Remote surface enhanced Raman spectroscopy imaging via a nanostructured optical fiber bundle Valérie Guieu,1 Patrick Garrigue,1 François Lagugné-Labarthet,2* Laurent Servant,1 Neso Sojic, 1* and David Talaga1 1
2
Université Bordeaux 1, Institut des Sciences Moléculaires, UMR CNRS 5255, 351 Cours de la Libération 33405 Talence cedex, France University of Western Ontario, Department of Chemistry, 1151 Richmond Street, London, On, N6A5B7, Canada *
[email protected]
Abstract: Remote surface enhanced Raman spectroscopy (SERS) imaging of an adsorbed monolayer was demonstrated through a nanostructured array of conical tips inscribed onto the distal face of a 30 cm optical fiber bundle. Despite intense Raman signal from the germanium oxide doped fibers, the Raman signal of an adsorbed monolayer of a reference compound (benzene thiol) was detected in the fingerprint region. This opens up the possibility of local remote imaging through an optical fiber that embeds a SERS active platform. ©2009 Optical Society of America OCIS codes: (000.0000) General; (000.2700) General science.
References and links 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.
R. A. Alvarez-Puebla, D. S. dos Santos, Jr., and R. F. Aroca, “SERS detection of environmental pollutants in humic acid–gold nanoparticle composite materials,” Analyst (Lond.) 132(12), 1210–1214 (2007). Y. Liu, K. Chao, M. S. Kim, D. Tuschel, O. Olkhovyk, and R. J. Priore, “Potential of Raman spectroscopy and imaging methods for rapid and routine screening of the presence of melamine in animal feed and foods,” Appl. Spectrosc. 63(4), 477–480 (2009). C. L. Brosseau, K. S. Rayner, F. Casadio, C. M. Grzywacz, and R. P. Van Duyne, “Surface-enhanced Raman spectroscopy: a direct method to identify colorants in various artist media,” Anal. Chem. 81(17), 7443–7447 (2009). C. Matthäus, T. Chernenko, J. A. Newmark, C. M. Warner, and M. Diem, “Label-free detection of mitochondrial distribution in cells by nonresonant Raman microspectroscopy,” Biophys. J. 93(2), 668–673 (2007). A. Barhoumi, D. Zhang, F. Tam, and N. J. Halas, “Surface-enhanced Raman spectroscopy of DNA,” J. Am. Chem. Soc. 130(16), 5523–5529 (2008). D. L. Jeanmaire, and R. P. Van Duyne, “Surface Raman spectroelectrochemistry: Part I. Heterocyclic, aromatic, and aliphatic amines adsorbed on the anodized silver electrode,” J. Electroanal. Chem. 84(1), 1–20 (1977). J. A. Dieringer, A. D. McFarland, N. C. Shah, D. A. Stuart, A. V. Whitney, C. R. Yonzon, M. A. Young, X. Zhang, and R. P. Van Duyne, “Surface enhanced Raman spectroscopy: new materials, concepts, characterization tools, and applications,” Faraday Discuss. 132, 9–26 (2006). N. Marquestaut, A. Martin, D. Talaga, L. Servant, S. Ravaine, S. Reculusa, D. M. Bassani, E. Gillies, and F. Lagugné-Labarthet, “Raman enhancement of azobenzene monolayers on substrates prepared by LangmuirBlodgett deposition and electron-beam lithography techniques,” Langmuir 24(19), 11313–11321 (2008). V. Guieu, F. Lagugné-Labarthet, L. Servant, D. Talaga, and N. Sojic, “Ultrasharp optical-fiber nanoprobe array for Raman local-enhancement imaging,” Small 4(1), 96–99 (2008). L. Novotny, “Optical antennas tuned to pitch,” Nature 455(7215), 887 (2008). S. Cintra, M. E. Abdelsalam, P. N. Bartlett, J. J. Baumberg, T. A. Kelf, Y. Sugawara, and A. E. Russell, “Sculpted substrates for SERS,” Faraday Discuss. 132, 191–199, discussion 227–247 (2006). P. Matousek, “Deep non-invasive Raman spectroscopy of living tissue and powders,” Chem. Soc. Rev. 36(8), 1292–1304 (2007). H. G. Schulze, L. S. Greek, C. J. Barbosa, M. W. Blades, B. B. Gorzalka, and R. F. B. Turner, “Measurement of some small-molecule and peptide neurotransmitters in-vitro using a fiber-optic probe with pulsed ultraviolet resonance Raman spectroscopy,” J. Neurosci. Methods 92(1-2), 15–24 (1999). C. Viets, and W. Hill, “Single-fibre surface-enhanced Raman sensors with angled tips,” J. Raman Spectrosc. 31(7), 625–631 (2000). E. J. Smythe, M. D. Dickey, J. Bao, G. M. Whitesides, and F. Capasso, “Optical antenna arrays on a fiber facet for in situ surface-enhanced Raman scattering detection,” Nano Lett. 9(3), 1132–1138 (2009).
#120087 - $15.00 USD
(C) 2009 OSA
Received 17 Nov 2009; revised 2 Dec 2009; accepted 2 Dec 2009; published 16 Dec 2009
21 December 2009 / Vol. 17, No. 26 / OPTICS EXPRESS 24030
16. J. P. Scaffidi, M. K. Gregas, V. Seewaldt, and T. Vo-Dinh, “SERS-based plasmonic nanobiosensing in single living cells,” Anal. Bioanal. Chem. 393(4), 1135–1141 (2009). 17. D. J. White, A. P. Mazzolini, and P. R. Stoddart, “Fabrication of a range of SERS substrates on nanostructured multicore optical fibres,” J. Raman Spectrosc. 38(4), 377–382 (2007). 18. P. R. Stoddart, and D. J. White, “Optical fibre SERS sensors,” Anal. Bioanal. Chem. 394(7), 1761–1774 (2009). 19. A. Lucotti, and G. Zerbi, ““Fiber-optic SERS sensor with optimized geometry,” Sensor. Actuat, B 121(2), 356– 364 (2007). 20. G. Kostovski, D. J. White, A. Mitchell, M. W. Austin, and P. R. Stoddart, “Nanoimprinted optical fibres: Biotemplated nanostructures for SERS sensing,” Biosens. Bioelectron. 24(5), 1531–1535 (2009). 21. V. Guieu, D. Talaga, L. Servant, N. Sojic, and F. Lagugné-Labarthet, “Multitip-Localized Enhanced Raman Scattering from a Nanostructured Optical Fiber Array,” J. Phys. Chem. C 113(3), 874–881 (2009). 22. C. N. LaFratta, and D. R. Walt, “Very high density sensing arrays,” Chem. Rev. 108(2), 614–637 (2008). 23. H. H. Gorris, and D. R. Walt, “Mechanistic aspects of horseradish peroxidase elucidated through single-molecule studies,” J. Am. Chem. Soc. 131(17), 6277–6282 (2009). 24. A. Rasmussen, and V. Deckert, “New dimension in nano-imaging: breaking through the diffraction limit with scanning near-field optical microscopy,” Anal. Bioanal. Chem. 381(1), 165–172 (2005). 25. G. S. Henderson, D. R. Neuville, B. Cochain, and L. Cormier, “The structure of GeO2–SiO2 glasses and melts: A Raman spectroscopy study,” J. Non-Cryst. Solids 355(8), 468–474 (2009). 26. E. C. Le Ru, C. Galloway, and P. G. Etchegoin, “On the connection between optical absorption/extinction and SERS enhancements,” Phys. Chem. Chem. Phys. 8(26), 3083–3087 (2006). 27. J. Grand, M. de la Chapelle, J.-L. Bijeon, P.-M. Adam, A. Vial, and P. Royer, “Role of localized surface plasmons in surface-enhanced Raman scattering of shape-controlled metallic particles in regular arrays,” Phys. Rev. B 72(3), 033407 (2005). 28. N. Félidj, J. Aubard, G. Lévi, J. R. Krenn, A. Hohenau, G. Schider, A. Leitner, and F. R. Aussenegg, “Optimized surface-enhanced Raman scattering on gold nanoparticle arrays,” Appl. Phys. Lett. 82(18), 3095–3097 (2003). 29. A. D. McFarland, M. A. Young, J. A. Dieringer, and R. P. Van Duyne, “Wavelength-scanned surface-enhanced Raman excitation spectroscopy,” J. Phys. Chem. B 109(22), 11279–11285 (2005).
1. Introduction The local identification of analytes at a molecular level with a high sensitivity, and possibly remotely, is a major issue of significance in many practical or fundamental fields, ranging from pollutant monitoring [1], security screening [2], artwork authentication [3], to single cell investigation studies [4] or biochemical analysis [5]. Optical methods, and more specifically, Raman spectroscopy, are of particular interest, as they can provide direct molecular fingerprint identification of the sample without any labeling procedure or sample preparation. Despite its intrinsic weak signal due to small scattering cross-sections of about 10−30 cm2, new methods to enhance the electromagnetic field have led to the conception of devices and modified surfaces that allow a significant decrease of the acquisition time, the laser irradiance, and of the sample size necessary for the investigation. Since the discovery of the enhancements effects from metallic surfaces reported in the pioneering work of Jeanmaire and Van Duyne [6], the fundamental comprehension of surface enhanced Raman spectroscopy (SERS) has outlined several requirements to obtain large field enhancement at the sample. The architecture of organized 2D or 3D sharp metallic nanostructures arrays can be tailored to control the frequency of the plasmon polariton [7,8], as well as the development of localized antennas [9–11]. The technical facilities presently available to engineer optimized and reproducible platforms with nanoscale dimensions have fostered the re-discovery of field enhancement opportunities from metallic surfaces, in particular when combined with optical fibers. The use of optical fibers for non-invasive [12] and in vitro [13] Raman measurements has been successfully implemented with modern optical instrumentation, and most recent works combining optical fibers with SERS capabilities have revealed new opportunities to probe functionalized surfaces or to insert them in an aqueous medium or living cells [14–19]. It is noteworthy that in the work of Capasso and co-authors [15], a SERS 2D gold array with a well defined geometry prepared by electron beam lithography, was transferred from a silicon wafer onto one facet of a single core 35 cm long fiber. The enhancement factor was estimated to be about 5x105 for benzenethiol and 2-[(E)-2-pyridin-4-ylethenyl]pyridine. Such an approach is quite interesting, since the SERS structures can be finely controlled in terms of geometry and the plasmonic properties can be perfectly tuned using state of the art electron beam lithography. The transfer of the SERS platform onto the facet of the fiber presents, #120087 - $15.00 USD
(C) 2009 OSA
Received 17 Nov 2009; revised 2 Dec 2009; accepted 2 Dec 2009; published 16 Dec 2009
21 December 2009 / Vol. 17, No. 26 / OPTICS EXPRESS 24031
however, a major technical challenge in the production of such devices on a large scale. An elegant approach based on the replication of a natural structure on the distal facet of a single core optical fiber was successfully accomplished by Kostovski et al. using nanoimprinting [20]. Large SERS enhancements were reported, but fiber lengths were limited to 2-4 cm. On the other hand, the SERS active fiber developed by Vo-Dinh and co-authors was based on the deposition of a thin silver film onto a tapered single core fiber [16]. This sensor was further modified with para-mercaptobenzoic acid for the detection of local pH changes in biological cells. In this study, it is remarkable that the single fiber SERS probe was successfully inserted inside the cell without inducing apoptotic response of the cells. To complement the single core optical fiber used for Raman-SERS measurements mentioned above, optical fiber bundles are also very versatile devices, in that they can be used for Raman SERS remote measurements. For such an approach, a facet of the fiber is spatially structured using chemical etching and leads to a collection of identical structures, such as microwells or nanotips. The shape can be controlled by changing the etchant concentration [9,17,21]. Multicore optical fibers have also been successfully developed and commercialized for high-density local sensing (Illumina Inc. and Quanterix Inc.) combined with fluorescence measurements [22,23]. This remote imaging approach is very powerful, but since it is based on fluorescence, it is still restrained by the intrinsic limitations of the principle of detection such as autofluorescence, poor selectivity, sample preparation or labeling using exogenous agents. In this work, we have developed a new method for fabricating a gold coated multi-core optical fiber array, the extremity of which displays a collection of identical sharp metallic tips. The optical fiber bundle embeds 6000 individually-cladded 3 µm fibers made of silica and doped with GeO2. The tips, obtained by selective chemical etching, were coated with gold to enhance the electromagnetic field at their apexes. The nanostructured fiber bundle was combined with surface-enhanced Raman imaging measurements to probe molecules adsorbed at its surface. The incident irradiation was performed through fibers that were 30-50 cm long and the Raman signal was collected in the back-scattering collection (Fig. 1) mode through the same bundle. Such devices offer a considerable advantage for remote analysis in a complex environment, with limited access to in vivo or in vitro measurements. In addition to the local signal enhancement due to the surface geometry of the nanostructured imaging bundle, one can gain a significant improvement in terms of spatial resolution: the enhancement being generated at a very confined location on the tip apexes, thus the observed local Raman enhancement will be spatially confined to a very small sample area, surpassing the classical spatial resolution limited to ~λ/2 [21,24]. 2. Methods SERS fiber bundle preparation: The nanotip array was prepared by a wet-etching procedure of an optical fiber bundle (Sumitomo, IGN-035/06, length 30-50 cm). Both facets were first polished using 30-15-3-0.3 µm lapping paper. Silica imaging fiber bundle with a diameter of 270 µm was used. One polished side was left for 90 minutes in a buffer solution of hydrofluoric acid (HF) consisting of 40% aqueous NH4F and 48% HF in deionized water in proportion 5/1/1, and was then rinsed with deionized water. The fluorine cladding was primarily dissolved, whereas a spatial structural change of the GeO2 doped core as sharp angular tips with a cone angle of (35 ± 3)° was observed as shown in Fig. 1. The etched fiber was then sputter-coated with a 30 nm thick gold film. The nanostructured facet was immersed in a 10−2 M solution of benzenethiol (99.9%, Sigma Aldrich) in absolute ethanol (99.9%, Sigma Aldrich) for 4 hours and rinsed with ethanol to remove benzenethiol molecules that are not covalently bound.
#120087 - $15.00 USD
(C) 2009 OSA
Received 17 Nov 2009; revised 2 Dec 2009; accepted 2 Dec 2009; published 16 Dec 2009
21 December 2009 / Vol. 17, No. 26 / OPTICS EXPRESS 24032
Fig. 1. Principles of SERS through the fiber bundle in the backscattering geometry. The tip array (SEM image) coated with gold (30 nm) is located on the distal face while the Raman signal is collected at the proximal face. The Raman map (right) shows the color-coded variations of the intensity of a Raman band of adsorbed species that correlates with the tip apexes locations. Each pixel on the Raman map is a spectrum that was collected from an individual tip position.
3. Results and discussion As reported in a previous work using near field scanning optical microscopy, the light coupled to the etched nanostructured facet is confined to the apexes of the individual tip fibers [21]. More importantly, this result has also outlined the regularity of the tip arrangement in a hexagonal array. This characteristic is very important since it allows one to correlate a Raman spectrum together with a specific tip location in the bundle as outlined in Fig. 1. To further confirm the capabilities of these fiber bundles to perform local Raman remote measurements through the fiber, surface enhanced experiments were carried out with a reference sample adsorbed onto the surface of the gold-coated array of nanotips. The array was functionalized with a monolayer of benzenethiol (BT) by immersion in a solution (CBT = 1.0x10−2 M), to estimate possible enhanced effects by the nanostructured array through the fiber. BT was used as a reference sample to avoid any resonance Raman effect and to compare the efficiency of our substrate with previously published results using the same molecule on different SERS substrates [15,18]. The Raman measurements were first performed in a back-scattering geometry. Results shown here were obtained with a fiber of 30 cm length. Excitation laser light from a Kr+ laser (λ = 752 nm, 4 mW) was focused onto the polished proximal face of the optical fiber bundle and guided by total internal reflection in the individual fiber cores through the bundle. The light reaching the distal face was confined to the apex of the tip and induced the enhancement of the Raman signal of the adsorbed BT molecules. A fraction of the Raman signal was backscattered and collected by the same core, transmitted, and eventually detected at the proximal face. (Fig. 2) To achieve this, the section of the proximal side was raster scanned with a piezoelectric stage and the signal was collected with a 100X, N.A. = 0.9 microscope objective (confocal hole open). As shown in Fig. 2(a), the Raman spectra all show a very strong broad peak at 430 cm−1 which can be assigned to the symmetric stretching mode of bridging oxygens in Ge-O-Ge and Si-O-Si 6 membered rings. Contributions at 565 cm−1 can be assigned to bending/deformation of Ge-O-Ge, and as well as TO-LO splitting while contribution at 797 cm−1 is attributed to the TO mode of Si-O-Si. The peak at 682 cm−1 can be tentatively assigned to a defect breathing mode between oxygen and Ge/Si [25]. The Raman spectra of the BT monolayer adsorbed onto the tip gold-coated array is very weak but distinguishable from the fiber spectrum and has sharp peaks in the fingerprint region, i.e. between ~1000 and 1600 cm−1. Figure 2(b) shows mainly vibrational modes of the aromatic ring at 1001, 1027, 1076, 1099 and 1576 cm−1 and can be assigned to ν12, ν18a, ν1, ν6a + ν7a and ωC = C + ν8a/b modes, respectively. The C-S stretching mode cannot be observed due to overlap with the
#120087 - $15.00 USD
(C) 2009 OSA
Received 17 Nov 2009; revised 2 Dec 2009; accepted 2 Dec 2009; published 16 Dec 2009
21 December 2009 / Vol. 17, No. 26 / OPTICS EXPRESS 24033
fiber spectrum. The identification of adsorbed molecules onto the fiber arrays can be assigned unambiguously, provided that there is little overlap with the fiber spectrum. A Raman map was acquired by integrating the Raman mode of BT at 1576 cm−1. (Fig. 2(c), the integration was performed using the [1535-1611] cm−1 region) The map shows that the maximum enhancement was observed at the tip positions. Previous studies have shown that the focusing of the collection objective was very important and that the larger enhancements effects were observed at the tip apexes [21]. The regularity of the “hot spots” can be overlayed with the arrangement of the individual tips located at the surface of the bundle. A set of (X,Y) coordinates can be assigned to each individual fiber, which opens the possibility to spatially address individual tips through the fiber and correlate the signal from a given tip with a defined position of the sample.
Fig. 2. (a) Tip enhanced Raman spectra collected at the proximal face of a 30 cm fiber bundle in the backscattering geometry. An acquisition time of 1.5 s per spectrum was used. The intensity of the signal collected between the fiber cores is scaled by a factor 100. (b) Fingerprint region between 800 and 1800 cm−1. (c) Raman mapping though a 30 cm long fiber bundle of the BT vibrational mode at 1576 cm−1.
Interestingly, the Raman surface mapping was also performed on the distal face of the structured fiber in a back scattering geometry without going through the fiber. As shown in Fig. 3(a), neither fluorescence nor fiber background were observed. The lower frequency region shows sharps peaks at 417, 694 and 732 cm−1 assigned to ν7a + νC-S, ν6a + νC-S and γCCC modes. Raman spectra of the BT monolayer were also obtained in locations between the fibers, but with a much weaker intensity. The associated Raman mapping [Fig. 3(b)] was obtained with a shorter acquisition time of 0.5 s per spectrum and with a laser power of 0.4 mW at the microscope objective. Such short acquisition time and low laser power demonstrates how this setup provides efficient electromagnetic enhancement to induce local SERS effect at the sample. The estimation of the Raman enhancement factor in the backscattering geometry is rather imprecise, due to the strong scattering of the fiber bundle itself. Nevertheless, we have shown in a previous work that for the measurements in the transmission geometry, an enhancement factor of ~5x104 was precisely measured at the extremities of the tips [21]. This order of magnitude is comparable to the values reported by Smythe et al. on a single core fiber with a nanostructured facet [15]. The limitation caused by the length of the optical bundle is also a critical point. The longer the fiber the stronger its intrinsic Raman signal will be. This emphasizes the importance of further development of optical fiber with low loss and low scattering for either single or multi core optical fibers. Thus far, optical fibers bundle with length of 30 to 50 cm have shown successful results for remote Raman applications. Similarly the spectral region accessible is dependent on the composition of the fiber. In our work the fiber has a strong signal center at 400 cm−1 while in the work of Smythe et al. a strong feature is centered at 1330 cm−1 [15]. Nevertheless, the vibrational spectra of organic molecules adsorbed on such a surface can still be detected, since some spectral windows are free of the fiber signal.
#120087 - $15.00 USD
(C) 2009 OSA
Received 17 Nov 2009; revised 2 Dec 2009; accepted 2 Dec 2009; published 16 Dec 2009
21 December 2009 / Vol. 17, No. 26 / OPTICS EXPRESS 24034
It is also important to mention that the signal/noise ratio of the SERS experiments can certainly be improved by tuning the plasmon frequency of the metallic nanostructure located at the facet of the fiber, so that matching with the excitation wavelength is effective. More precisely, even though a matching between the absorption of the surface plasmon and the excitation wavelength is not a generalized prerequisite condition [26], it has been pointed out by several groups [27–29], that the adequate condition for a given Raman vibration to be λ + λexcitation where λplasmon, λRaman and λexcitation are the enhanced is met when λ plasmon ≈ Raman 2 wavelengths of the plasmon band (at its maximum), the Raman vibrational mode of interest and the excitation wavelength, respectively. This simplified relation shows the importance to precisely control the geometry of the features of the SERS platform. For our samples, the geometry of the distal facet upon etching could be tailored depending on the etching conditions. The plasmon frequency of these various metalized structures should definitely play an important role on the final value observed for the enhancement factor and should systematically be determined for optimized enhancements.
Fig. 3. (a) Raman spectra collected in the transmission geometry. The acquisition time was set to 0.5 s per spectrum. (b) Variation of the Raman intensity of the benzenethiol mode at 1575 cm−1. The integration range is [1535-1611] cm−1.
Conclusion
In conclusion, we have demonstrated the ability to remotely detect the Raman signals of a benzenethiol monolayer through a nanostructured optical fiber bundle. Much larger enhancement would be expected if resonant conditions were used or if the plasmon frequency of the structure inscribed on the fiber facet better matches the excitation wavelength. Since the maximum enhancement is confined at the apex of each tip, Raman guided signal can be correlated to a specific position, therefore generating high spatial resolution imaging through the bundle of fibers. This new imaging probe could be used for various applications, such as enhanced Raman endoscopy or remote in vitro detection in complex micro-environments that can integrate such an optical SERS platform. Compared to more technologically demanding approaches using state-of-the-art nanofabrication techniques, our fabrication method relies only on the chemical etching of a fiber followed by metal deposition. Acknowledgements
The authors thank the Agence Nationale pour la Recherche (Programme en Nanosciences et Nanotechnologies ANR-05-NANO-048), the Région Aquitaine and the University Bordeaux 1 for financial support. F.L.L. acknowledges Natural Sciences and Engineering Research Council of Canada for Discovery Grant, as well as the Nanofabrication Facility at Western.
#120087 - $15.00 USD
(C) 2009 OSA
Received 17 Nov 2009; revised 2 Dec 2009; accepted 2 Dec 2009; published 16 Dec 2009
21 December 2009 / Vol. 17, No. 26 / OPTICS EXPRESS 24035
