Open Access
Add to BibSonomy
Abstract
Fiber-optics based surface-enhanced Raman scattering (FO-SERS) has an unique advantage of being able to remotely detect analyte molecules because the fiber length can be adjusted as desired. However, the Raman signal of the fiber-optic material is so strong that it is an important challenge in utilization of optical fiber for remote SERS sensing. In this study, we found that the background noise signal was greatly reduced by ca. 32% compared to conventional fiber-optics with a flat surface cut. To confirm the feasibility of FO-SERS detection, silver nanoparticles labeled with 4-fluorobenzenethiol were attached onto the end surface of an optical fiber to form a SERS-signaling substrate. The SERS intensity from the fiber-optics with a roughened surface as SERS substrate was increased significantly with respect to signal-to-noise ratio (SNR) values compared to optical fibers with flat end surface. This result implies that the fiber-optics with roughened surface could be used as an efficient alternative for FO-SERS sensing platform.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Surface-enhanced Raman scattering (SERS) is an optical phenomenon that enhances the Raman signal of an analyte owing to localized surface plasmon resonance (LSPR) generated by metal nanoparticles such as gold and silver when light is irradiated. In particular, when the analyte molecule is located between metal nanoparticles called hot-spots, a significant increase in the Raman signals of the molecule occurs [1–4]. Benefiting such a high sensitivity of SERS, it has been applied for detection of pathogens [5–8], pesticides [4,9,10], and other dangerous materials [11–13]. When fiber optic is used for SERS detection (FO-SERS), the length of the optical fiber can be adjusted as desired, allowing the detection of distant analytes while keeping the high sensitivity of SERS [14–16]. Therefore, it can be useful for the remote detection of toxic or potentially hazardous analytes [17,18]. In addition, the optical alignment in fiber-optic Raman equipment is easier than the conventional Raman equipment using a microscope due to the merits of optical fiber such as long delivery of lights with little loss during the passage and flexibility of light passage.
However, most FO-SERS research to date still uses conventional confocal Raman measurement equipment [17,19–22] since the strong background signal of the optical fiber itself strongly interferes with the Raman signal of the sample at the fiber end surface [23]. Looking at closely the fiber-optic SERS measurement, we focus on three parts of light-material interaction. First one is the optical fiber material itself through the beam delivery, which generates noise signals depending on the length of the optical fiber. The second one is the reflection at the end surface of optical fiber, which part of light does not reach to the SERS-active nanoparticle but contributes to noise signal. The third one is effective collection of SERS light from SERS-active nanoparticles radiating lights random direction. Therefore, minimizing the optical noise of the optical fiber and maximizing the optical signal is an important issue in FO-SERS for practical use in potential applications.
In this study, we fabricated the surface of optical fibers by different treatments to lower the background signal of the optical fiber itself and optimize the signal-to-noise ratio (SNR) values for an efficient fiber-optic SERS sensing platform. The FO-SERS advantage of simple alignment was also utilized by constructing SERS measurement equipment consisting of optical fibers only.
2. Experimental section
2.1 Chemicals and materials
Tetraethyl orthosilicate (TEOS), 3-mercaptopropyltrimethoxysilane (MPTS), ethylene glycol (EG), silver nitrate (AgNO3), 4-fluorobenzenethiol (4-FBT), 11-mercaptoundecanoic acid (11-MUA) and hexadecylamine were purchased from Sigma Aldrich (St. Louis, MO, USA). Ethanol (99.5%) and ammonium hydroxide (NH4OH, 28-30%) were purchased from Daejung (Siheung, Korea). Multimode optical fibers with core/cladding diameters of 105/125 $\mathrm{\mu m}$ were purchased from Thorlabs (Newton, NJ, USA).
2.2 Fabrication of fiber-optic end surfaces by different surface cutting
The surface cutting of optical fiber was done in three different ways. A flat surface was fabricated by cutting with a commercial optical fiber cleaver. An angled surface was fabricated by cutting with a specific angle at the end of the optical fiber using a modified cleaver to tilt the fiber at the desired angle. A rough surface fiber was prepared using a fiber jacket stripper to physically create random roughness at the end surface of the optical fiber. Figure 1 shows a schematic of optical fibers with different end surfaces.
2.3 Preparation of sample fiber
The optical fiber of which the end surface was treated with silver nanoparticles as SERS substrate was prepared and called as the sample fiber. The end surface was treated by adsorbing silver nanoparticles (AgNS) labeled with 4-FBT. AgNS was synthesized by a previously reported method [24]. Briefly, using the Stöber method [25], we prepared silica nanoparticle of ca. 150 nm in diameter. After modifying the silica core surface with a thiol group using MPTS, a silica core surrounded with silver nanoparticles was synthesized by adding AgNO3, EG and hexadecylamine. After that, 4-FBT was added as a Raman label molecule to the silver nanoparticle dispersion to finally form SERS-signaling silver nanoparticles (AgNS4-FBT). The sample fiber was prepared by repeatedly dropping and drying 1.5 $\mathrm{\mu}\textrm{L}$ of the AgNS4-FBT particle dispersion on the end of the optical fiber three times.
2.4 FO-SERS measurement scheme composed of optical fiber
For SERS measurements, a 785-nm laser-line (Cobolt 08 NLD, Cobolt, Sweden) was used as a photo-excitation source, and the sample fiber was irradiated through a 2-by-1 fiber coupler (BM Laser, Korea). The laser power was 40 mW at the end of the sample fiber, and the acquisition time at each measurement was 5 s. A spectrometer (SR-303i-A, Andor Technology, UK) equipped with a CCD (DV401A-BV, Andor Technology, UK) and a customized optical fiber-spectrometer coupling module (Dongwoo Optron, Korea) collected the lights scattered from the sample fiber modified as SERS substrates and the noise from the optical fiber itself. Figure 2 shows a photo-image of the SERS measurement equipment and a schematic diagram of the signal collection process.
Fig. 2. SERS measurement equipment composed of an optical fiber. (a) Photo-image of the Raman scattering measurement setup. (b) Schematic diagram of optical component alignment for collecting light from the sample optical fiber.
3. Results and discussion
3.1 Characterization of optical fibers with different end surfaces
Each fiber-optic was fabricated by cutting the end surface differently, as shown in Supplement 1, Fig. S1. The Raman spectra obtained from different optical fiber surfaces represent optical noise patterns caused by the interaction of laser lights and the optical fiber material itself through light passage (Fig. 3(a)).
Fig. 3. Raman signals from the optical fiber itself under an excitation of 785 nm. (a) Raman spectra of optical fibers with different end surfaces. (b) Signal reproducibility of rough-surface fiber fabrication.
The black line is the spectrum of light scattered by a flat surface fiber, the red one is that of a 10°-angled surface, the blue one is that of a 20°-angled surface, and the magenta one is that of a rough surface. It was observed that the noise signal decreased as the angle of the fiber end increased. In the case of the rough surface fiber, the Raman signals of the optical fiber itself was reduced than the flat surface fiber for the entire measured area, as shown in Supplement 1, Fig. S2. However, despite decreased signal intensity, it still shows strong Raman signals in areas below 1000 cm-1, so only after 1000 cm-1 was presented as a detection window. The noise signal at the 1042 cm-1 band, the highest intensity in the Fig. 3(a), decreased by ca. 32% in the rough surface fiber compared to the flat surface fiber.
Since the rough cutting method is simply hand-cutting with a fiber jacket stripper, it was necessary to verify the reproducibility of the cutting process. Figure 3(b) presents the noise intensity at 1042 cm-1 in the Raman spectra of eight different rough-surface fiber samples. As an indicator of signal reproducibility, the coefficient of variability (CV) is calculated as follows:
The average and the standard deviation of the signal intensity at 1042 cm-1 were 4244.4 and 40.1 counts, respectively. Therefore, the CV value was 0.9%, confirming excellent reproducibility of the rough-cutting method. The change of rough-cutting angle may be considered for better performance. However, the degree of local cut angle in rough cut surface is so large as tens of degree (Fig. 4 and Supplement 1, Fig. S1) that the angle effect seems to be included already in the rough cutting while the angle control of rough-cut surface is not practically possible. The decrease in the noise signal from the optical fiber itself in the rough surface fiber can be explained by the relationship between surface roughness and specular reflectance [26]. The intensity of light reflected from the rough surface is expressed as follows:
Fig. 4. Comparison of a flat surface fiber and rough surface fiber. (a) Scanning electron microscopy images of fiber-optic end surfaces. The left image is a flat surface, and the right image is a rough surface. (b) Schematic diagram of the intensity of light reflected from the optical fiber surface. The left diagram is the reflection from the flat surface and the right diagram is the reflection from the rough surface.
${R_0}$ is the reflected light intensity on the smooth surface, and ${R_s}$ is the reflected light intensity on the rough surface of the same material. When the laser is irradiated through an optical fiber, the light intensity at the rough surface decreases exponentially with the surface roughness factor ($\mathrm{\sigma }$).
As shown in Fig. 4(a), the flat surface fiber had a smooth surface, and the rough surface fiber had a roughened surface. The decrease in the amount of reflected light due to surface roughness means that the light reflected at the interface of the optical fiber end and the air was decreased, and presumably, more light could interact with the nanoparticles attached to the fiber surface. However, this needs to be verified by SERS measurements.
3.2 SERS sensitivity of sample fibers with different end surfaces
To verify the efficiency of different optical fiber-cuts with respect to SERS measurements, well-confirmed SERS-active nanoparticles were employed. As shown in Supplement 1, Fig. S2(a), it was confirmed that silver nanoparticles (AgNS) formed a shell around the silica core and 4-FBT labelled silver nanoparticles (AgNS4-FBT) were synthesized as intended. The SERS activity of AgNS4-FBT particles was confirmed by measuring their SERS spectrum using a micro-Raman equipment, as shown in Supplement 1, Fig. S2(b). Since the SERS signal of the 4-FBT molecule was strong enough, AgNS4-FBT particles were loaded at the end-surface of optical fibers as SERS substrates. The SERS spectra obtained by measuring the fiber samples with different end surfaces are shown in Fig. 5(a). All spectra are presented by subtracting them from the Raman spectrum of a bare fiber sample.
Fig. 5. Signal-to-Noise Ratio (SNR) comparison of fiber optics with different end shapes. (a) SERS spectra of different end-shaped optical fibers after subtracting each bare fiber signal. (b) Scatter plot of the SNR of different end-shaped optical fibers. (SERS signal was based on the 1072 cm-1 peak in Fig. 5(a), and the noise was calculated as the standard deviation in a range from 1515 to 1535 cm-1, which was a flat section in all sample spectra in Fig. 5(a).)
The black line is for a flat surface, the red line is for a 10°-angled surface, the blue line is for a 20°-angled surface, and the magenta line is for a rough surface. All SERS spectra depict the most prominent band of 4-FBT at 1072 cm-1. The rough surface fiber and the 20° angled fiber showed weaker bands of 4-FBT appearing at 1486 and 1583 cm-1. As the cutting angle increased, the SERS signal increased and was the highest for the rough-cut surface. To compare the sensitivity as a SERS sensing platform, we adopted the SNR value rather than Raman intensity since the effective photo-excitation power at the SERS-active nanoparticles could not be measured due to different reflectance at the fiber end surface, even though the external illumination laser power could be controlled. The SNR value is calculated as follows:
Ianalyte is the observed intensity of the SERS band at 1072 cm-1, and Inoise is the standard deviation at the region from 1515 to 1535 cm-1, where the Raman signal does not appear in Fig. 5(a). As shown in Fig. 5(b), the SNR value increased as the tip angle increased, while the rough-surface fiber showed a dramatic increase in the SNR value by ca. 27 times compared to the flat-surface fiber. This amount of SNR increase cannot be explained by the 32% decrease in the background signal from the flat surface fiber to the rough-surface fiber. The 32% decrease in the background signal simply implies a 32% increase in laser power delivered to the SERS-active nanoparticles attached to the optical fiber end.
However, the intensity at the 1072 cm-1 SERS band was increased by ca. 9 times by rough-cutting, as shown in Fig. 6. Such an unexpected increase in SERS intensity could be due to an increased light collection efficiency. By considering light passage near the optical fiber end, light collection efficiency can be approximately estimated for the flat-surface fiber. The collection efficiency was defined as the solid angle of light passing through the optical fiber compared to the Raman scattering occurring in all-directions at the end of the optical fiber. If the numerical aperture (NA) of the optical fiber is known, the collection efficiency can be calculated, and the scattered angle of the optical fiber can be found by estimating conversely. The scheme and a more detailed description of the calculation are described in Supplement 1 (Fig. S3). The collection efficiency was calculated as 1.2% for the flat-surface fiber using the NA of the optical fiber we used. Using the obtained increase in SERS intensity for the rough-surface fiber, the collection efficiency of the rough-surface fiber was estimated to be 8.5%. This means that rough-surface fiber increased the collection efficiency as the scattered angle of light, which can propagate within the optical fiber, was increased from 13 to 34 degrees. The collaboration between increasing the laser power delivered to the SERS-active nanoparticles and increasing the collection efficiency seems to have resulted in a 9-fold signal enhancement. The increase in the SNR from the sample fiber with a roughened surface was due to an increase in the SERS signal intensity itself, as well as a decrease in the Raman signal of the bare rough surface fiber. Collection efficiency increased due to surface roughness, resulting in an increase in SERS intensity, which led to an increase in the SNR. This shows that the rough-surface fiber as an FO-SERS sensing platform can achieve high sensitivity.
4. Conclusions
We observed enhanced the SNR value in FO-SERS measurements by rough-cutting the end of optical fiber. Due to the randomly roughened surface of the fiber, the noise signal of the optical fiber itself was reduced by about 32% compared to the flat-surface fiber. In the case of SERS measurements using the rough-surface fiber as a SERS sensing platform, the SNR values were calculated to be the highest compared to the other types of tip-shaped fibers. Especially, about 27 times higher values were obtained than those from the flat-surface fiber. Since SNR enhancement by reducing the Raman signals of the optical fiber itself is an important challenge in FO-SERS research, this study is expected to be utilized in various FO-SERS fields.
Funding
National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT), (NRF-2021M3C1C3097205, 2021R1A45031762); 2021 Cultural Heritage Smart Preservation & Utilization R&D Program from the Cultural Heritage Administration National Research Institute of Cultural Heritage (Project Name: Development of in-situ analysis and diagnosis, deterioration prediction technology of organic colorants) (2021A01D02-001); and Korea Institute of Planning and Evaluation for Technology in Food, Agriculture, Forestry and Fisheries (IPET) through Crop Viruses and Pests Response Industry Technology Development Program, funded by the Ministry of Agriculture, Food and Rural Affairs (321107-03).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
Supplemental document
See Supplement 1 for supporting content.
References
1. J. Yoon, N. Choi, J. Ko, K. Kim, S. Lee, and J. Choo, “Highly sensitive detection of thrombin using SERS-based magnetic aptasensors,” Biosens. Bioelectron. 47, 62–67 (2013). [CrossRef]
2. 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]
3. B. N. Khlebtsov, V. A. Khanadeev, E. V. Panfilova, D. N. Bratashov, and N. G. Khlebtsov, “Gold Nanoisland Films as Reproducible SERS Substrates for Highly Sensitive Detection of Fungicides,” ACS Appl. Mater. Interfaces 7(12), 6518–6529 (2015). [CrossRef]
4. C. Song, Y. Wang, Y. Lei, and J. Zhao, “SERS-Enabled Sensitive Detection of Plant Volatile Biomarker Methyl Salicylate,” J. Phys. Chem. C 126(1), 772–778 (2022). [CrossRef]
5. Y. Yang, G. Li, P. Wang, L. Fan, and Y. Shi, “Highly sensitive multiplex detection of foodborne pathogens using a SERS immunosensor combined with novel covalent organic frameworks based biologic interference-free Raman tags,” Talanta 243, 123369 (2022). [CrossRef]
6. H. Kearns, R. Goodacre, L. E. Jamieson, D. Graham, and K. Faulds, “SERS Detection of Multiple Antimicrobial-Resistant Pathogens Using Nanosensors,” Anal. Chem. 89(23), 12666–12673 (2017). [CrossRef]
7. R. M. Jarvis and R. Goodacre, “Characterisation and identification of bacteria using SERS,” Chem. Soc. Rev. 37(5), 931–936 (2008). [CrossRef]
8. K. Gracie, E. Correa, S. Mabbott, J. A. Dougan, D. Graham, R. Goodacre, and K. Faulds, “Simultaneous detection and quantification of three bacterial meningitis pathogens by SERS,” Chem. Sci. 5(3), 1030–1040 (2014). [CrossRef]
9. Z. Zhang, T. Si, J. Liu, and G. Zhou, “In-Situ Grown Silver Nanoparticles on Nonwoven Fabrics Based on Mussel-Inspired Polydopamine for Highly Sensitive SERS Carbaryl Pesticides Detection,” Nanomaterials 9(3), 384 (2019). [CrossRef]
10. J. Jiang, S. Zou, L. Ma, S. Wang, J. Liao, and Z. Zhang, “Surface-Enhanced Raman Scattering Detection of Pesticide Residues Using Transparent Adhesive Tapes and Coated Silver Nanorods,” ACS Appl. Mater. Interfaces 10(10), 9129–9135 (2018). [CrossRef]
11. A. Hakonen, P. O. Andersson, M. Stenbaek Schmidt, T. Rindzevicius, and M. Käll, “Explosive and chemical threat detection by surface-enhanced Raman scattering: A review,” Anal. Chim. Acta 893, 1–13 (2015). [CrossRef]
12. R. Gao, X. Song, C. Zhan, C. Weng, S. Cheng, K. Guo, N. Ma, H. Chang, Z. Guo, L.-B. Luo, and L. Yu, “Light trapping induced flexible wrinkled nanocone SERS substrate for highly sensitive explosive detection,” Sens. Actuators, B 314, 128081 (2020). [CrossRef]
13. R. Gao, H. Qian, C. Weng, X. Wang, C. Xie, K. Guo, S. Zhang, S. Xuan, Z. Guo, and L.-B. Luo, “A SERS stamp: Multiscale coupling effect of silver nanoparticles and highly ordered nano-micro hierarchical substrates for ultrasensitive explosive detection,” Sens. Actuators, B 321, 128543 (2020). [CrossRef]
14. L. Li, S. Deng, H. Wang, R. Zhang, K. Zhu, Y. Lu, Z. Wang, S. Zong, Z. Wang, and Y. Cui, “A SERS fiber probe fabricated by layer-by-layer assembly of silver sphere nanoparticles and nanorods with a greatly enhanced sensitivity for remote sensing,” Nanotechnology 30(25), 255503 (2019). [CrossRef]
15. Z. Chen, Z. Dai, N. Chen, S. Liu, F. Pang, B. Lu, and T. Wang, “Gold Nanoparticles-Modified Tapered Fiber Nanoprobe for Remote SERS Detection,” IEEE Photonics Technol. Lett. 26(8), 777–780 (2014). [CrossRef]
16. G. F. 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]
17. T. B. Pham, T. H. C. Hoang, V. C. Nguyen, D. C. Vu, H. Bui, and V. H. Pham, “Improved versatile SERS spheroid end-facet optical fiber substrate based on silver nano-dendrites directly planted with gold nanoparticles using dual-laser assisted for pesticides detection,” Opt. Mater. (Amsterdam, Neth.) 126, 112196 (2022). [CrossRef]
18. Y. Long, H. Li, W. Wang, X. Yang, and Z. Liu, “Ultrasensitive detection of Cr(VI) using a novel SERS optical fiber probe modified by dual-functional methimazole,” J. Alloys Compd. 910, 164916 (2022). [CrossRef]
19. J. Zhang, S. Chen, T. Gong, X. Zhang, and Y. Zhu, “Tapered Fiber Probe Modified by Ag Nanoparticles for SERS Detection,” Plasmonics 11(3), 743–751 (2016). [CrossRef]
20. G. Quero, G. Zito, S. Manago, F. Galeotti, M. Pisco, A. C. De Luca, and A. Cusano, “Nanosphere Lithography on Fiber: Towards Engineered Lab-On-Fiber SERS Optrodes,” Sensors 18(3), 680 (2018). [CrossRef]
21. Y. Ran, P. Strobbia, V. Cupil-Garcia, and T. Vo-Dinh, “Fiber-optrode SERS probes using plasmonic silver-coated gold nanostars,” Sens. Actuators, B 287, 95–101 (2019). [CrossRef]
22. Y. Long, H. Li, X. Yang, Y. Yuan, and M. Zheng, “Controlling silver morphology on a cramped optical fiber facet via a PVP-assisted silver mirror reaction for SERS fiber probe fabrication,” New J. Chem. 45(8), 4004–4015 (2021). [CrossRef]
23. J. May and Y.-S. Li, “Fiber Raman background study and its application in setting up optical fiber Raman probes,” Appl. Opt. 35(15), 2527–2533 (1996). [CrossRef]
24. H. Kang, T. Kang, S. Kim, J.-H. Kim, B.-H. Jun, J. Chae, J. Park, D.-H. Jeong, and Y.-S. Lee, “Base Effects on Fabrication of Silver Nanoparticles Embedded Silica Nanocomposite for Surface-Enhanced Raman Scattering (SERS),” J. Nanosci. Nanotechnol. 11(1), 579–583 (2011). [CrossRef]
25. W. Stöber, A. Fink, and E. Bohn, “Controlled growth of monodisperse silica spheres in the micron size range,” J. Colloid Interface Sci. 26(1), 62–69 (1968). [CrossRef]
26. H. E. Bennett and J. O. Porteus, “Relation Between Surface Roughness and Specular Reflectance at Normal Incidence,” J. Opt. Soc. Am. 51(2), 123–129 (1961). [CrossRef]
