metal or metal-decorated nanostructures due to their localized surface plasmonic resonance, which can dramatically amplify Raman scattering signals of target molecules absorbed on nanostructures [3]. Thus, for practical applications, one of the key research topics is to develop facile, low-cost methods to construct large-area substrates that are uniformly covered by high density plasmonic nanostructures (i.e. hot spots), which is the prerequisite for simultaneously guaranteeing the sensitivity and reproducibility of SERS signals [5–13]. To date, various nanostructured arrays (e.g. nanocone [11], nanorod with narrow gaps [8] and 3D hybrid nanostructures [5,6]) have been fabricated by using top-down

⇑ Corresponding authors. E-mail addresses: (R. Sun).

[email protected] (J. Li),

https://doi.org/10.1016/j.apsusc.2018.02.247 0169-4332/Ó 2018 Elsevier B.V. All rights reserved.

[email protected]

lithography (e.g. nanosphere-assisted lithography [11] and optical lithography [12]) and template-assisted technology (e.g. nanoimprinting [9] and electrodeposition [6,7]), which have made a great breakthrough in realizing the high sensitivity and reproducibility of SERS signals. However, top-down lithography is timeconsuming and requires sophisticated equipment [10–12]; Meanwhile, template-assisted technology is restricted by available templates, which is always followed by complicated processing to obtain the highly sensitive nanostructures [5–9]. Thus, developing facile, low-cost methods to fabricate uniformly distributed, largearea plasmonic nanostructured arrays remains a challenge. Another issue for SERS is that it needs an efficient way to bring the target molecules to the surface of plasmonic nanostructures [14]. For dilute aqueous solutions, nanostructures always need to be dipped into the solution for more than 8 h before detection to achieve uniform analyte absorption. The diffuse rate of analyzed molecules to the nanosurface is very slow (ca. 60 h/cm) [15]. To solve this problem, scientists have turned to the substrates own both SERS and superhydrophobicity [14–20]. The nanostructure not only can act as hot spots for SERS detection, but also enrich the analyte of a spherical droplet into a small area due to the concentration effect [19]. To simultaneously achieve these two functions above, it needs to skillfully design the nanostructure. To

Y. Li et al. / Applied Surface Science 443 (2018) 138–144

date, various SERS and superhydrophobic structures have been reported, such as nanotip or nanorod arrays [14], hybrid microand nanostructure [16,17], nanoflakes [18] and so on. However, each method still requires significant advancement in their fabrication techniques to realize the practical applications of SERS and superhydrophobic substrates. Porous anodic alumina formed by electrochemical anodization is considered to be a good SERS-active substrate due to its advantages in terms of stability, low cost, operating convenience and technique compatibility [21–24]. However, in traditional anodization, it needs long time (>8 h) to obtain self-ordered nanodents on Al surface [25]. The conventional cylindrical alumina nanopore arrays are not the proper structure either for SERS [23] or superhydrophobic property [26]. In principle, alumina hybrid NTNPAs should realize the both SERS and superhydrophobic functions after deposited with Ag nanoparticle and modified with low-free-energy fluorosilane (FS). First, the nanotips can excite intensive electromagnetic field and avoid the capillary effect due to the interrupted cell wall [23,26]. Second, the nanopores can provide a stable airpocket to ensure superhydrophobic property [26]. However, the SERS performance of superhydrophobic NTNPAs has not been reported. It is still a great challenge to fabricate NTNPAs in a facile and rapid way. The present study shows that superhydrophobic NTNPAs with high SERS performance can be facilely, rapidly and economically fabricated by a designed process—one-step voltage-variation anodization of Al foils followed by Ag nanoparticle deposition and FS modification. The structural parameters of alumina NTNPAs can be controlled by anodization time, and the one under 1560 s anodization followed by 400 s Ag deposition exhibited the highest SERS intensity. The preliminary study showed that this SERS-active structure can uniformly distribute in a large area (5  5 cm2), which leads to good reproducible SERS signals on the surfaces of different batches of samples. Finally, the SERS performance of the superhydrophobic NTNPAs with different FS modification and Ag deposition orders were studied. R6G with a concentration of 10 8 M in a 20 mL droplet can be rapidly detected on NTNPAs with FS modification followed by Ag deposition. 2. Experimental 2.1. Reagents and materials Highly pure (99.999%) aluminum foils with a thickness of 0.2 mm were used as starting materials (Cuibolin Nonferrous Metals Co., Ltd, Bejing, China). Analytical reagents were produced by Tianjing TianLi Chemical Reagents Ltd (China), including phosphoric acid (H3PO4, 85 wt%), chromium trioxide (CrO3, 99 wt%), perchloric acid (HClO4, 70–72 wt%). Ag targets (99.99%) were purchased form Thermo Fisher. The de-ionized water was produced by a pure water system.

139

sity (j) reach the peak value of 20 mA/cm2, after that the voltage exponentially decreased with the fixed j value. The anodization was stopped at the desired time. j was calculated by dividing the current values measured from the power supply system (IT 6874A, ITECH) by the anodized sample area. After reaction, anodized aluminium foils were rinsed with deionized water and then dried by nitrogen airflow. To obtain the SERS-active surface, Ag nanoparticles were uniformly deposited on the surface of alumina NTNPAs at the rate of 1 Å/s under 3  10 6 Pa by a customized DE 400 electron beam evaporator (DE Technology Inc, China). 2.3. Fluorosilane modification To obtain the superhydrophobic property, the as-prepared NTNPAs without or with deposited Ag nanoparticles were placed, together with 10 lL heptadecafluorodecyltrimethoxysilane liquid (Sigma-Aldrich Trading Co., Ltd., China), into a sealed glass container and then heated at 120 °C for 2 h. 2.4. Characterizations Top views of the as-prepared NTNPAs with different Ag nanoparticle deposition durations were directly studied under high-resolution scanning electronic microscope (SEM, Nova NanoSEM 450, USA). Side views of the sample were obtained after sputtering a 15-nm thickness of Au layer on alumina NTNPAs instead of Ag nanoparticle deposition. Water contact angles were measured at ambient temperature by an optical contact angle meter (Dataphysics OCA20, Germany). The hemispherical reflection spectra were measured by a spectrophotometer (Perkin Elmer lambda 950, USA) equipped with an integrating sphere for wavelengths of 400–800 nm. 2.5. SERS measurement For the samples without fluorosilane modification, the samples were immersed in identical 1 mM R6G aqueous solution for 8 h to make sure the surface adsorbed a layer of R6G molecules. Subsequently, the samples were dried by nitrogen gas flow after rinsed with de-ionized water. For the samples with fluorosilane modification, droplets (20 mL) with different R6G concentrations were directly deposited on the samples and the contrast sample. The droplet was evaporated at room temperature. Raman spectra were collected by a confocal Raman Spectrometer (Horiba Jobin-Yvon LabRAM HR Evolution) with a 50 objective lens and an excitation wavelength of 532 nm. The effect power of the laser source was 2 mW at the surface of the samples, which had an average spot size of 3 mm in diameter. The acquisition time was 1 s. 3. Results and discussion 3.1. Synthesis process of superhydrophobic NTNPAs for SERS detection

2.2. Fabrication of hybrid nanotip and nanopore arrays Al foils (50 mm  50 mm  0.2 mm) were firstly electropolished in a mixture of perchloric acid and ethanol (V/V = 1:4) for 8 min (20 V, 0 °C). The electropolished Al foil was then placed in a custom-tailored electrochemical cell, where the Al foil was used as the anode and a platinum electrode was employed as a counter electrode. The distance of these two electrodes was fixed at 3.5 cm. Alumina hybrid nanotip and nanopore arrays (NTNPAs) were fabricated by one-step voltage-variation anodization of Al foils in 40 °C phosphoric acid solution, which was under violent agitation. Al foils was anodized at 25 V for 30 s, and then the anodization voltage linearly increases with the rate of 0.5 V/10 s until current den-

The fabrication procedure for SERS and superhydrophobic NTNPAs is shown in Fig. 1. Alumina NTNPAs was firstly prepared by one-step voltage-variation anodization in high temperature phosphoric acid (Fig. 1a). The formation of NTNPAs should be ascribed to skillfully utilize the co-effects of enhanced aluminum anodization and alumina etching in high temperature phosphoric acid electrolyte, voltage-variation procedure and the different chemical composition of anodic alumina. Firstly, compared with the conventional pore growth (using 4 °C phosphoric acid) [27], our adopted electrolyte (40 °C phosphoric acid) can simultaneously realize the fast aluminum anodization and good etching ability upon the alumina nanopore wall. Secondly, voltage-variation

140

Y. Li et al. / Applied Surface Science 443 (2018) 138–144

the FS modified nanopores can guarantee the solid-liquid contact line on the Ag out layer, which made the sample superhydrophobic. For the FS-Ag nanosample (Fig. 1c), the stable air pocket in the nanopores retained the surface superhydrophobic, but the solid-liquid contact line was pulled inward due to the hydrophilic nature of deposited Ag out layer, which led to a smaller contact angle of analyte solution droplet compared with that on the AgFS nanosample [19]. Smaller contact angle deteriorated the concentration effect of the FS-Ag nanosamples, and the condensed analyte covered a relative larger area compared with the Ag-FS nanosamples (the bottom images of Fig. 1b and c). However, the SERS intensity of every condensed analyte molecule on the FS-Ag nanosamples was stronger than that on the Ag-FS nanosample, which may be attributed to the analyte molecule being close to the Ag nanoparticle where the electromagnetic field was generated. Thus, both the SERS performances of these two type superhydrophobic substrates were investigated. 3.2. Fabrication of NTNPA with optimal SERS activity

Fig. 1. A schematic showing the fabrication of superhydrophobic nanotip and nanopore arrays (NTNPAs) for SERS detection. (a) Alumina hybrid NTNPAs were firstly prepared by the one-step voltage-variation anodization process; (b, c) SERS and superhydrophobic functions were obtained by Ag nanoparticle deposition followed by fluorosilane (FS) modification (b) or FS modification followed by Ag nanoparticle deposition (c) on alumina NTNPAs.

anodization can accelerate nanopore growth and the period enlargement of nanopore along the anodization time (bottom part of Fig. 1a); thus, the previously formed alumina nanopores (top layer) experienced a long etching time, and the consequently formed nanopores produced thick cell walls for etching. Thirdly, the cell border of the alumina consisted of virtually pure alumina, whereas the walls were impure and doped with PO34 anions (the top part of Fig. 1a) [28]. Compared with the pure alumina at the cell borders, the walls made of contaminated alumina can be dissolved easily. The relatively thin top-layer alumina nanopores that formed initially had a long etching time in the phosphoric acid solution as they formed the separated nanotips. The underlying nanopores were widened due to a relatively shorter etching time. Ag deposition and FS modification on NTNPAs can generate high-density hot spots for SERS detection and make the sample superhydrophobic, respectively. However, the SERS activity of the substrate may be influenced by the formed monolayer if the FS modification was the last step. Whereas the superhydrophobicity may deteriorate if Ag deposition was the last step. The top images of Fig. 1b and c show the solid-liquid interface (not to scale) on the nanosamples with Ag nanoparticle deposition followed by FS modification (Ag-FS) and FS modification followed by Ag nanoparticle deposition (FS-Ag), respectively. For the Ag-FS nanosample (Fig. 1b), the air between the FS-modified Ag nanoparticle and

Alumina NTNPAs comprise the base nanostructure, which determines the SERS property. Thus it is necessary to firstly find out the proper electrochemical conditions to obtain the optimal structure for SERS detection. The formation mechanism of alumina NTNPAs indicates that anodization time is important in controlling the morphology. Thus, we investigated the SERS activity of nanostructured surfaces at seven typical anodization times by fixing Ag deposition duration at 400 s. The anodization time was chosen according to the voltage and current variation. As shown in Fig. 2a, the initial Al foil (0) was first anodized at 25 V for 30 s (1), and then the anodization potential (U) linearly increased with the rate of 0.5 V/10 s to 98 V at 1460 s (3). Current density (j) exponentially increased to the peak value of 20 mA/cm2 (from 3 to 6), and then U exponentially decreased with a fixed j value. Fig. 2b presents the SERS spectra of 1 mM R6G obtained from the seven samples, whose fluorescent backgrounds have been subtracted for the convenience of comparison. Before conducting SERS detection, the samples were dipped into the solution for 8 h and then dried at room temperature after rinsed with de-ionized water. It can be seen that the Raman signals were too weak to be detected on sample 1, which was highly similar to those on sample 0, because only a uniform barrier layer existed on Al surface (Fig. 2c). When anodization time was prolonged, SERS signal began to appear on nanosample 2 (Fig. 2b), as tiny nanopores formed on the surface (Figs. 2c and S1). The low protrusions around the nanopores can generate enhanced electromagnetic field [22]. However, several characteristic peaks of R6G, including 1310, 1361, 1574 and 1650 cm 1 had not been excited (Fig. 2b). When the anodization time was prolonged further, the SERS spectrum of R6G on Sample 3 showed all the characteristic peaks, because aligned and separated nanotips began to appear on the vertex of nanopores. However, the signal intensities were much weaker than those on Sample 4 and 5, because the low height of the nanotips (Figs. 2c and S1) affected the electromagnetic enhancement [23]. Sample 4 and 5 showed identical SERS spectra due to the long and separated nanotips (Figs. 2c and S1). However, the SERS activity decreased with the further elongation of anodization time to 1760 s (Sample 6) due to the aggregated nanotips (Figs. 2c and S1). Thus, the best anodization time was found to be between 1560 s and 1660 s. Ag nanoparticle deposition also plays a critical role in tailoring the top surface morphology of NTNPAs, and thus influencing the overall SERS effect. Fig. S2a shows the scanning electron microscope (SEM) top views of NTNPAs with different Ag deposition durations by fixing the anodization time at 1560 s. As can be seen, SERS intensities increased as the deposition duration increased

Y. Li et al. / Applied Surface Science 443 (2018) 138–144

141

Fig. 2. (a) Recorded current densities (j) and potential varied with time during one-step voltage-variation anodization process. Here the Arabic numbers of 0, 1, 2, 3, 4, 5, 6 that marked on the graph stand for anodization time of 0 s, 30 s, 940 s, 1460 s, 1560 s, 1660 s and1760 s, respectively; (b) SERS spectra of 1 mM R6G adsorbed on the nanostructured surfaces with Ag nanoparticle deposition duration of 400 s, whose anodization time corresponding to the Arabic numbers in (a), respectively; (c) Four representive SEM top views of the nanostructured surfaces as anodization time increasing from 30 s to 1760 s.

from 50 s to 400 s and then decreased as the duration extend to 600 s. Compared with the border of nanopores, nanotips on the top layer of nanopores were not the prefer deposition site of the Ag nanoparticle, which resulted in a faster deposition rate of the Ag nanoparticle at the border than that at the nanotips. Thus, the Ag nanoparticle cannot sufficiently cover the nanotips given that deposition duration was only 50 s; the nanotips began to connect to one another and turn to the pore border when deposition duration was 600 s (Fig. S2a). Consequently, the ideal SERS activity cannot be obtained in these two situations due to the lack of hot spots (Fig. S2b). When deposition duration was between 50 and 600 s, the nanotips became thicker as deposition duration increased, which gradually enhanced the coupling of the localized surface plasmons between the neighboring nanotips [8]. Under the current experimental conditions, Ag deposition duration of 400 s produced the optimal SERS substrate with the highest SERS sensitivity, whose average enhanced factor (EF) was 1.34  106 (the calculation of EF see Fig. S3). 3.3. SERS signal uniformity and reproducibility of NTNPAs Notably, this technique has intrinsic advantages for large area and batch production. Samples with 25 cm2 area were preliminary realized in our laboratory. Fig. 3a showed the nanostructured sample with uniform black-red color, which was fabricated by anodizing Al for 1560 s followed by 400 s Ag deposition. In contrast, the flat sample was similar to a mirror. Fig. 3b shows lowmagnification SEM images of the nanostructured sample shown in Fig. 3a. Note that, Au sputtering instead of Ag deposition on

the cross-section of alumina NTNPAs was employed to obtain a clearly side view. It was evident that aligned nanotips were uniformly distributed on the whole surface with a density of 1.04  108 per mm2. Further magnified SEM images revealed that the nanotips arranged with diameters of 42 ± 3 nm and heights of 253 ± 36 nm sit vertically on the vertexes of the nanopores, whose pore diameter and period were 104 nm and 149 nm, respectively. To evaluate the SERS signal uniformity and reproducibility of NTNPAs, we measured the SERS signal of absorbed 1 mM R6G on 30 randomly selected spots from one nanostructured sample (area 5  5 cm2) and points from three samples of different batches. Fig. 3c shows eight representative SERS spectra from each sample. No changes were observed in the characteristic peaks compared with the spectra from the different samples, indicating that the SERS signals from different spectra were all of comparable intensity. To obtain a statistically meaningful result, the standard deviation and average value of the five peaks at 612, 772, 1361, 1510 and 1650 cm 1 were shown in Fig. 3d. The relative stand deviation of the intensity at 1650 cm 1 was estimated to be 11.28% for one sample and 11.83% for different samples. The homogeneous distribution of the Raman intensity strongly supports the notion that NTNPAs are uniformly distributed in a large area, which can generate SERS signal with good uniformity and reproducibility. The uniform black-red color of nanostructured sample suggested the nanosample should have low reflectance. Fig. 4 shows that the nanosample had broadband antireflective property with a significantly low average hemispherical reflectance of 13% in the range of 400–800 nm, which can be ascribed to the varied refractive index along the short nanotips [29]. In comparison, the

142

Y. Li et al. / Applied Surface Science 443 (2018) 138–144

Fig. 3. (a) Optical images of as-prepared nanostructured sample (left, black-red color) and contrast flat sample (right, like mirrors), which were all deposited with Ag nanoparticles for 400 s; (b) SEM top-view (top) and side-view (bottom) of the as-synthesized nanostructured sample in (a). The inset showing the magnified details of toplayer nanotips; (c) SERS spectra of 1 lM R6G absorbed on 3 identical nanostructured samples, and each one was randomly chosen 8 spectra; (d) The standard deviation and average value of the five prominent peaks at 612, 772, 1361, 1510 and 1650 cm 1, respectively, which obtained from 3 identical nanostructured samples and 30 spots each.

Fig. 4. The nanostructured surface had broadband antireflective property with significantly low average hemispherical reflectance of 13% in the range of 400–800 nm. In comparison, the average reflectance of mirror-like flat surface was as high as 99%.

average reflection of mirror flat sample was as high as 99%. Clearly, this substrate can effectively harvest the energy of incident light, which resulted in resonance enhancement [6]. To gain a deeper insight into the SERS enhancement of the NTNPAs, the distribution of the near-field electromagnetic field was investigated by finitedifference time-domain (FDTD) simulations (Fig. 5). The maximum electric-field intensity was 10 at the edge of the nanotips. 3.4. SERS activity and superhydrophobicity of NTNPAs with different modification order This SERS-active nanosurface can become superhydrophobic after being modified with low-free-energy FS molecules. As seen

Fig. 5. The distribution of the electromagnetic field was investigated by finitedifference time-domain (FDTD) simulations. (a) Hexagonally arranged Ag nanotips with a diameter of 42 nm were constructed as model, and the gap between the nanotips is 40 nm which was surrounded by air. (b) The calculated electric-field intensity distribution. The incident laser line with wavelength of 532 nm was assumed to be normal to the sample and polarizations of the incident light are along the arrows shown below.

in Fig. 6a, the water static contact angle of NTNPAs with Ag deposition (Ag), Ag deposition followed by FS modification (Ag-FS) and FS modification followed by Ag deposition (FS-Ag), which were 46°, 152°, and 149°, respectively. The Ag out layer only had a subtle influence on the superhydrophobicity because Ag nanoparticle cannot deposit on cell wall of the pore underneath; meanwhile, the stable air-pocket in the nanopores can retain the surface superhydrophobicity [26]. The morphology evolution of a 10-lL droplet during evaporation on FS-Ag sample was recoded (Fig. 6b). Before 50 min, due to the low adhesion force between the drop and the surface, the drop reduced in volume while maintaining its quasi-spherical shape (self-similar geometrical transformation), which made the droplet solution became more and more concentrated. After that, the droplet collapsed due to reach a condition of instability, and then continuously and completely evaporated in a confined region [14].

Y. Li et al. / Applied Surface Science 443 (2018) 138–144

143

and FS modification to realize low-cost, large-area fabrication of superhydrophobic NTNPAs for sensitive SERS detection. The optimal structural parameters of NTNPAs for SERS detection were obtained by adjusting the anodization time and Ag deposition duration. NTNPAs with FS modification followed by Ag deposition can realize detection of 10 8 M R6G in a time-saving and spacesaving way. It should be noted that this method is simple, rapid and inexpensive, only needs 1560 s and very simple and inexpensive equipment to fabricate uniformly distributed alumina NTNPAs at large area (5  5 cm2). This method is very promising to be evolved to an industry compatible process by replacing the Ag physical vapor deposition to electrodeposition. This large-scale nanostructure with high sensitivity and reproducibility is likely to be a new practical SERS substrate that can be used to develop novel biological and chemical sensors. Acknowledgment This work was supported by the National Natural Science Foundation of China (21403285, 11474192), Natural Science Foundation of Shaanxi Province (2017JM2009), Fundamental Research Funds for the Central Universities (GK201603019, GK201601008, 2016TS042, GK201704002, 2017TS015), Scientific Research Foundation of Shaanxi Normal University. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.apsusc.2018.02.247. References

Fig. 6. (a) Optical images of a 4-lL water droplet on the alumina NTNPAs with Ag nanoparticle deposition (Ag), Ag nanoparticle deposition followed by FS modification (Ag-FS) and FS modification followed by Ag nanoparticle deposition (FS-Ag), respectively; (b) The morphology evolution of a 10-lL droplet during evaporation on the FS-Ag nanosample under ambient conditions; (c) Photographs show a 20-lL droplet of 1 lM R6G solution (top) and the corresponding concentrated spot (bottom) after evaporation on FS-Ag nanosample; (d) SERS spectra of dried 1 lM R6G droplet on Ag, Ag-FS and FS-Ag nanosamples, respectively; (e) SERS spectra of dried 20-lL R6G droplet with concentration of 10 8 M on FS-Ag and Ag-FS nanosample, respectively.

The Ag-FS sample showed the similar trend. Fig. 6c intuitively shows that a 20-lL droplet of 1 mM R6G solution can be concentrated into a circle spot with diameter of 1 mm on the FS-Ag sample. Thus, the SERS intensity at 612 cm 1 of a directly dried 10 6 M R6G droplet (20 lL) on FS-Ag nanosample was approximately 3 times higher than that on unmodified sample (Fig. 6d). The SERS signal intensity on FS-Ag nanosample was slightly better than that on the Ag-FS, because the absorbed R6G molecular was closely contact with the Ag layer without the inter FS layer. When the R6G concentration of the droplet was relatively high (1 lM), the SERS intensity at 1361 cm 1 on the FS-Ag nanosample was approximately 1.5 times higher than that on the Ag-FS nanosamples. When the R6G concentration of the droplet further decrease to 10 8 M (Fig. 6e), the characteristic peak of R6G at 612, 1361 and 1574 cm 1 still can be clearly detected on the FS-Ag nanosample, but noting can be detected on the Ag-FS nanosample. 4. Conclusions In summary, we have demonstrated a facile method of one-step voltage-variation anodization Al foils followed by Ag deposition

[1] J. Kneipp, Interrogating cells, tissues, and live animals with new generations of surface-enhanced raman scattering probes and labels, ACS Nano 11 (2017) 1136–1141. [2] L.A. Austin, B. Kang, M.A. EI-Sayed, Probing molecular cell event dynamics at the single-cell level with targeted pasmonic gold nanoparticles: a review, Nano Today 10 (2015) 542–558. [3] P.D. Howes, S. Rana, M.M. Stevens, Plasmonic nanomaterials for biodiagnostics, Chem. Soc. Rev. 43 (2014) 3835–3853. [4] H. Sun, H. Liu, Y. Wu, A. Green, Reusable SERS film with high sensitivity for insitu detection of thiram in apple juice, Appl. Sur. Sci. 416 (2017) 704–709. [5] X. Zhang, Y. Zheng, X. Liu, W. Lu, J. Dai, D.Y. Lei, D.R. MacFarlane, Hierarchical porous plasmonic metamaterials for reproducible ultrasensitive surfaceenhanced Raman spectroscopy, Adv. Mater. 27 (2015) 1090–1096. [6] C. Zhu, G. Meng, P. Zheng, Q. Huang, Z. Li, X. Hu, X. Wang, Z. Huang, F. Li, N. Wu, A Hierarchically ordered array of silver-nanorod bundles for surface-enhanced raman scattering detection of phenolic pollutants, Adv. Mater. 28 (2016) 4871–4876. [7] K. Chen, X. Zhang, Y. Zhang, D.Y. Lei, H. Li, T. Williams, D.R. MacFarlane, Highly ordered Ag/Cu hybrid nanostructure arrays for ultrasensitive surfaceenhanced raman spectroscopy, Adv. Mater. Interfaces 3 (2016) 1600115. [8] Q. Fu, Z. Zhan, J. Dou, X. Zheng, R. Xu, M. Wu, Y. Lei, Highly reproducible and sensitive SERS substrates with Ag inter-nanoparticle gaps of 5 nm fabricated by ultrathin aluminum mask technique, ACS Appl. Mater. Interfaces 7 (2015) 13322–13328. [9] B. Daglar, T. Khudiyev, G.B. Demirel, F. Buyukserin, M. Bayindir, Soft biomimetic tapered nanostructures for large-area antireflective surfaces and SERS sensing, J. Mater. Chem. C 1 (2013) 7842–7848. [10] T. Li, K. Wu, T. Rindzevicius, Z. Wang, L. Schulte, M.S. Schmidt, A. Boisen, S. Ndoni, Wafer-scale nanopillars derived from block copolymer lithography for surface-enhanced raman spectroscopy, ACS Appl. Mater. Interfaces 8 (2016) 15668–15675. [11] A. Horrer, C. Schäfer, K. Broch, D.A. Gollmer, J. Rogalski, J. Fulmes, D. Zhang, A.J. Meixner, F. Schreiber, D.P. Kern, M. Fleischer, Parallel fabrication of plasmonic nanocone sensing arrays, Small 9 (2013) 3987–3992. [12] F. Gentile, M.L. Coluccio, N. Coppedè, F. Mecarini, G. Das, C. Liberale, L. Tirinato, M. Leoncini, G. Perozziello, P. Candeloro, F.D. Angelis, E.D. Fabrizio, Superhydrophobic surfaces as smart platforms for the analysis of diluted biological solutions, ACS Appl. Mater. Interfaces 4 (2012) 3213–3224. [13] Z.-Y. Lv, A.-Q. Li, Y. Fei, Z. Li, J.-R. Chen, A.-J. Wang, J.-J. Feng, Facile and controlled electrochemical route to three-dimensional hierarchical dendritic gold nanostructures, Electrochim. Acta 109 (2013) 136–144. [14] F.D. Angelis, F. Gentile, F. Mecarini, G. Das, M. Moretti, P. Candeloro, M.L. Coluccio, G. Cojoc, A. Accardo, C. Liberale, R.P. Zaccaria, G. Perozziello, L. Tirinato, A. Toma, G. Cuda, R. Cingolani, E.D. Fabrizio, Breaking the diffusion

144

[15]

[16]

[17]

[18]

[19]

[20]

Y. Li et al. / Applied Surface Science 443 (2018) 138–144 limit with super-hydrophobic delivery of molecules to plasmonic nanofocusing SERS structures, Nat. Photonics 5 (2011) 682–687. M. Fan, F. Cheng, C. Wang, Z. Gong, C. Tang, C. Man, A.G. Brolo, SERS optrode as a ‘‘ fishing rod ” to direct pre-concentrate analytes from auperhydrophobic surfaces, Chem. Commun. 51 (2015) 1965–1968. Q.-X. Zhang, Y.-X. Chen, Z. Guo, H.-L. Liu, D.-P. Wang, X.-J. Huang, Bioinspired multifunctional hetero-hierarchical micro/nanostructure tetragonal array with self-cleaning, anticorrosion, and concentrators for the SERS detection, ACS Appl. Mater. Interfaces 5 (2013) 10633–10642. Z.-X. Yan, Y.-L. Zhang, W. Wang, X.-Y. Fu, H.-B. Jiang, Y.-Q. Liu, P. Verma, S. Kawata, H.-B. Sun, Superhydrophobic SERS substrates based on silver-coated reduced graphene oxide gratings prepared by two-beam laser interference, ACS Appl. Mater. Interfaces 7 (2015) 27059–27065. H. Jeong, J. Kim, Electrodeposition of nanoflake Pd structures: structuredependent wettability and SERS activity, ACS Appl. Mater. Interfaces 7 (2015) 7129–7135. J.-A. Huang, Y.-L. Zhang, Y. Zhao, X.-L. Zhang, M.-L. Sun, W. Zhang, Superhydrophobic SERS chips based on a Ag coated natural taro-leaf, Nanoscale 8 (2016) 11487–11493. Z. Yu, W. Song, L. Chen, Y. Park, B. Zhao, Q. Cong, Y.M. Jung, Simple immersion to prepare a Zn/Ag biomimetic superhydrophobic surface and exploring its applications on SERS. Colloids Surf. A: Physicochem. Eng. Asp. 467 (2015) 224– 232.

[21] A. Santos, T. Kumeria, D. Losic, Nanoporous anodic alumina: a versatile platform for optical biosensors, Materials 7 (2014) 4297–4320. [22] D. Choi, Y. Choi, S. Hong, T. Kang, L.P. Lee, Self-organized hexagonal-nanopore SERS array, Small 6 (2010) 1741–1744. [23] K. Sun, G. Meng, Q. Huang, X. Zhao, C. Zhu, Z. Huang, Y. Qian, X. Wang, X. Hu, Gap-tunable Ag-nanorod arrays on alumina nantip arrays as effective SERS substrates, J. Mater. Chem. C 1 (2013) 5015–5022. [24] C. Sui, K. Wang, S. Wang, J. Ren, X. Bai, J. Bai, SERS activity with tenfold detection limit optimization on a type of nanoporous AAO-based complex multilayer substrate, Nanoscale 8 (2016) 5920–5927. [25] W. Lee, R. Ji, U. Gösele, K. Nielsch, Fast fabrication of long-range ordered porous alumina membranes by hard anodization, Nat. Mater. 5 (2006) 741– 747. [26] J. Li, W. Zhang, Y. Luo, J. Zhu, X. Gao, Facile fabrication of anodic alumina rodcapped nanopore films with condensate microdrop self-propelling function, ACS Appl. Mater. Interfaces 7 (2015) 18206–18210. [27] A.P. Li, F. Müller, A. Birner, K. Nielsch, U. Gösele, Hexagonal pore arrays with a 50–420 nm interpore distance formed by self-organization in anodic alumina, J. Appl. Phys. 84 (1998) 6023–6026. [28] G.E. Thompson, G.C. Wood, Porous anodic film formation on aluminum, Nature 290 (1981) 230–231. [29] J. Li, J. Zhu, X. Gao, Bio-inspired high-performance antireflection and antifogging polymer films, Small 13 (2014) 2578–2582.