Open Access
Add to BibSonomy
Abstract
In order to enhance the sensitivity, integration, and practical application capability of Raman detection systems, we propose a multi-channel microfluidic integrated D-shaped optical fiber SERS (Surface-enhanced Raman scattering) probe structure. Firstly, a microfluidic polydimethylsiloxane (PDMS) channel was fabricated using a self-designed multi-channel microfluidic template. Secondly, a uniform layer of silver nanoparticles was deposited on the D-shaped optical fiber using the liquid-liquid interface method. Finally, the D-shaped optical fiber was plasma-bonded to the multi-channel microfluidic channel and a cover glass, resulting in a microfluidic integrated D-shaped optical fiber SERS probe. The prepared sample exhibited excellent detection performance for R6G (rhodamine 6 G) with a detection limit as low as 10−11 mol/L and an enhancement factor of 1.14 × 109. Moreover, the multi-channel configuration enables simultaneous detection of multiple molecules, demonstrating excellent multi-channel multiplexing capability.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Surface-enhanced Raman scattering spectroscopy is a powerful vibrational spectroscopic technique based on the non-elastic scattering of incident photons by molecules [1]. It possesses outstanding features such as ultra-sensitivity, non-destructiveness, rapidity, and fingerprint identification, making it highly valuable in the fields of analysis and sensing [2–4].
Since the discovery of the Raman signal enhancement phenomenon by Fleischmann et al. in 1974 [5], scientists have been dedicated to studying various forms of plasmonic SERS nano substrates for enhancing Raman signals. For instance, Hu et al. [6] constructed a high-performance, stable, and uniform Au@AgNRs flexible film SERS substrate, which retained 98.6% of its SERS performance even after 90 days of storage. Sha et al. [7] utilized a seed-mediated growth method to prepare silver nanoparticles with different sizes and then deposited them onto a polydimethylsiloxane (PDMS) film, resulting in a controllable and highly homogeneous SERS substrate for the detection of low concentrations of R6G. Napatsakorn et al. [8] employed thiol-functionalized aptamer-modified gold nanoparticles as SERS substrates for the specific detection of glyphosate.
Compared to the aforementioned substrates, fiber-based SERS offers unique advantages such as suitability for complex environmental detection, high biocompatibility, small size, and remote signal transmission [9]. Among the various fiber-based SERS probe structures, D-shaped optical fiber SERS probes have gained considerable attention due to their large active SERS area, low cost, and ease of fabrication [10,11]. For instance, Yin et al. [12] fabricated D-shaped optical fiber SERS probes by femtosecond laser etching and chemically depositing AgNPs, which could detect R6G at a concentration of 10−7 mol/L. Liu et al. [13] prepared D-shaped optical fiber SERS probes through side polishing and chemical deposition of AgNPs. They collected Raman signals perpendicular to the D-shaped surface and achieved a detection limit of 10−9 mol/L for Methylene blue (MB).
Microfluidic technology is an integrated technique that allows for the analysis and manipulation of substances within microchannel. It offers advantages such as small size and low sample volume requirement. Properly designed microfluidic chips can enhance the efficiency of substance detection [14,15]. The combination of SERS and microfluidic creates a hybrid system with fast detection speed, high sensitivity, good reproducibility, and high integration [16]. Consequently, many researchers have integrated fiber-based SERS technology with microfluidic to develop a range of microfluidic fiber-based SERS probes. For instance, Li et al. [17] fabricated concave double-clad fiber SERS probes and embedded them into a microfluidic chip, creating an all-fiber Raman scattering light detection system. This system achieved a detection limit of 10−5 mol/L for R6G. Fan et al. [18] integrated end-face fiber SERS probes into a microfluidic chip, achieving a detection sensitivity of 10−10 mol/L for R6G.
Although the aforementioned microfluidic fiber-based SERS probes have achieved low-concentration detection of substances, their detection limits are limited due to the method of metal particle enrichment and the relatively small SERS active area. In this study, combining the unique advantages of the D-shaped optical fiber, we employed the seed-mediated growth method and the liquid-liquid interface method [19] to prepare a D-shaped optical fiber@AgNPs composite structure. microfluidic channels were fabricated using a molding method, and a microfluidic integrated D-shaped optical fiber SERS probe was prepared through plasma surface bonding. The sensitivity and repeatability of the basal Raman detection were investigated in this study.
2. Experiments
2.1 Preparation of D-shaped fiber
D-shaped optical fibers were prepared using a grinding method [20]. The specific steps are as follows. Firstly, a 45 cm multimode fiber with a core/cladding diameter of 62.5/125 µm was taken. The fiber coating was removed by fiber stripping pliers for approximately 3 cm, and then it was cleaned with anhydrous ethanol to maintain cleanliness. Next, the fiber was placed in a V-groove (width: 144 µm, height: 125 µm, angle: 60°) and secured with UV adhesive. Finally, it was carefully ground using 0.2 µm roughness sandpaper until the fiber reached the same height as the glass V-groove. At this moment, the remaining cross-section of the optical fiber in the V-groove is exactly shaped like a semi-circle (D-shape), resulting in a D-shaped fiber as shown in Fig. 1(a).
Fig. 1. Experimental process flowchart. (a) Preparation process of DSF-AgNPs composite structure; (b) fabrication process of microfluidic integrated D-shaped fiber SERS probe; (c) photograph of the fabricated sample.
2.2 Preparation of D-shaped fiber@AgNPs composite structure
The preparation process of the D-shaped fiber@AgNPs composite structure (DSF-AgNPs) is shown in Fig. 1(a). Firstly, AgNPs with a diameter of approximately 100 nm were synthesized using a seed-mediated growth method and transferred onto the interface [19]. Note that Ag nanoparticle with the diameter of 100 nm has a better LSPR in our samples [20]. Then, the prepared D-shaped fiber was inserted into the liquid surface at a 45°angle, raised vertically, and dried to obtain the composite structure of D-shaped fiber and AgNPs.
In Fig. 1(a), we characterized the prepared composite structure using Scanning Electron Microscope (SEM) and performed particle size statistics. It can be observed that the surface of our prepared sample is covered with a layer of silver nanoparticles, which have a uniform size distribution with an average diameter of approximately 92 nm.
2.3 Preparation of PDMS microchannel
The PDMS microchannel was fabricated using the molding method, as illustrated in Fig. 1(b). Firstly, PDMS prepolymer and curing agent (provided by Suzhou Wenhao microfluidic Technology Co., Ltd.) were mixed in a ratio of 10:1 and thoroughly blended to remove air bubbles. Then, the mixture was poured onto a mold (provided by Suzhou Wenhao microfluidic Technology Co., Ltd.) with a length of approximately 5 cm, width of approximately 3 cm, channel width of 180 µm, and height of 250 µm. The inlet and outlet were punctured with a diameter of 1 mm, and the detection area had an area of 2.25 cm2. The PDMS microchannel was cured on the mold and subsequently peeled off gently to obtain the final PDMS microchannel.
2.4 Preparation of microfluidic integrated D-shaped fiber SERS probe
Sample preparation was performed using the plasma surface bonding method [21]. The specific procedure is as follows: Firstly, the glass slide and PDMS channel were placed together in a plasma cleaner for plasma treatment. After treatment, the prepared DSF-AgNPs composite structure was carefully placed on the glass slide, as shown in step ① of Fig. 1. Then, the PDMS channel was placed on the glass slide with DSF-AgNPs, as shown in step ② of Fig. 1, to complete the bonding. After bonding, a small amount of PDMS mixture was added to seal the contact area between the PDMS microchannel and the fiber on both sides. The inlet and outlet of the microchannel were connected using a needle with an inner diameter of 0.8 mm and an outer diameter of 1 mm, as well as a PTFE tube with an inner diameter of 0.8 mm and an outer diameter of 1 mm. The prepared sample is shown in Fig. 1(c). With strict process control, the microfluidic-integrated D-shaped optical fibers we prepared exhibit excellent consistency.
2.5 Materials and characterization equipment
The glass slide and PDMS microchannel were subjected to surface modification using a plasma cleaner (TS-PL02, Shenzhen Dongxin Gaoke Automation Equipment Co., Ltd.). The morphology of the samples was characterized using a scanning electron microscope (SEM) (Quattro S, Thermo Fisher Scientific). Raman testing experiments were conducted using a confocal Raman spectrometer (LabRAM HR Evolution, Horiba Jobin Yvon S.A.S). Due to the absorption spectrum of different AgNPs and their LSPR, we chose 532 nm laser as the incident light [22]. The output power of excite laser source is 50 mW with a 2.5% filter and an integration time of 10 seconds, in order to avoid thermal effects. The LapSpec software was used to remove the baseline from the raw Raman spectra.
3. Results and discussion
3.1 SEM characterization and electromagnetic field simulation calculations
We characterized the prepared DSF-AgNPs composite structure using SEM, and the results are shown in Fig. 2(a). Statistical analysis of the gaps between silver nanoparticles revealed an average particle spacing of approximately 11.3 nm. Furthermore, to investigate the Raman enhancement mechanism of microfluidic-integrated D-shaped fiber SERS probes, we employed FDTD calculations to determine the spatial distribution of the electromagnetic field of silver nanoparticles in different regions.
Fig. 2. Theoretical analysis. (a) SEM characterization and gap statistics of probe surface; (b) simulation model; (c)(d)(e) E-field simulation of regions 1, 2, and 3 at different incident angles; (f) statistical analysis of E-field intensity for the two types of structures.
When light propagates inside the D-shaped optical fiber, the main factor contributing to the Raman enhancement is the electric field enhancement generated between the metal nanoparticles by the leakage of light outside the fiber. Therefore, a structure as shown in Fig. 2(b) can be constructed, where the vibration direction is in the xoz plane, and the wavelength of the incident light is 532 nm, and the angle θ is less than the total internal reflection critical angle (43.6° in the figure). Considering the statistical analysis of the silver nanoparticles size and gap, we set the silver nanoparticles size to be 92 nm and the gap to be 11.3 nm. Simulations were performed on the representative structures of silver nanoparticles within regions 1, 2, and 3 for five cases: θ = 0°, θ = 10°, θ = 20°, θ = 30° and θ = 40°. The results, shown in Figs. 2(c) (d) and (e), indicate a significant electric field enhancement between the silver nanoparticles in the simulations, with a maximum electric field (E-field) amplitude of 37 V/m.
We also conducted relevant simulations for other diameters and gaps. Combining the information from Figs. 2(c) (d) and (e), it can be observed that the structure with two spheres and an incident angle of θ = 0° exhibits the maximum electric field. Therefore, we chose this case to construct two types of structures: “0°-92 nm-x nm” and “0°-y nm-11.3 nm”. We set x to be 4, 8, 11.3, 14, and 18 nm, and y to be 50, 70, 92, 110, and 130 nm for simulation (x represents the gap, and y represents the diameter). The results are shown in Fig. 2(f).
We use the enhancement factor (EF) based on electro-magnetic enhancement theory to characterize the Raman enhancement effect. Its calculation expression is as follows.
3.2 Raman characterization
3.2.1 Detection limit
To evaluate the sensitivity of the microfluidic-integrated D-shaped fiber SERS probe, we prepared R6G solutions with concentrations ranging from 10−7 mol/L to 10−11 mol/L. The solutions were sequentially injected into the probe channel in increasing order of concentration. After filling the detection area with the R6G solution, it was left undisturbed for two minutes before performing Raman measurements. Figure 3(a) shows the detection results for R6G concentrations ranging from 10−7 mol/L to 10−10 mol/L, where each Raman characteristic peak is clearly visible. Figure 3(b) displays the Raman detection result for the 10−11 mol/L R6G concentration, indicating that the Raman characteristic peaks at 611 cm−1 and 1650 cm−1 are more pronounced. Which demonstrates that the microfluidic-integrated D-shaped fiber SERS probe can still detect R6G at a concentration of 10−11 mol/L.
Fig. 3. Raman detection results of different probe molecules using the microfluidic-integrated D-shaped optical fiber SERS probe. (a) Concentrations ranging from 10−7 mol/L to 10−10 mol/L; (b) concentration of 10−11 mol/L; (c) multi-molecule detection results.
We employ the analytical enhancement factor (AEF) to characterize the Raman enhancement effect of our experiments, and its calculation expression is as follows.
where ISERS and cSERS represent the arbitrary characteristic peak intensity of the prepared SERS substrate under the detection limit and the corresponding concentration of the sample at that time. IRS and cRS represent the Raman signal intensity of the aforementioned corresponding characteristic peaks obtained by directly detecting the sample on a bare silicon wafer and the sample concentration. Since this study focuses on the 611 cm−1 peak, and ISERS = 107.3, cSERS = 10−11 mol/L, IRS = 94.1, and cRS = 10−2 mol/L, therefore, through calculation, the sample's AEF is determined to be 1.14 × 109.It can be observed that there is a certain discrepancy between the actual AEF of the sample and the theoretical results from the simulation analysis. The main reasons for this are as follows.
The simulation results only reflect the contribution of electromagnetic field enhancement and do not take into account the chemical enhancement effect.
In the simulation model, silver nanoparticles are assumed to be uniformly distributed with a spacing of 11.3 nm. However, during the detection of Raman signals, there may be slight aggregation phenomena within the signal detection area, resulting in smaller distances than the set 11.3 nm. This can lead to stronger localized surface plasmon resonance characteristics.
Therefore, these factors contribute to the difference between the actual AEF and the simulated theoretical results.
3.2.2 Detection of multi-molecule
In order to investigate the multi-molecule detection capability of the microfluidic-integrated D-shaped optical fiber SERS probe, we prepared a mixture containing 10−4 mol/L malachite green oxalate (MG) solution, 10−6 mol/L crystal violet (CV) solution, and 10−8 mol/L R6G solution. The three test solutions were injected into the PDMS channels through different inlets and Raman signals were collected in the detection area.
Figure 3(c) shows the Raman testing results of the R6G, CV, and MG probe molecules in the microfluidic-integrated D-shaped optical fiber SERS probe. The 611 cm−1 peak of R6G, the 912 cm−1 peak of CV, and the 1216 cm−1 peak of MG can be detected under various mixing conditions. Which indicates that the microfluidic-integrated D-shaped optical fiber SERS probe retains its ability to distinguish multiple molecules in a mixed solution.
3.2.3 Testing of repeatability
To test the repeatability of our prepared samples, we first injected a 10−7 mol/L R6G solution into the microfluidic integrated channel and collected the R6G Raman signals in the detection area. Then, we flushed the channel with anhydrous ethanol through the same inlet and collected the signals again, repeating this process for a total of five cycles. Figure 4(a) shows the results of the repeatability testing of the microfluidic integrated D-shaped optical fiber SERS probe. Figure 4(b) shows the Raman intensity of the R6G molecule at 611 cm−1 during the cycling test. From the results, it can be observed that anhydrous ethanol effectively cleans the R6G molecules on the microchannel and fiber probe, indicating that the microfluidic integrated D-shaped optical fiber SERS probe exhibits good repeatability.
Fig. 4. The results of the reproducibility testing. (a) Raman spectra of the microfluidic D-shaped optical fiber SERS probe during 5 cycles of testing in a 10−7 mol/L R6G solution; (b) Raman intensity at 611 cm−1 during the 5 cycles of testing.
3.2.4 Impact of PDMS thickness on the detection results
To investigate the impact of PDMS thickness on the microfluidic D-shaped optical fiber SERS probe, we prepared microfluidic fiber probes with different PDMS thicknesses by varying the total amount of PDMS prepolymer and curing agent. A 10−10 mol/L R6G solution was used as the analyte, and Raman testing was conducted on the microfluidic fiber probes. For comparison purposes, an open D-shaped optical fiber SERS probe without PDMS coating was set as the control group.
Based on the experimental results from Fig. 5, as the PDMS thickness gradually increased from 0 to 2 mm, there was a slight overall decrease in the intensity of the various characteristic peaks of the R6G molecule. However, the decrease was limited, indicating that the PDMS cover has minimal impact on the performance of the microfluidic D-shaped optical fiber SERS probe within the range of 0-2 mm thickness.
Fig. 5. The influence of PDMS thickness on the signal intensity. (a) Raman spectra of 10−10 mol/L R6G in microfluidic D-shaped optical fiber SERS probes with different thicknesses; (b) relationship between the signal intensity of the R6G Raman characteristic peak and the PDMS thickness.
4. Conclusion
This study utilized a molding method to fabricate PDMS microchannel and combined the D-shaped optical fiber@AgNPs composite structure with the microfluidic channels through plasma surface bonding, resulting in a multi-inlet microfluidic integrated D-shaped optical fiber SERS probe. The sample exhibited a detection limit of 10−11 mol/L for R6G and demonstrated good stability, repeatability, and capability for multi-molecule detection. Furthermore, the influence of PDMS thickness on the detection performance of the microfluidic integrated D-shaped optical fiber SERS probe was investigated. The experimental results indicated that the probe's detection performance was minimally affected when the PDMS thickness was within 2 mm. This microfluidic integrated fiber SERS probe holds potential for integrated detection of various trace substances, offering promising applications.
Funding
National Natural Science Foundation of China (62175023).
Acknowledgments
We would like to thank Dr. Gong X.N. at Analytical and Testing Centre of Chongqing University for sample characterization.
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.
References
1. S. L. Kitaw, Y. S. Birhan, and H. C. Tsai, “Plasmonic surface enhanced Raman scattering nano-substrates for detection of anionic environmental contaminants: Current progress and future perspectives,” Environ. Res. 221, 115247 (2023). [CrossRef]
2. X. M. Wang, C. Chen, R. Q. Wang, X. G. Qiao, G. I. N. Waterhouse, and Z. X. Xu, “Performance evaluation of novel Ag@ GO-biomaterial SERS substrates for the ultrasensitive detection of neomycin in foods,” Sens. Actuators, B 380, 133250 (2023). [CrossRef]
3. Z. Y. Zhang, S. X. Yang, R. F. Zhao, J. D. Chen, S. S. Wang, J. Choo, and L. X. Chen, “Recyclable surface-enhanced Raman scattering substrate-based sensors for various applications,” ACS Sustainable Chem. Eng. 11(4), 1278–1293 (2023). [CrossRef]
4. H. F. Zhou and J. Kneipp, “Potential regulation for surface-enhanced raman scattering detection and identification of carotenoids,” Anal. Chem. 95(6), 3363–3370 (2023). [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. B. X. Hu, H. B. Pu, and D. W. Sun, “Flexible Au@AgNRs/CMC/qPCR film with enhanced sensitivity, homogeneity and stability for in-situ extraction and SERS detection of thiabendazole on fruits,” Food Chem. 423, 135840 (2023). [CrossRef]
7. H. Y. Sha, Z. K. Wang, Y. Zhu, and J. Zhang, “Open nanocavity-assisted Ag@PDMS as a soft SERS substrate with ultra-sensitivity and high uniformity,” Opt. Express 31(10), 16484–16494 (2023). [CrossRef]
8. N. Kamkrua, T. Ngernsutivorakul, S. Limwichean, P. Eiamchai, C. Chananonnawathorn, V. Pattanasetthakul, R. Ricco, K. Choowongkomon, M. Horprathum, and N. Nuntawong, “Au nanoparticle-based surface-enhanced raman spectroscopy aptasensors for paraquat herbicide detection,” ACS Appl. Nano Mater. 6(2), 1072–1082 (2023). [CrossRef]
9. T. Li, Z. N. Yu, Z. K. Wang, Y. Zhu, and J. Zhang, “Optimized tapered fiber decorated by Ag nanoparticles for Raman measurement with high sensitivity,” Sensors 21(7), 2300 (2021). [CrossRef]
10. Y. Zhang, C. Gu, A. M. Schwartzberg, and J. Z. Zhang, “Surface-enhanced Raman scattering sensor based on D-shaped fiber,” Appl. Phys. Lett. 87(12), 123105 (2005). [CrossRef]
11. H. Y. Li, R. Chu, J. Y. Cao, F. Zhou, K. K. Guo, Q. M. Zhang, H. C. Wang, and Y. Liu, “Sensitive and reproducible on-chip SERS detection by side-polished fiber probes integrated with microfluidic chips,” Measurement 218(15), 113203 (2023). [CrossRef]
12. Z. Yin, Y. F. Geng, Q. L. Xie, X. M. Hong, X. L. Tan, Y. Z. Chen, L. L. Wang, W. J. Wang, and X. J. Li, “Photoreduced silver nanoparticles grown on femtosecond laser ablated, D-shaped fiber probe for surface-enhanced Raman scattering,” Appl. Opt. 55(20), 5408–5412 (2016). [CrossRef]
13. M. Liu, W. D. Zhang, C. Meng, G. H. Zhang, L. Zhang, D. Mao, and T. Mei, “Lab on D-shaped fiber excited via azimuthally polarized vector beam for surface-enhanced Raman spectroscopy,” Opt. Express 28(8), 12071–12079 (2020). [CrossRef]
14. A. G. Niculescu, D. E. Mihaiescu, and A. M. Grumezescu, “A Review of microfluidic Experimental Designs for Nanoparticle Synthesis,” Int. J. Mol. Sci. 23(15), 8293 (2022). [CrossRef]
15. H. Li, S. D. Yu, D. Wang, X. Y. Huang, Q. Fu, D. L. Xu, L. L. Zhang, S. Z. Qian, and X. B. Qiu, “An automated microfluidic system with one-dimensional beads array for multiplexed torch detection at point-of-care testing,” Biomed. Microdevices 24(4), 38 (2022). [CrossRef]
16. J. C. Guo, F. Y. Zeng, J. H. Guo, and X. Ma, “Preparation and application of microfluidic SERS substrate: Challenges and future perspectives,” J. Mater. Sci. Technol. 37, 96–103 (2020). [CrossRef]
17. S. Y. Li, L. Xia, W. Li, X. Chen, Z. Yang, and J. C. Xia, “All-fiber SERS sensing with a depressed double cladding fiber probe embedded in a microfluidic chip,” Appl. Opt. 58(29), 7929–7934 (2019). [CrossRef]
18. M. K. Fan, P. H. Wang, C. Escobedo, D. Sinton, and A. G. Brolo, “Surface-enhanced Raman scattering (SERS) optrodes for multiplexed on-chip sensing of nile blue A and oxazine 720,” Lab Chip 12(8), 1554–1560 (2012). [CrossRef]
19. H. Y. Sha, Z. K. Wang, and J. Zhang, “SiO2 microsphere array coated by Ag nanoparticles as Raman enhancement sensor with high sensitivity and high stability,” Sensors 22(12), 4595 (2022). [CrossRef]
20. K. Yang, Z. K. Wang, and J. Zhang, “High sensitivity D-shaped fiber decorated with self-assembly Ag nanoparticles for Raman enhancement via collecting backward and vertical direction scattering signals,” Opt. Laser Technol. 164, 109533 (2023). [CrossRef]
21. G. Bar, L. Delineau, A. Häfele, and M. H. Whangbo, “Investigation of the stiffness change in, the indentation force and the hydrophobic recovery of plasma-oxidized polydimethylsiloxane surfaces by tapping mode atomic force microscopy,” Polymer 42(8), 3627–3632 (2001). [CrossRef]
22. Z. K. Wang, H. Y. Sha, K. Yang, Y. Zhu, and J. Zhang, “Self-assembled monolayer silver nanoparticles: Fano resonance and SERS application,” Opt. Laser Technol. 157, 108771 (2023). [CrossRef]
