A highly-sensitive optical fiber surface-enhanced Raman scattering (SERS) sensor has been developed by interference lithography. While one facet of the optical fiber is patterned with silver-coated nanopillar array as a SERS platform, the other end of the probe is used, in a remote end detection, to couple the excitation laser into the fiber and send the SERS signal to the spectrometer. SERS performance of the probe is characterized using trans-1,2-bis(4-pyridyl)-ethylene (BPE) monolayer and an enhancement factor of 1.2 × 107 can be achieved by focusing the laser directly onto the nanopillar array (front end detection). We also demonstrate that this probe can be used for in situ remote sensing of toluene vapor by the remote end detection. Such a fiber SERS probe shows great potential for molecular detection in various sensing applications.
©2012 Optical Society of America
Surface-enhanced Raman scattering (SERS) is a powerful spectroscopic technique for molecular detection due to its high sensitivity and molecular specificity [1, 2]. With the strong electromagnetic field enhancement by the surface plasmon resonance (SPR) of metallic nanostructures and the surface chemical enhancement, SERS can provide a nondestructive and ultrasensitive detection down to a single molecule level [3, 4]. Since its discovery in 1974 , SERS has attracted extensive attention and been widely used in various chemical and biological identifications [6–9].
While SERS provides the molecular “fingerprint” information with the high sensitivity, optical fibers have been used as SERS probes because of their low cost, flexibility, compactness, and remote sensing capability [10–19]. The original single multimode fiber SERS probe was first demonstrated by Mullen et al. in 1991 . In the following studies, different shapes of fiber sensing tips were tested, such as flat and angled fibers [11, 12]. Alternatively, wet chemical etching was utilized to fabricate tapered fiber probes for localized SERS sensing [13, 14]. Metal nanoparticles have been used for the dip coating of the fiber facet as SERS-active substrates as well [15, 16]. And recently, Zhu et al. developed the oblique angle deposition technique to deposit silver nanorod array on the fiber facet for SERS detection in a forward scattering configuration .
To date, several lithography techniques have also been employed for nanopatterning and nanofabrication on the fiber facet. Focused ion-beam lithography (FIB) has been used to define gold nanostructures on the fiber tip but the inadvertent doping with gallium ions altered the optical response of the SERS substrate [20, 21]. For electron-beam lithography (EBL), it is difficult to be directly applied to the optical fiber; therefore, an elegant transfer technique has been developed by Capasso et al. to transfer the EBL-defined metal nanostructures onto the fiber facet . However, this technique is unsuitable for high-throughput fabrication. Nanoimprint lithography (NIL) might be a potential candidate technique for the mass production; but the nanoimprint resist introduces additional coupling loss and Raman background, which will degrade the SERS performance of the fiber probe . Interference lithography (IL) has also been implemented to fabricate waveguide grating structures on the facet of an optical fiber for refractive-index sensing; however, no SERS-active substrate fabricated on the fiber facet with similar approach has ever been reported .
In this paper, we report a highly-sensitive fiber SERS sensor based on IL-defined two-dimensional rectangular array of nanopillars. A high-precision nanofabrication process is developed for the optical fiber with high uniformity and reproducibility, which is well suited for the mass production. With the closely-spaced silver-coated nanopillar array (pitch = 317 nm here), a very high density of SERS “hotspots” can be defined. Using trans-1,2-bis(4-pyridyl)-ethylene (BPE) monolayer as a characterization analyte, we estimate the enhancement factor (EF) of the fiber SERS substrate to be ~1.2 × 107, when the laser is directly focused onto the nanopillar array (front end detection). Furthermore, we demonstrate that this fiber probe, even in remote end detection, has strong signals and enables in situ remote sensing of toluene vapor. This is, to the best of our knowledge, the first time that a single conventional optical fiber SERS probe can be used for vapor/gas detection at room temperature. These results show the great potential of this fiber SERS probe as a highly-sensitive sensor for in situ remote sensing applications.
2.1 Fabrication of fiber SERS probe
To build a highly integrated optical SERS sensing system for practical applications, it is desirable to fabricate the SERS substrate on the facet of an optical fiber. The main challenge during fabrication is how to control the position and orientation of the fiber tip in the processes of spin coating, lithography, etching, and vapor deposition, as the fiber has a small diameter and a large aspect ratio. In this study, we successfully fabricated the nanopillar array on the facet of a regular multimode fiber by mounting the fiber with the help of a fiber ferrule and employing IL to pattern the nanopillar array onto the fiber facet.
The fiber SERS probe was made from a standard silica multimode optical fiber (OFS Fitel, LLC., model: BF06864, NA = 0.22), with a 50 µm core diameter and a 125 μm cladding diameter. The typical fiber length was around 10 cm. A custom-designed ceramic fiber ferrule was attached at one fiber end for the spin coating processes, as shown in Fig. 1(a) . The ferrule had an outer diameter of 3 mm and was 12 mm long. The epoxy gap between the optical fiber and the ferrule was less than 2 μm and the whole area of the ferrule including the fiber was polished with 0.02 µm polishing film. Since this procedure of attaching a fiber ferrule at the fiber end is a standard process in fiber-optic industry, the fiber sample can be easily fabricated at low-cost and high throughput. Photograph and SEM images of the fiber ferrule were shown in Fig. 1(b) and (c), respectively. The fiber facet with the fiber ferrule was first spin coated with a 260 nm thick antireflection layer and then coated with a 700 nm thick photoresist. The nanopillar array was then fabricated onto the fiber facet by IL, using the same method as that in the fabrication of similar array on 4-inch fused silica wafers in our previous studies [24–27]. The laser wavelength for IL was 413 nm and the dose was around 80 mJ/cm2. The resultant photoresist pattern was a two-dimensional periodic nanopillar array with a 317 nm pitch and a 160 nm pillar diameter. An ion milling deep reactive ion etching step was used to remove the antireflection layer between the photoresist nanopillars (mask), after which the unprotected silica area was etched down to 600 nm and then the residual photoresist mask on the top of the nanopillars was washed away. Finally a 60 nm layer of silver was e-beam evaporated at a deposition rate of 0.1 nm/s onto the fiber facet at an angle of 60° to make it SERS-active. We have tested various deposition angles, such as 0°, 30°, and 60°. The 60° deposition angle resulted in the best SERS signal for the remote end detection in the current study.
2.2 SERS measurements of BPE monolayer and toluene vapor
Both BPE powder (purity > 99.9%) and toluene liquid (purity > 99.9%) were obtained commercially through Sigma-Aldrich. For the BPE measurement, it followed a procedure described previously : the patterned fiber facet was submerged in 5 mM BPE solution in methanol for 24 hours, then gently rinsed in methanol, and dried under a stream of nitrogen. For the toluene measurement, the fiber facet was placed above the toluene liquid surface. The toluene vapor was estimated to be at the saturated concentration (2~3%) at room temperature .
For the SERS measurement, two configurations were used in this study. One was the front end detection, in which the laser light was focused directly by the objective lens (20 × , NA = 0.40) onto the patterned fiber facet and the SERS signal was collected via the backscattering geometry. The other was remote end detection, which is the typical optrode geometry for remote sensing: the laser light was coupled from the unpatterned fiber end, propagated through the fiber, and triggered the SERS activity at the remote patterned facet; the SERS signal was then collected from the distal fiber end and coupled back to the Raman spectrometer through the fiber. The SERS measurements were performed using a Renishaw InVia Raman Microscope system and the excitation laser wavelength was 514 nm. All spectra were baseline corrected to remove the broad fiber background.
3. Results and discussion
Figure 2 shows the SEM images of the silver-coated nanopillar array patterned on the fiber core area. It can be seen that the nanopillar array has a high uniformity which allows for the robustness and repeatability of the measurement . Due to the 60° angle deposition, the nanopillars are mainly coated by the silver on one side and the strong “shadowing” effect leads to much less silver on the substrate surface (at the bottom of the nanopillars), which is crucial in reducing the metal absorption for the remote end SERS detection.
For the optical characterization of the fiber SERS probe, reflectance spectra were measured in both the front end and the remote end configurations using Nanospec reflectometer with the white light source illumination. Figure 3 shows that there is a SPR at 558 nm for the front end configuration and a SPR at 537 nm for the remote end configuration. In the remote end configuration, the broader and deeper resonance dip is due to the excitation of the nanopillar array on the whole fiber core area (diameter: 50 µm) instead of only the ones at the focal spot (diameter: 20 µm), while the lower overall reflectivity is mainly due to the coupling loss. There is also a difference in numerical aperture that needs to be accounted for different angles of excitation involved; the numerical aperture changes from 0.40 to 0.22 in the front and remote end configuration, respectively. More crucially, we believe that the difference in the wavelength position of SPR in two configurations is mainly due to the asymmetry of this complex 3D dielectric-metal structure with respect to the excitation (especially in the out-of-plane z-direction). Further modeling to decipher the electromagnetic distribution in both configurations will be carried out in the future. To optimize the enhancement for both the excitation laser and the Raman scattered signal, we have chosen 514 nm laser light for the excitation of the surface plasmon in our experiment.
Figure 4(a) shows the SERS spectra of BPE monolayer in both the front end and the remote end configurations. The laser power was 0.2 mW and the integration time was 10 s. Both plots show the characteristic SERS peaks of BPE at 1200 cm−1 (C = C stretching), 1604 cm−1 (aromatic ring stretching), and 1635 cm−1 (in-plane ring mode), which are consistent with what is reported in the literature . For the remote end configuration, the SERS sensitivity is around 1/5 of that obtained in the front end configuration. The bulk Raman spectrum of 0.1 M BPE solution in methanol is shown in Fig. 4(b). This spectrum is obtained with a laser power of 2 mW and an integration time of 10 s. Besides the three main peaks from BPE, the broad peak at 1456 cm−1 is from the methanol solvent. While the remote end configuration is more useful in sensing applications, characterization from the front end configuration provides a more direct measurement of the enhancement factor (EF) of the SERS substrate itself to give an indication of the quality and performance of our substrate as compared to other technologies. Therefore, in order to characterize the SERS performance of this fiber probe, we estimate the EF based on the front end configuration using the following expression:
Taking the 1200 cm−1 SERS band as a reference, the SERS intensity underneath this band is calculated to be 126157 (counts), while the Raman intensity underneath this band is calculated to be 1246 (counts). We assume a monolayer coverage density for BPE on silver surface to be 1014 molecules/cm2 . The depth of focus is estimated to be ~200 μm for our system. The number of molecules in SERS measurement is the molecular coverage density (1014 molecules/cm2) multiplied by the metal surface area under the laser beam spot area (diameter: ~5 µm). The number of molecules in Raman measurement is the molecular concentration (5 mM) used in the study multiplied by the total interaction volume, which is assumed to be the laser beam spot area multiplied by the depth of focus (~200 μm). Therefore, the EF value is estimated to be 1.2 × 107 for the front end configuration. This value is a little bit lower than that (on the order of ~108) obtained previously on our fused silica substrates [25, 27]. One possible reason is that in order for the remote sensing application, there is a tradeoff between minimizing the metal absorption loss and maximizing the SERS enhancement. Therefore, the metal deposition angle and coating thickness are different from that used previously [25, 27]. Also, the geometry of the nanopillar in this study has not been optimized yet. However, comparing to other existing technologies such as the EBL-transfer technique developed by Capasso et al. which has an EF value of 2.6 × 105~5 × 105 , our probe has a much higher sensitivity as a fiber SERS sensor.
Recently, we have demonstrated that the nanopillar array SERS substrate on fused silica wafer can be used for the toluene vapor detection . Therefore, it is very straightforward to employ this technique on the optical fiber for the remote sensing application. Curve A in Fig. 5 shows the fiber SERS probe used in the remote sensing of the toluene vapor (2~3%). In this experiment, the laser power was 2 mW and the integration time was 10 s. It can be seen that both the 1002 cm−1 and 1597 cm−1 peaks (C = C stretching) are clearly observable. Curve B in Fig. 5 shows a comparison experiment in which an unpatterned optical fiber without silver coating in the toluene vapor with the same configuration cannot detect any signal. And we also tried a patterned fiber without the silver coating and an unpatterned fiber with the silver coating, both of which did not result in any detectable signal.
With the high sensitivity provided by the silver-coated nanopillar array SERS substrate on the fiber tip, this is the first time, to the best of our knowledge, that a single conventional optical fiber SERS sensor can be used for the remote detection of the vapor/gas samples at room temperature. This key component would facilitate the development of a highly-sensitive integrated portable all-fiber based SERS sensing system for remote environmental control in the future work .
In summary, we have successfully demonstrated a highly sensitive fiber SERS probe based on a silver-coated nanopillar array fabricated by using IL. Using the BPE monolayer as a characterization analyte in its SERS measurement, an EF of 1.2 × 107 has been achieved in the front end configuration. We have also demonstrated that this fiber SERS probe can be applied in the remote sensing of the toluene vapor, which shows the great potential as a powerful and compact analytical tool for molecular detection and identification in environmental control. Future work includes pushing the sensitivity of the fiber probe by optimizing the nanopillar geometry and the metal coating strategy.
We acknowledge support from the National Science Foundation (NSF), ECCS-0823921. X.Y. acknowledges financial support by the Lawrence Scholar Program at LLNL. This work was performed under the auspices of the U. S. Department of Energy by LLNL under Contract DE-AC52-07NA27344, LLNL-JRNL-575152. We thank Dr. Bin Chen at NASA Ames Research Center for offering the Raman instrument.
References and links
1. A. Campion and P. Kambhampati, “Surface-enhanced Raman scattering,” Chem. Soc. Rev. 27(4), 241–250 (1998). [CrossRef]
2. K. Kneipp, H. Kneipp, I. Itzkan, R. R. Dasari, and M. S. Feld, “Surface-enhanced Raman scattering and biophysics,” J. Phys. Condens. Matter 14(18), R597–R624 (2002). [CrossRef]
4. K. Kneipp, Y. Wang, H. Kneipp, L. T. Perelman, I. Itzkan, R. R. Dasari, and M. S. Feld, “Single molecule detection using surface-enhanced Raman scattering,” Phys. Rev. Lett. 78(9), 1667–1670 (1997). [CrossRef]
5. M. Fleischmann, P. J. Hendra, and A. J. McQuillan, “Raman spectra of pyridine adsorbed at a silver electrode,” Chem. Phys. Lett. 26(2), 163–166 (1974). [CrossRef]
6. K. T. Carron and B. J. Kennedy, “Molecular-specific chromatographic detector using modified SERS substrates,” Anal. Chem. 67(18), 3353–3356 (1995). [CrossRef]
7. A. Tao, F. Kim, C. Hess, J. Goldberger, R. He, Y. Sun, Y. Xia, and P. Yang, “Langmuir-Blodgett silver nanowire monolayers for molecular sensing using surface-enhanced Raman spectroscopy,” Nano Lett. 3(9), 1229–1233 (2003). [CrossRef]
8. S. Shanmukh, L. Jones, J. Driskell, Y. Zhao, R. Dluhy, and R. A. Tripp, “Rapid and sensitive detection of respiratory virus molecular signatures using a silver nanorod array SERS substrate,” Nano Lett. 6(11), 2630–2636 (2006). [CrossRef] [PubMed]
9. M. S. Schmidt, J. Hübner, and A. Boisen, “Large area fabrication of leaning silicon nanopillars for surface enhanced Raman spectroscopy,” Adv. Mater. (Deerfield Beach Fla.) 24(10), OP11–OP18 (2012). [CrossRef] [PubMed]
10. K. I. Mullen and K. T. Carron, “Surface-enhanced Raman spectroscopy with abrasively modified fiber optic probes,” Anal. Chem. 63(19), 2196–2199 (1991). [CrossRef]
11. C. Viets and W. Hill, “Comparison of fiber-optic SERS sensors with differently prepared tips,” Sens. Actuators B Chem. 51(1-3), 92–99 (1998). [CrossRef]
12. C. Viets and W. Hill, “Single-fiber surface-enhanced Raman sensors with angled tips,” J. Raman Spectrosc. 31(7), 625–631 (2000). [CrossRef]
13. A. Lucotti and G. Zerbi, “Fiber-optic SERS sensor with optimized geometry,” Sens. Actuators B Chem. 121(2), 356–364 (2007). [CrossRef]
14. V. Guieu, P. Garrigue, F. Lagugné-Labarthet, L. Servant, N. Sojic, and D. Talaga, “Remote surface enhanced Raman spectroscopy imaging via a nanostructured optical fiber bundle,” Opt. Express 17(26), 24030–24035 (2009). [CrossRef] [PubMed]
15. X. Yang, C. Gu, F. Qian, Y. Li, and J. Z. Zhang, “Highly sensitive detection of proteins and bacteria in aqueous solution using surface-enhanced Raman scattering and optical fibers,” Anal. Chem. 83(15), 5888–5894 (2011). [CrossRef] [PubMed]
16. G. F. S. Andrade, M. Fan, and A. G. Brolo, “Multilayer silver nanoparticles-modified optical fiber tip for high performance SERS remote sensing,” Biosens. Bioelectron. 25(10), 2270–2275 (2010). [CrossRef] [PubMed]
17. Y. Zhu, R. A. Dluhy, and Y. Zhao, “Development of silver nanorod array based fiber optic probes for SERS detection,” Sens. Actuators B Chem. 157(1), 42–50 (2011). [CrossRef]
18. X. Yang, C. Shi, D. Wheeler, R. Newhouse, B. Chen, J. Z. Zhang, and C. Gu, “High-sensitivity molecular sensing using hollow-core photonic crystal fiber and surface-enhanced Raman scattering,” J. Opt. Soc. Am. A 27(5), 977–984 (2010). [CrossRef] [PubMed]
19. C. Gu, X. Yang, J. Zhang, R. Newhouse, and L. Cao, “Fiber sensors for molecular detection,” Proc. SPIE 7851, 785105, 785105-14 (2010). [CrossRef]
20. A. Dhawan, M. Gerhold, A. Madison, J. Fowlkes, P. E. Russell, T. Vo-Dinh, and D. N. Leonard, “Fabrication of nanodot plasmonic waveguide structures using FIB milling and electron beam-induced deposition,” Scanning 31(4), 139–146 (2009). [CrossRef] [PubMed]
21. 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). [CrossRef] [PubMed]
22. 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). [CrossRef] [PubMed]
23. S. Feng, X. Zhang, H. Wang, M. Xin, and Z. Lu, “Fiber coupled waveguide grating structures,” Appl. Phys. Lett. 96(13), 133101 (2010). [CrossRef]
24. A. Fernandez, H. T. Nguyen, J. A. Britten, R. D. Boyd, M. D. Perry, D. R. Kania, and A. M. Hawryluk, “Use of interference lithography to pattern arrays of submicron resist structures for field emission flat panel displays,” J. Vac. Sci. Technol. B 15(3), 729–735 (1997). [CrossRef]
25. M. R. Gartia, Z. Xu, E. Behymer, H. Nguyen, J. A. Britten, C. Larson, R. Miles, M. Bora, A. S. P. Chang, T. C. Bond, and G. L. Liu, “Rigorous surface enhanced Raman spectral characterization of large-area high-uniformity silver-coated tapered silica nanopillar arrays,” Nanotechnology 21(39), 395701 (2010). [CrossRef] [PubMed]
26. M. Bora, B. J. Fasenfest, E. M. Behymer, A. S. P. Chang, H. T. Nguyen, J. A. Britten, C. C. Larson, J. W. Chan, R. R. Miles, and T. C. Bond, “Plasmon resonant cavities in vertical nanowire arrays,” Nano Lett. 10(8), 2832–2837 (2010). [CrossRef] [PubMed]
27. A. S. P. Chang, M. Bora, H. T. Nguyen, E. M. Behymer, C. C. Larson, J. A. Britten, J. C. Carter, and T. C. Bond, “Nanopillars array for surface enhanced Raman scattering,” Proc. SPIE 8024, 80240I, 80240I-8 (2011). [CrossRef]
28. A. Gutes, I. Laboriante, C. Carraro, and R. Maboudian, “Silver nanostructures on silicon based on galvanic displacement process,” J. Phys. Chem. C 113(39), 16939–16944 (2009). [CrossRef]
29. A. Kim, F. S. Ou, D. A. A. Ohlberg, M. Hu, R. S. Williams, and Z. Li, “Study of molecular trapping inside gold nanofinger arrays on surface-enhanced Raman substrates,” J. Am. Chem. Soc. 133(21), 8234–8239 (2011). [CrossRef] [PubMed]
30. X. Yang, Z. Tanaka, R. Newhouse, Q. Xu, B. Chen, S. Chen, J. Z. Zhang, and C. Gu, “Portable fiber sensors based on surface-enhanced Raman scattering,” Rev. Sci. Instrum. 81(12), 123103 (2010). [CrossRef] [PubMed]