Open Access
Add to BibSonomy
Abstract
This study presents the development of an in-situ background-free Raman fiber probe, employing two customized double-cladding anti-resonant hollow-core fibers (AR-HCFs). The Raman background noise measured in the AR-HCF probe is lower than that of a conventional multi-mode silica fiber by two orders of magnitude. A plug-in device for fiber coupling optics was designed that was compatible with a commercially available confocal Raman microscope, enabling in-situ Raman detection. The numerical aperture (NA) of both AR-HCF claddings exceeds 0.2 substantially enhancing the collection efficiency of Raman signals at the distal end of the fiber probe. The performance of our Raman fiber probe is demonstrated by characterizing samples of acrylonitrile–butadiene–styrene (ABS) plastics, alumina ceramics, and ethylene glycol solution.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
The use of optical fibers as probes play a key role in in-situ and in-vivo Raman endoscopy, which has widespread applications in real-time in-vivo tissue imaging [1,2], cancer detection [3–5], and the diagnosis of diseases such as esophageal and gastric dysplasia [6,7].
To mitigate the inherent Raman background noise from the silica glass fiber host, a dual-fiber probe configuration is often adopted, where the delivery of the pump laser and collection of the Raman signal is spatially segregated [8,9]. At the distal end of the dual-fiber probe, the inclusion of a narrow band-pass filter/film is necessary to attenuate the considerable broadband silica Raman noise originating from the delivery fiber for the pump laser, giving rise to the challenges of probe design towards lower cost, higher compactness and smaller volume.
The anti-resonant hollow-core fiber (AR-HCF) is a newly emerging technology that guides light through its hollow core. It is characterized by low loss, low optical nonlinearity, low dispersion, and higher optical damage [10]. Compared to other microstructures hollow-core fibers such as Kagome and photonic bandgap hollow core fibers, AR-HCFs offer a much wider low-loss transmission window and a broader tuning range of the central wavelength, spanning from deep ultraviolet to mid-infrared. Consequently, AR-HCFs have been applied in various fields, including sensing [11,12], laser transmission [13,14], communication [15,16], and gas hollow core fiber laser [17,18]. The reduction in modal overlap with silica glass in the cladding results in a significant decrease of Raman noise from the host material guided in the AR-HCF [10]. In 2018, Keratitis et al. demonstrated that using AR-HCFs resulted in a 1,000-fold reduction in silica background Raman emission compared to conventional solid fibers, while maintaining an equivalent collection efficiency [19].
In this study, we demonstrated in-situ Raman detection using a customized background-free Raman fiber probe based on two types of double-cladding AR-HCFs compatible with a commercial Renishaw Raman confocal microscope. Different from the study [19] which built a complete set of AR-HCF based Raman spectrometer, our research goal is to design a detachable probe connected to a commercial Raman confocal microscope. In addition, two types of double-cladding AR-HCFs with simpler structure compared to [19] are used, which helps to reduce the difficulty of fabricating fibers. The numerical aperture (NA) of the double cladding is more than 0.2 to enhance the collection efficiency of the Raman signal at the distal end. A lens-coupling device was designed and resembled, which could be installed directly in confocal Raman microscopes as a replacement for the objective lens. We demonstrate the performance of our Raman fiber probe by analyzing solid and liquid samples, including acrylonitrile–butadiene–styrene (ABS) plastics, alumina ceramics, and ethylene glycol. Our background-free fiber probe of with a minimal size holds promising potential for applications in biomedical sensing and other related fields.
2. Fabrication and characterization of AR-HCFs
2.1 Fiber fabrication
As shown in Fig. 1, two types of double-cladding AR-HCFs were designed and fabricated using the traditional stack-and-draw method. A lower refractive index coating and fluoride-doped silica glass tube were used respectively (denoted by AR-HCF A and B) so as to enable a high collection efficiency of Raman signal from the sample. The geometric parameters of the two AR-HCFs are listed in Table 1. Similar designs of AR-HCF A are reported in [20,21] for application in two-photon microscopy.
Fig. 1. (a) and (b) show the SEM images of the two double-cladding AR-HCFs. The AR-HCF in (a) is coated with a low-index acrylic coating, and one in (b) has a fluorine doped silica tube outside a pure silica cladding. (c) and (d) are optical microscopy images of AR-HCFs in (a) and (b) respectively, when the Koehler illumination is applied.
Table 1. Geometric parameters of AR-HCF A and B
2.2 Fiber characterization
To minimize the impact of bending in the attenuation measurement, the AR-HCF was loosely rewound with a diameter of approximately 0.9 m firstly. A tungsten lamp (Thorlabs, SLS301) served as the broadband light source. A short piece of SMF-28 was used to collect light, and butt-couple with AR-HCF. This approach facilitated a broadband excitation of core mode of AR-HCF instead of cladding modes. AR-HCF A was cut from 203 to 50 m and B from 88.6 to 30 m for attenuation measurement. An optical spectrum analyzer (OSA, YOKOGAWA, AQ6374) was used.
The short-length transmission spectrum shown in Fig. 2(a) indicates that the low-loss transmission window of AR-HCF A spans approximately 600-1300 nm. At wavelength of 633 nm, the attenuation was simulated to be approximately 0.2 dB/m. In Fig. 2(b), the low-loss transmission window of AR-HCF B encompasses the range of 500-1100 nm, with an attenuation measured 0.031 dB/m at 633 nm.
Fig. 2. Transmission spectra and calculated attenuations of AR-HCF A and B. (a) The total length of AR-HCF A was 203 m and cut to 50 m; (B) The total length of AR-HCF B was 88.55 m and cut to 30 m.
2.3 Numerical aperture of AR-HCFs
As shown in Fig. 3, the NA of the cladding and core of the AR-HCFs were characterized by measuring the far-field patterns emitted from the fiber core and cladding respectively. A 633 nm bandpass filter was employed. The modes in the fiber cladding and core are selectively excited by varying the coupling conditions, and their respective far-field patterns are recorded. The NAs for the AR-HCF A and B core were measured to be 0.028 and 0.025 respectively, whereas the NAs for the cladding were 0.252 and 0.224 at a wavelength of 633 nm, almost ten times higher than that of the core.
3. Setup of the Raman fiber probe
3.1 Commercial confocal Raman microscope used in the experiment
The Raman fiber probe was constructed using a versatile Renishaw India Raman microscope as illustrated in Fig. 4(a). A HeNe laser emitting a wavelength of 633 nm was collimated, expanded, and ultimately coupled to the microscope. The microscope objectives focus the laser beam onto the sample surface, resulting in a diameter ranging from 1 to 5 microns. The far-field pattern of the laser beam before reaching the objective lens is shown in Fig. 4(b), which is in poor quality and affects the coupling efficiency with AR-HCFs. The Stokes signal of Raman scattering of the sample was collected by the objective lenses and subsequently analyzed by the system using dispersive devices and a CCD detector with high sensitivity.
Fig. 3. (a) Experimental schematic for measuring NA of AR-HCF. L1 and L2 are optical lenses; M1 and M2 are silver mirrors (with average reflection R > 97%@633 nm); A CMOS camera is mounted on a uniaxial displacement stage to record the change of the far-filed patterns emitted by the AR-HCF along the z direction. (b) (c) The relationship between displacement z and the diameters of the far-field patterns emitted from the AR-HCF A and B cladding and core, respectively.
Fig. 4. (a) Renishaw Inia Raman microscope's internal structure; (b) The far-field pattern of HeNe laser beam before the incidence of objective lenses.
3.2 Customized Raman fiber probe using AR-HCF
Figure 5(a) shows schematic of the connection a double-cladding AR-HCF probe and the Renishaw Raman microscope. The pump laser beam was steered out of the microscope using three silver mirrors and resized by a Keplerian beam expander design. The coupling efficiencies at the incident ends of AR-HCF A and B reach approximately 78%. Figure 5(c) illustrates the prepared distal ends of two types of AR-HCFs. AR-HCF A necessitates the unstriped coating for Raman signal collection rather than a bare end like AR-HCF B.
Fig. 5. (a) Design of optical path connecting AR-HCF probe and Renishaw Raman microscope; (b) The appearance of customized device of fiber coupling by lens in the CAD file; (c) Distal end of two AR-HCF probes.
Figure 6(a) compares the silica glass Raman backgrounds of probes using AR-HCFs and a silica multimode fiber with the same length of half a meter. We demonstrate the spectrum of 200 to 1000 cm−1 because the intrinsic Raman signal of quartz is mainly in the low wave number segment. Under the same conditions of 0.7 mW incident power and 1 s integration time, the measured Raman background of each AR-HCFs was approximately two orders of magnitude lower than that of the traditional silica fiber.
Fig. 6. (a) Measured silica background of solid core fiber, AR-HCF A and AR-HCF B; (b) Heat-shrinkable plastic protective can be used to protect the probe from the pollution; (c) Measured extra Raman noise from the heat-shrinkable plastic protective cover.
Fig. 7. Measurement of ABS plastic sample by using (a) AR-HCF A and (b) AR-HCF B, with the fiber ends contacting (orange) and staying far away from the sample surface (blue). (7 mW incident power, 10 s integration time) The direct measurement of sample with the Renshaw confocal microscope is plotted in yellow. (0.3 mW incident power, 10s integration time).
For the measurement of liquid sample, a heat-shrinkable plastic film was wrapped outside the fiber probe as a protective cover, as shown in Fig. 6(b). This method is found more resilient against damage than the household plastic wrap reported [19]. The Raman noise introduced by the heat-shrinkable plastic was characterized by wrapping half-meter AR-HCF B. As illustrated in Fig. 6(c), the presence of the plastic protective cover only resulted in some noticeable Raman signals at 2900 cm−1 and beyond when the incident power was set at 7 mW and the integration time was 10 s.
4. Application and discussion
4.1 Measurement of solid sample
As shown in Fig. 7, the performance of AR-HCF probe was testified in measuring the Raman signal of ABS plastics and alumina ceramic. An incident power of 7 mW and integration time of 10 s were set. Notably, within the spectrum, a prominent signal in the lower wave number range (particularly 0-500 cm−1) was identified as originating from the silica Raman background. In addition, the spectral line exhibited numerous robust peaks, largely consistent with previously reported lines [22]. However, owing to the variations in the content ratio of the three main components of diverse ABS plastics, the intensities of the spectral line peaks differ.
Notably, the two spectra exhibit prominent peaks at 1554.9 and 2328.9 cm−1, corresponding to the distinctive Raman signals of oxygen and nitrogen molecules, respectively, derived from the atmospheric air contained in the core of AR-HCF [23]. These findings are not readily observable using the original Renishaw Invia Raman system due to the significantly lower particle population density of gases compared to solids and liquids, thereby rendering the detection of gas samples considerably more arduous [24].
Figure 8 shows the measured Raman signal of alumina sample. To reduce spectral peak intensity while avoiding signal saturation, the power of the excitation laser was decreased to 0.7 mW. Peaks at 1368.5 and 1398.7 cm−1 imply as $\alpha - A{l_2}{O_3}$ in [25].
Fig. 8. Measurement of alumina ceramic sample by using (a) AR-HCF A and (b) AR-HCF B, with the fiber ends contacting (orange) and staying far away from the sample surface (blue). (0.7 mW incident power, 10 s integration time) The direct measurement of sample with the Renshaw confocal microscope is plotted in yellow. (20 µW incident power, 10s integration time)
4.2 Measurement of liquid sample
For the sake of exploring the ability of the probe to measure liquids, ethylene glycol was used and configured in nine concentrations ranging from 0% to 100% by volume fraction. The obtained results, as depicted in Fig. 9, revealed distinct peaks at wave numbers of 883.5, 1053.2, 1096.4, 1276.9, 1453.3, and approximately 3000 cm−1 when the incident power was set at 7 mW and the integration time was 10 s. These peaks correspond to the known Raman signal of ethylene glycol, as reported in [26]. However, the presence of the silica Raman background prevented the observation of one weak intrinsic signal peak of ethylene glycol at 433.9 cm−1 in the spectrum.
Fig. 9. Measurement of glycol solutions of different concentrations by using (a) AR-HCF A and (c) AR-HCF B. (b) and (d) are the relationship between the Raman signal intensity at 883.5 cm−1 and the concentration of ethylene glycol solution, where (b) is measured by AR-HCF A, (d) is measured by AR-HCF B. (7 mew incident power, 10 s integration time) The direct measurement of 100% ethylene glycol with the Renshaw confocal microscope is plotted in red. (0.3 maw incident power, 10s integration time).
To explore the limit of detection (LOD) of probes, the ethylene glycol Raman signal peak at 883.5 cm−1 was specifically chosen for feature analysis owing to its superior signal-to-noise ratio compared to other characteristic peaks in the low wave number range. The correlation between the concentration of ethylene glycol solution and the Raman signal intensity at 883.5 cm−1 is shown in Fig. 9(b) and (d). The analysis revealed that the utilization of both types of AR-HCFs yielded a linear correlation in the detection outcomes for concentrations exceeding 20%, whereas concentrations below 20% exhibited disorderliness at low signal-to-noise ratios. Consequently, the LOD for the glycol solution by using our probes was about 20%.
It is noted that LOD of AR-HCF probe can be further enhanced by utilization of a gold-plated container (LOD of 5%) [27] or injection of liquid sample into the core (LOD of 0.04%) [28].
5. Conclusion
In this study, we successfully demonstrated the application of double-cladding AR-HCFs in Raman sensing by developing a specialized fiber probe. The AR-HCF based probe exhibited a remarkable capability of suppressing the silica Raman background noise by two orders of magnitude, even in the presence of a moderate excitation laser quality. The distal end of the probe has a compact size of less than 0.4 mm in diameter. We systematically character the performance of our probe by measuring solid and liquid samples. LOD for ethylene glycol was at a concentration of 20%. Our method can provide an immediate solution to update the in-situ measurement function of nearly all the commercial versatile confocal Raman microscopes. The single-fiber probe design makes the fiber becomes disposable and its significantly lower cost exhibits promising potential for a broad range of in-situ vivo Raman endoscopy applications. However, continuing to suppress the silica background noise and improve NA of AR-HCFs are still problems to be explored.
Funding
Strategic Priority Research Program of the Chinese Academy of Sciences (XDB0650000); Chinese Academy of Sciences (ZDBS-LY-JSC020); Key R&D Program of Shandong Province (2021CXGC010202); National Natural Science Foundation of China (61935002, 62075200, 62127815).
Acknowledgments
We would like to thank Miss Hui Liu for kind guidance and help in using the Renishaw Invia Raman microscope.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are available in Dataset 1, Ref. [29].
Supplemental document
See Supplement 1 for supporting content.
References
1. Z. Huang, S. Teh, W. Zheng, et al., “Integrated Raman spectroscopy and trimodal wide-field imaging techniques for real-time in vivo tissue Raman measurements at endoscopy,” Opt. Lett. 34(6), 758–760 (2009). [CrossRef]
2. M. S. Bergholt, K. Lin, W. Zheng, et al., “In vivo, real-time, transnasal, image-guided Raman endoscopy: defining spectral properties in the nasopharynx and larynx,” J. Biomed. Opt. 17(7), 077002 (2012).
3. M. S. Bergholt, W. Zheng, K. Y. Ho, et al., “Fiber-optic Raman spectroscopy probes gastric carcinogenesis in vivo at endoscopy,” J. Biophotonics 6(1), 49–59 (2013). [CrossRef]
4. D. Petersen, P. Naveed, A. Ragheb, et al., “Raman fiber-optical method for colon cancer detection: cross-validation and outlier identification approach,” Spectrochim. Acta, Part A 181, 270–275 (2017). [CrossRef]
5. H. P. S. Heng, C. Shu, W. Zheng, et al., “Advances in real-time fiber-optic Raman spectroscopy for early cancer diagnosis: pushing the frontier into clinical endoscopic applications,” Transl. Biophotonics 3(1), e202000018 (2021). [CrossRef]
6. M. S. Bergholt, W. Zheng, K. Y. Ho, et al., “Fiberoptic confocal raman spectroscopy for real-time in vivo diagnosis of dysplasia in Barrett's esophagus,” Gastroenterology 146(1), 27–32 (2014). [CrossRef]
7. J. Wang, K. Lin, W. Zheng, et al., “Fiber-optic Raman spectroscopy for in vivo diagnosis of gastric dysplasia,” Faraday Discuss. 187, 377–392 (2016). [CrossRef]
8. Y. Komachi, H. Sato, K. Aizawa, et al., “Micro-optical fiber probe for use in an intravascular Raman endoscope,” Appl. Opt. 44(22), 4722–4732 (2005). [CrossRef]
9. J. Wang, M. S. Bergholt, W. Zheng, et al., “Development of a beveled fiber-optic confocal Raman probe for enhancing in vivo epithelial tissue Raman measurements at endoscopy,” Opt. Lett. 38(13), 2321–2323 (2013). [CrossRef]
10. F. Yu and J. Knight, “Negative curvature hollow core optical fiber,” IEEE J. Sel. Top. Quantum Electron. 22(2), 146–155 (2016). [CrossRef]
11. Z. Li, Z. Wu, Z. Li, et al., “High-sensitivity and fast-response fiber optic temperature sensor using an anti-resonant reflecting optical waveguide mechanism,” Photonics Res. 11(8), 1397–1407 (2023). [CrossRef]
12. Y. Bai, D. Xiong, Z. Yao, et al., “Analysis of CH4, C2H6, C2H4, C2H2, H2, CO, and H2S by forward Raman scattering with a hollow-core anti-resonant fiber,” J. Raman Spectrosc. 53(5), 1023–1031 (2022). [CrossRef]
13. Y. Cui, W. Huang, Z. Zhou, et al., “Highly efficient and stable coupling of kilowatt-level continuous wave laser into hollow-core fibers,” Chin. Opt. Lett. 20(4), 040602 (2022). [CrossRef]
14. M. A. Cooper, J. Wahlen, S. Yerolatsitis, et al., “2.2 kW single-mode narrow-linewidth laser delivery through a hollow-core fiber,” Optica 10(10), 1253–1259 (2023). [CrossRef]
15. Y. Hong, T. Bradley, N. Taengnoi, et al., “Hollow-core NANF for high-speed short-reach transmission in the S+ C+ L-bands,” J. Lightwave Technol. 39(19), 6167–6174 (2021). [CrossRef]
16. A. Nespola, S. Straullu, T. D. Bradley, et al., “Transmission of 61 C-band channels with l-band interferers over record 618 km of hollow-core-fiber,” 2020 Optical Fiber Communications Conference and Exhibition (OFC) (IEEE, 2020).
17. Zefeng Wang, F. Yu, W. J Wadsworth, et al., “Efficient 1.9 µm emission in H2-filled hollow core fiber by pure stimulated vibrational Raman scattering,” Laser Phys. Lett. 11(10), 105807 (2014). [CrossRef]
18. X. Zhu, F. Yu, D. Wu, et al., “Low-threshold continuous operation of fiber gas Raman laser based on large-core anti-resonant hollow-core fiber,” Chin. Opt. Lett. 20(7), 071401 (2022). [CrossRef]
19. S. Yerolatsitis, F. Yu, S. McAughtrie, et al., “Ultra-low background Raman sensing using a negative-curvature fibre and no distal optics,” J. Biophotonics 12(3), e201800239 (2019). [CrossRef]
20. C. Wang, H. Liu, H. Cui, et al., “Two-photon endomicroscopy with microsphere-spliced double-cladding antiresonant fiber for resolution enhancement,” Opt. Express 30(15), 26090–26101 (2022). [CrossRef]
21. C. Wang, H. Liu, J. Ma, et al., “Spiral scanning fiber-optic two-photon endomicroscopy with a double-cladding antiresonant fiber,” Opt. Express 29(26), 43124–43135 (2021). [CrossRef]
22. G. Bikulčius, I. Ignatjev, and A. Ručinskienė, “Rapid method to determine suitability of ABS plastics for metallisation,” Trans. Inst. Met. Finish. 92(1), 47–51 (2014). [CrossRef]
23. Sutherland, Gordon Brims Black Mcivor, “Experiments on the Raman effect at very low temperatures,” Proceedings of the Royal Society of London. Series A, Containing Papers of a Mathematical and Physical Character 141(845), 535–549 (1933).
24. William S.M. Brooks, Matthew Partridge, I. A. K. Davidson, et al., “Development of a gas-phase Raman instrument using a hollow core anti-resonant tubular fibre,” J. Raman Spectrosc. 52(10), 1772–1782 (2021). [CrossRef]
25. S. Y. Reyes-López, “Analysis of the phase transformation of aluminum formate Al (O2CH)3 to α-alumina by Raman and infrared spectroscopy,” Journal of Ceramic Processing Research 14(5), 627–631 (2013).
26. K. Krishnan and R. S. Krishnan, “Raman and infrared spectra of ethylene glycol,” Proceedings of the indian academy of sciences-section A. Vol. 64. No. 2. Springer India: New Delhi, 1966.
27. H. Cai, X. Yu, Q. Chu, et al., “Hollow-core fiber-based Raman probe extension kit for in situ and sensitive ultramicro-analysis,” Chin. Opt. Lett. 17(11), 110601 (2019). [CrossRef]
28. X. Wang, S. Li, S. Gao, et al., “Microfluidic Raman sensing using a single ring negative curvature hollow core fiber,” Biosensors 11(11), 430 (2021). [CrossRef]
29. S. Luan, S. Chen, X. Zhu, et al., “In-situ background-free Raman probe using double-cladding anti-resonant hollow-core fibers,” Science Data Bank, 2023, https://doi.org/10.57760/sciencedb.14392.
