A new fluorescence enhancement technical platform based on anodic aluminum oxide (AAO) nanostructure substrate is reported for the first time. Several fluorophores have been examined on the AAO nanostructure substrates. Systematic experiments found that the enhancement factor can be up to two orders of magnitude compared to the fluorescence signals on a glass substrate, indicating its great potential for ultrasensitive fluorescence detection. Given the simple and cost-effective fabrication process of lithographically patterned AAO nanostructure, this type of AAO nanostructure platform has great potential applications, especially its integration with microdevices and microfluidic devices for fluorescence-based biological analysis.
©2012 Optical Society of America
Fluorescence sensing and detection is a very important technique and has been widely utilized in many different fields, such as medical imaging, a variety of biology detections (e.g., DNA arrays and gene sequencing), clinical chemistry, and environmental monitoring [1,2]. Fluorescence methods provide many advantages over other biodetection techniques, including high sensitivity and multiplexing capabilities even though the labeling of fluorescence dyes to biomolecules is still a tedious work. While fluorescence offers high sensitivity, how to further improve its sensitivity, to facilitate the detection of ultra-low concentration of targeted species, has been an active research topic of significant importance. To this end, optimum design of the fluorophores , improved detection method and apparatus  and advanced fluorescence enhancement substrate [5–8] are among the major research efforts.
Of all the advanced fluorescence substrates, the metallic nanostructured substrates have been mostly and successfully exploited. As a result, the metal-enhanced fluorescence (MEF) has become a widely used technique [5–8]. Specifically, gold (Au), silver (Ag) or aluminum (Al) nanoparticles or nanostructures have been widely used as fluorescence enhancers. In this case, usually the fluorophores are located in close proximity to the surface of the nanostructures. It is well known that the physical mechanism for the fluorescence enhancement behind the MEF is due to the interactions of the excited fluorophores with surface plasmon resonances in metals. In other words, when the fluorophores are in close proximity to the surface of the metallic nanostructures, their intrinsic decay rate and the quantum yield increase, leading to a remarkable decrease in the lifetime of the fluorophores and thus an enhanced fluorescence signals [5–8]. It has been found that the fluorescence enhancement has clear gap dependence between the fluorophores and the metallic surface , and thus an optimum distance for the enhancement exists. Due to the quenching effect, a thin layer of dielectric material (e.g., SiO2 of 5 nm) has to be applied to the metal nanoparticles or nanostructures to separate the metal nanostructures and the fluorophores. Some examples of MEF substrates include silver fractal nanostructures , gold surface nanoscale grating , aluminum bowtie nano-apertures , etc. However, besides the expensive metals such as Au or Ag, the construction of these types of metallic nanostructures also usually requires expensive nanofabrication process such as electron-beam lithography and focused ion beam milling, preventing the production of inexpensive nanostructured fluorescence enhancement substrates in an efficient manner.
In addition to the aforementioned metallic nanostructured substrates, some other non-metallic nanomaterials and nanostructures have been also discovered and exploited to increase the fluorescence signals and sensitivity. Some representatives include the nanoscaled zinc oxide (ZnO) substrate [9,10], ZnO/SiO2 core/shell nanorod substrate  and nanoscaled SnO2 substrate . The anodic aluminum oxide (AAO) nanostructures have also been explored and utilized for fluorescence detection [13,14]. One approach is to arrange an AAO layer on a Kretschmann-Raether (ATR) setup to achieve surface-plasmon resonance (SPR)-induced fluorescence platform . The other work was mainly focused on polarization-dependent analysis of the optical properties of AAO nanostructures on the fluorescence signals . Very little work has been reported to directly use the biocompatible AAO nanostructures as a fluorescence enhancement platform. Additionally, the AAO nanostructures usually have been fabricated from a high purity Al (99.9999%) sheet or the large Al thin film , no work has been reported to fabricate AAO nanostructures from the lithographically patterned Al thin film.
Herein, we report fluorescence enhancement substrates based on the patterned AAO nanostructures for the first time, which are fabricated in a cost-effective manner by combining a lift-off and one-step anodization process thus paving a way for developing disposable fluorescence enhancement substrates, and can be readily integrated with microdevices or microfluidic devices for a variety of fluorescence-based biological and biomedical applications.
2. Fabrication of nanostructured alumina and experimental procedure
An illustration of fabricating arrayed aluminum oxide nanostructure (AAO) patterns from lithographically patterned Al thin film is given in Fig. 1 . The detailed fabrication procedure has been reported in another paper. Briefly, start with a glass substrate coated with indium tin oxide (ITO). ITO thickness is 435~465nm. It is cleaned by using sonication with soapy DI water, acetone, ethanol, and then DI water, respectively. Between each solution of the clean process, the substrate is rinsed by DI water and then dried using a nitrogen gun. Then, Al of 1.5-2 µm with Ti of 10 nm as an adhesion layer is deposited on the substrate using E-beam evaporation. 2 × 2 Al patterns connected with each other with Al lines using a lift-off process. Third, one-step anodization process  is performed to form anodic aluminum oxide (AAO) nanopores: The anodization was performed in 0.3M sulfuric acid at a voltage of 17 V and a temperature of 2 °C. Monitoring and controlling the current flow in the Al thin film is the key during the anodization step. The AAO nanopore size and the spacing among nanopores can be tuned by changing the process parameters. By changing the layout of the patterns in Fig. 1, tens and hundreds of arrayed AAO patterns can be fabricated.
An example of the fabricated AAO nanostructures on ITO glass substrate is shown in Fig. 2 . The wafer-scale AAO nanostructure is given in Fig. 2(a), and lithographically patterned 2 × 2 AAO nanostructure is given in Fig. 2(b). Both of them are optically transparent. SEM images of the AAO nanostructure layer is shown in Fig. 2(c). As we can see, the nanopores with size of 10 ± 2 nm are formed in the Al thin film, while the Al thin film consists of Al grains, typical size is in the range of 50 nm to 400 nm. The measured roughness of the film is ~50-80 nm by using atomic force microscope (AFM) in Fig. 2(d). Compared to AAO fabricated from Al foil thin film using one-step or two-step anodization process, the distinct difference is that the topology of the surface of the AAO nanostructure layer. For latter case, the Al thin film is quite smooth without any nanoscale Al grains.
Fluorophores Rhodamine 6G (R6G) (Lightning Powder, Inc), fluorescein sodium salt (FSS) (Sigma, Inc), Calceim AM (Sigma, Inc) and fluorescent brightening agents (FBA) (Sigma, Inc) were used for the technical demonstrations. All of them are dissolved in dionized (DI) water or isopropyl alcohol (IPA). Solutions of fluorophores are uniformly applied on the AAO nanostructures and ITO glass by spin-coating. All the fluorescent images have been taken after the solutions become dried using a fluorescence microscope (Olympus, Inc), which has three filter sets: FITC (excitation filter: 475-490 nm; barrier filter: 500-540 nm); TRITC (excitation filter: 545-565 nm; barrier filter: 580-620 nm) and DAPI (excitation filter: 330-385 nm; barrier filter: 420-460 nm). The setup for the measurement of fluorescence spectra utilizes the similar filter sets and the fluorescent signals are recorded by a spectrometer (Ocean Optics, Inc.).
3. Experimental results and discussion
Systematic experiments have been carried out on a series of substrates coated with different types of fluorophores, resulting in consistent fluorescence enhancement. In Fig. 3(a) , the same amount of R6G is uniformly coated on the two substrates: one AAO nanostructure and ITO glass substrate, one 10 nm Au coated AAO nanostructure and ITO glass substrate. Measurements were then taken after the substrates became dried. It reveals that the fluorescence of R6G on AAO nanostructure has been enhanced significantly compared to that on the ITO glass, even though the fluorescence of R6G on Au-coated AAO nanostructure has somewhat less enhancement, which might be due to the quenching effect since the R6G is directly coated on the Au thin film . In order to confirm the observed phenomenon, the Calceim AM is also uniformly coated on a substrate as shown in Fig. 3(b). Part of this substrate is coated with 5 nm Au and part of this substrate is not coated with Au. As can be seen in Fig. 3(b), as expected, the fluorescence enhancement has some difference on the substrate, which is clearly visible at the interface with and without Au coating. Based on these experiments, we used 4 patterned AAO nanostructures to obtain the fluorescent images of three fluorophores as shown in Fig. 3(c).
To quantify the fluorescence enhancement of the AAO nanostructure substrate compared to a glass substrate, the fluorescence spectra of R6G and FSS on an AAO nanostructure substrate and a glass substrate of the same size have been measured. The optical scattering spectra of the same AAO nanostructure substrate and glass substrate prior to attaching any fluorophores have been measured as a reference. After subtracting the optical scattering spectra from the fluorescence spectra, the corrected fluorescence spectra of R6G and FSS are obtained in Fig. 4 . The fluorescence enhancement factor of the AAO nanostructure substrate compared to the glass substrate for both R6G and FSS is two orders of magnitude, which is comparable to that of the ZnO nanostructure substrate . Experiments found that the fluorescence intensities from FITC anti-IgG molecules on ZnO nanostructure substrate were three orders of magnitude higher than those from FITC anti-IgG molecules of the same concentration of 200 µg/mL on Si, Si nanowires, glass, quartz and PMMA substrates .
Some control experiments have been carried out to understand the fluorescence enhancement mechanism. In Fig. 5(a) , it shows that the fluorescence enhancement on totally anodized Al (pure AAO) is somewhat lower than that on the partially anodized Al (partial AAO). The partial AAO is not optically transparent as the pure AAO as shown in the bright field optical image in Fig. 5(b). A layer of opaque Al below a layer of AAO in the partial AAO causes enhanced reflectance of the excitation light, thereby resulting in an enhanced fluorescence. In order to confirm the critical role of AAO nanostructure in the fluorescence enhancement, the AAO layer has been etched away by immersing the AAO substrate in a mixture solution of phosphoric acid (0.4 M) and chromic acid (0.2 M) at 65 °C over night, followed by rigorous DI water rinse. The SEM image of the substrate after the AAO layer etching is given in Fig. 5(c-d), which is clearly different from those in Fig. 2(c). The remaining on the substrate might be anodized Ti. However it is too thin to be determined using an X-ray photoelectron spectroscopy (XPS). The fluorescent dye R6G is then uniformly coated on the substrate. The fluorescence images are shown in Fig. 5(e). No fluorescence enhancement can be observed.
The surface area effect of the AAO nanostructure on the fluorescence enhancement has been evaluated. First, the same amount of R6G and FSS is spin-coated on AAO nanostructure substrates and glass substrates of the same size, respectively. The fluorescence images and spectra of the 4 substrates have been taken after they dried. Then the substrates are immersed in DI water of the same amount in 4 beakers and are ultrasonically rinsed for several hours. Thereafter, the fluorescence images and spectra of the 4 washed substrates have been measured. After washing, essentially no R6G remains on the AAO nanostructure substrate, which is confirmed by the fluorescence image as shown in Fig. 6(a) . The fluorescence spectra of R6G on AAO after washing in Fig. 6(b) further confirm that no R6G was left on AAO nanostructure substrate. For FSS, very little amount of fluorescence dyes remains on the AAO nanostructure as confirmed in Fig. 6(a). The measured fluorescence spectra of the R6G and FSS solution washed from AAO nanostructure substrates and glass substrates are shown in Fig. 6(c) and Fig. 6(d), indicating the amount of R6G and FSS attached on the AAO nanostructure substrate is lower than that on glass substrate before washing. These results suggest that the surface area of AAO nanostructure for attaching fluorophores does not increase compared to that of a glass substrate with the same size. In contrast, the amount of fluorophores attached to the surface of AAO nanostructure is lower than that on the glass substrate based on systematic experiments, indicating the intrinsic property of AAO nanostructures plays a critical role in this enhancement, probably similar to that of the ZnO and SnO2 nanostructures [9–12].
Similar to other metal oxides such as ZnO nanorods, ZnO nanoscaled film and SnO2 nanostructures for fluorescence enhancement [9–12], the mechanism of the AAO nanostructure substrate for fluorescence enhancement is different from MEF since no metals exist in the substrate. The exact mechanism requires further study. Some possible explanations are summarized in the following: First, the surface scattering effects of the nanostructured AAO cause the redistribution of the electromagnetic fields with high surface intensities, eventually resulting in the enhanced fluorescence [17,18]. Second, based on the research on ZnO nanostructures for fluorescence enhancement [9–12], the following effects might also play very important roles in the observed fluorescence enhancement. Similar to the waveguiding property of the ZnO nanomaterials and its ability to enhance the intensity of the evanescent field, the average size of the grains in the AAO film (Fig. 2) is similar to that of ZnO nanorods to guide the visible light effectively, thereby enhancing the intensity of the evanescent field. Hence, the AAO nanostructures can serve as efficient evanescent waveguides, enabling the fluorescence enhancement tremendously .
The AAO nanostructure substrate for fluorescence enhancement is reported for the first time. Experiments found that the fluorescence enhancement factor can be up to two orders of magnitude compared to a glass substrate, indicating its ability to detect ultralow trace level of fluorophore-labeled biomolecules. Due to the easiness of large scale fabrication of patterned AAO nanostructure on a single chip, this technique might benefit the multiplexing fluorescence-based biosensing tremendously. The simple, cost-effective and disposable nature of this type of sensor is attractive for rapid point-of-care and field biodetection applications as well.
This research is supported in part by a grant of NSF Pfund-Louisiana 2012 and a NSF CAREER award.
References and links
6. E. M. Goldys, K. Drozdowicz-Tomsia, F. Xie, T. Shtoyko, E. Matveeva, I. Gryczynski, and Z. Gryczynski, “Fluorescence amplification by electrochemically deposited silver nanowires with fractal architecture,” J. Am. Chem. Soc. 129(40), 12117–12122 (2007). [CrossRef] [PubMed]
8. G. Lu, W. Li, T. Zhang, S. Yue, J. Liu, L. Hou, Z. Li, and Q. Gong, “Plasmonic-enhanced molecular fluorescence within isolated bowtie nano-apertures,” ACS Nano 6(2), 1438–1448 (2012). [CrossRef] [PubMed]
10. V. Adalsteinsson, O. Parajuli, S. Kepics, A. Gupta, W. B. Reeves, and J. I. Hahm, “Ultrasensitive detection of cytokines enabled by nanoscale ZnO arrays,” Anal. Chem. 80(17), 6594–6601 (2008). [CrossRef] [PubMed]
11. J. Zhao, L. Wu, and J. Zhi, “Fabrication of micropatterned ZnO/SiO2 core/shell nanorod arrays on a nanocrystalline diamond film and their application to DNA hybridization detection,” J. Mater. Chem. 18(21), 2459–2465 (2008). [CrossRef]
12. C. Gu, J. Huang, N. Ni, M. Li, and J. Liu, “Detection of DNA hybridization based on SnO2 nanomaterial enhanced fluorescence,” J. Phys. D Appl. Phys. 41(17), 175103 (2008). [CrossRef]
13. S. Cloutier, A. Lazareck, and J. Xu, “Detection of nano-confined DNA using surface-plasmon enhanced fluorescence,” Appl. Phys. Lett. 88(1), 0139041–0139043 (2006). [CrossRef]
14. R. Li and H. Grebel, “Surface enhanced fluorescence: polarization characteristics,” IEEE Sens. J. 10(3), 465–468 (2010). [CrossRef]
16. G. Sulka, S. Stroobants, V. Moshchalkov, G. Borghs, and J. P. Celis, “Synthesis of well-ordered nanopores by anodizing aluminum foils in sulfuric acid,” J. Electrochem. Soc. 149(7), D97–D103 (2002). [CrossRef]
17. N. Ganesh, W. Zhang, P. C. Mathias, E. Chow, J. A. Soares, V. Malyarchuk, A. D. Smith, and B. T. Cunningham, “Enhanced fluorescence emission from quantum dots on a photonic crystal surface,” Nat. Nanotechnol. 2(8), 515–520 (2007). [CrossRef] [PubMed]
18. T. Fujii, Y. Gao, R. Sharma, E. L. Hu, S. P. DenBaars, and S. Nakamura, “Increase in the extraction efficiency of GaN-based light-emitting diodes via surface roughening,” Appl. Phys. Lett. 84(6), 855–857 (2004). [CrossRef]