Long Period Grating in ZBLAN fiber using the pull-and-heat method
Abstract
This Letter demonstrates a method for long-period grating inscription in fluoride fibers using the pull-and-heat method. It also presents an analysis of fiber deformation when subjected to a combination of heating temperature, duration, and tension and suggests a methodology for high-precision micro-tapering fiber regions developing for low-loss LPG inscription. Finally, we show that the proposed methodology has the potential for even stronger LPGs, since it offers a high degree of uniformity between the periods and controllability of the micro-tapering regions to meet the optical requirements of the components in a wide range of applications.
Full text
Letter Vol. 50, No. 18/15 September 2025/Optics Letters 5678 Long period grating in ZBLAN fiber using the pull-and-heat method Antreas Theodosiou,* Yauhen Baravets, AND Pavel Peterka Institute of Photonics and Electronics, The Czech Academy of Sciences, Chaberska 57, 182 00 Prague 8–Kobylisy, Czechia *[email protected] Received 17 June 2025; revised 31 July 2025; accepted 4 August 2025; posted 8 August 2025; published 5 September 2025 This Letter demonstrates a method for long-period grating inscription in fluoride fibers using the pull-and-heat method. It also presents an analysis of fiber deformation when subjected to a combination of heating temperature, duration, and tension and suggests a methodology for high-precision micro-tapering fiber regions developing for low-loss LPG inscription. Finally, we show that the proposed methodology has the potential for even stronger LPGs, since it offers a high degree of uniformity between the periods and controllability of the micro-tapering regions to meet the optical requirements of the components in a wide range of applications. Published by Optica Publishing Group under the terms of the Creative Commons Attribution 4.0 License. Further distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI. https://doi.org/10.1364/OL.571074 Introduction. In the last decade, significant research interest was gained in the mid-infrared (mid-IR) electromagnetic part of the spectrum, which can be located somewhere between ∼2 and 20 μm. This wavelength range is particularly interesting for environmental and biochemical sensing applications, material micro-machining, medicine, and future communication networks [1,2]. Moreover, the two atmospheric windows, between 3–5 μm and 8–14 μm, are exceptionally promising for free-space transmission, especially for earth-to-satellite communications [3]. To achieve these technological achievements, exotic glasses must be used to develop appropriate optical fibers since the capabilities of silica fibers are limited to ∼2.2 μm. Such exotic glasses include, among others, fluoride and chalcogenide glasses [4,5]. The existence of particular exotic fibers dates back more than 40 years. However, only in the last decade did the maturity of those fibers reach an acceptable level of quality, especially the fluorozirconate or ZBLAN. This enabled a progressive scientific achievement in fiber lasers and reached significant milestones with respect to optical fiber components. Some of the latest progress in mid-IR fiber lasers includes efficient pumping methods to achieve higher quantum efficiencies [6], and supercontinuum sources extended up to 13 μm [7] and several octaves of spectrum [8] while mid-IR supercontinuum fiber lasers are already commercially available. Meanwhile, significant progress was achieved with respect to the development of optical fiber components, which are an essential and critical cornerstone for further development in the field. Ideally, we want to attain an equivalent component range similar to the silica counterparts. The necessary components that need to be developed include, among others, gratings, couplers, polarizing elements, pump and signal combiners, isolators, and others [9]. To the best of our knowledge, significant progress was achieved regarding the inscription of monolithic fiber Bragg grating (FBG) [10] and tilted FBGs [11], the development of optical fiber couplers [12,13] using multimode and single-mode fibers, and different designs for pump and signal combiners [14,15]. Except for the fiber lasers applications, the sensing applications is also an emerging and very interesting field growing in the mid-IR range [16,17]. In-fiber components like gratings are essential for such applications. Except for the FBG, another type of grating is the long-period gratings (LPGs), which mainly operate in transmission spectrum when compared to FBGs that can operate both in transmission and in reflection. The LPGs have been extensively studied for applications in optical communications and sensing systems, mainly in the NIR wavelength range [18,19]. In contrast to uniform fiber Bragg gratings (FBGs), the LPGs have periods larger than 100 μm and operate exclusively in transmission, coupling the core mode with co-propagating cladding modes, as described by the following phase-matching condition: 𝜆𝐿𝑃𝐺 =(𝑛𝑐𝑜,𝑒𝑓 𝑓 −𝑛𝑐𝑙,𝑒𝑓 𝑓)Λ,(1) where Λ is the period of index change, and nco,eff and ncl,eff are the effective refractive indices of the core and cladding, respectively [18,20]. Sensing applications of LPGs are of particular interest due to their high sensitivity to external perturbations such as temperature and surrounding refractive index [21–23]. Also, these components can be used for gain flattening, signal modulation, and as dispersion-managing elements in actively mode-locked laser sources. Considering the silica fibers, for the fabrication of these elements, there are two main methodologies; method 1, related to periodic refractive index modifications of the core refractive index either using lasers, for instance femtosecond lasers [24] or CO2lasers [25] and method 2, is related with the periodic deformation of cladding-core region using other techniques such as electric arc [26] or CO2lasers [27]. 0146-9592/25/000001-04 Journal © 2025 Optica Publishing Group
Letter Vol. 50, No. 18/15 September 2025/Optics Letters 5679 Table 1.Suggested Filament Parameters for the LPG Inscription Using Thorlabs IRZS23 Fiber Parameters Values Power (W) 9.1 Argon flow (l/min) 0.7 Duration (s) 0.7 Tension (gr) 63–64 Regarding the first method and the progress using fluoride fibers, Heck et al. [28], in 2018, reported on the LPG in a double-cladding Erbium-doped ZBLAN fiber using femtosecond laser inscription. The LPG consisted of 120 periods and a transmission depth of 24 dB, which, after an annealing protocol at 225 °C, was decreased to ∼6 dB. A year later, the same group used a similar LPG as a loss mechanism to mitigate parasitic lasing in the mid-IR amplifier [29]. In addition to his work, She et al. [30], in 2021, also inscribed an LPG using a femtosecond laser inscription, consisting of 100 periods and a strength of∼11 dB at 3400 nm. This group also demonstrated the LPG as a strain sensor, reporting a sensitivity of 4.23 pm/με [30]. These kinds of in-fiber components are essential to avoid intra-cavity splicing joints and, as a result, improve the overall performance of the fiber lasers. On the other hand, LPGs manufactured using the second method were developed using a filament splicer only recently, mostly performing micro-tapering regions. LPGs in ZBLAN fiber operating at 1870 nm [31] and 3100 nm [32] were designed with resonance dips not exceeding 20 dB. In addition, when developing LPGs with the micro-tapering method, there are limitations on tapering lengths; as a result, this is more appropriate for periods higher than 500 μm. However, LPGs developed with the second method are of particular interest not only for fiber laser applications but also for sensing purposes, enabling exciting applications. This is because the cladding deformation makes the light more easily coupled out of fiber and interacts with the ambient environment. As a result, refractive index and absorption measurements related to spectrum signatures of gases and liquids are possible. Moreover, previous studies in silica fibers have shown that the micro-tapering region’s geometric characteristics, such as diameter and length, can adjust and optimize the element’s overall performance, extending the available operation range and sensitivity. In this Letter, we present our results regarding the development of an LPG in ZBLAN fiber using the pull-and-heat technique. This LPG is based on the principles of the second method, applying a locally controlled head to induce periodic deformation of the optical fiber. We investigated the material according to the heating characteristics and showed how to control the waist diameter and waist length by controlling the heating profile. This is particularly interesting for sensing purposes since the dimensions of the tapering areas are expected to give different sensing characteristics, opening the routes for gas and liquid sensing applications. Experimental details. We used the Vytran GPX-3400 filament splicer and the graphite filament V2 for the LPG inscription. Before grating the inscription, we performed several investigation attempts to optimize the process parameters and explore the material. The main inscription parameters of the inscription process were the power, the argon flow, and the axial tension applied to the fiber. The final inscription parameters Fig. 1.Waist diameter and diameter difference between the pristine-unprocessed ZBLAN fiber for various argon flow adjustments and constant power and axial tension. Fig. 2.Waist length of the micro-tapering region for various argon flow adjustments and constant power, and axial tension. were decided based on the waist diameter, the length of the waist region, and the loss induced by each modified area as summarized in Table1. The fiber used for this work was IRZS23 from Thorlabs, with a core and cladding diameter of 9 μm and 125 μm, respectively. The numerical aperture (NA) of the fiber was 0.19, and the mode field diameter was 10.5 μm at 2500 nm. Our attempt was to start keeping the heating duration short at 0.7 s, the argon flow 0.3 l/min, and adjusting the power of the filament and the axial tension on the fiber before the heating procedure until a micro-tapering was created. We chose a very short heating duration to avoid any crystallization issues on the fiber. The axial fiber tension varied from 30 to 64 g, where we observed fiber diameter wasting. At this point, we tried to control the modified region by controlling the argon flow. Keeping the power and tension constant, as presented in Figs.1and2, it is possible to control the waist region diameter and length simply by adjusting the argon flow. As presented recently in Ref. [33], an adjustment of 0.01l/min of the argon is equivalent to a temperature change of ∼1.25 °C. The minimum length of the modified region that could be achieved was ∼180 μm. As a result, the smallest LPG period with a duty cycle of 50:50 is∼360 μm, which is more than adequate for the inscription of LPGs with resonance wavelengths from visible to mid-IR wavelength range. It is worth noting that different combinations of power, argon flow, and tension can be used to achieve similar micro-tapering results. Microscope images from three fiber samples showing the differences in waist diameter and length are presented in Fig.3.
5680 Vol. 50, No. 18/15 September 2025/Optics Letters Letter c) b) a) Waist diameter 110 μm modified length 280 μm Fig. 3.Microscope images indicating waist length and diameter of ZBLAN single cladding fibers that were developed keeping all the filament parameters the same and adjusting only the argon flow; (a) argon flow 0.4 l/min, waist diameter 110 μm, modified length 280 μm; (b) argon flow 0.6 l/min, waist diameter 113 μm, modified length 260 μm; (c) argon flow 0.8 l/min, modified diameter 120 μm, waist length 180μm. Fig. 4.Long-period grating spectra using a supercontinuum source SuperK NKT and Yokogawa AQ6375E optical spectrum analyzer, for the periods 17–20, and for the final LPG with 21 periods of 780 μm. Results. Following all the previous studies, we choose a period of the LPG of 780 μm, which results in an LPG duty cycle equivalent to 45/65. The inscription suggested that the inscription parameters are summarized in Table1. During the inscription process, the ZBLAN fiber was connected to a supercontinuum source, NKT SuperK compact emitting in the range of 450–2400 nm, and the transmission spectrum was monitored using an optical spectrum analyzer (OSA), Yokogawa AQ6375E. We choose parameters creating waist cladding diameters of ∼106–107 μm in order to ensure sufficient core deformation for optimum LPG inscription. The core diameter at the waist region is estimated at ∼8.34 μm compared to the unprocessed with a diameter of ∼9.6 μm, which is ∼15% reduced. The final LPG consisted of 21 periods, equivalent to ∼16.38 mm total length, with two LPG resonances, a strong one at 2305nm and a weaker one at 2245 nm. Figure4shows the evolution of the LPG for the periods 17to21, and we observe the gradual strengthening of the LPG, period by period with an increase of 3–4 dB per period, an indication which shows the uniform periodicity of the deformation, Waist Transition Unmodified Fig. 5.(a) The unmodified, transition, and waist region of the fiber were analyzed, and (b) mapping of the refractive index dependence with respect to the position on the transition fiber region for a single period. both of the period but also of the core modification. The final LPG has a transmission notch higher than 25dB at 2305nm with a full width at half maximum (FWHM) bandwidth of 1.5 nm. This is the strongest LPG inscribed in ZBLAN fiber reported until today. Moreover, it is important to note that the LPG is not saturated, and further increasing the LPG strength is possible using the specific methodology; however, in our case, we run out of a dynamic range of the OSA. At the same time, the second resonance wavelength at 2245 nm seems to be a coupling to another cladding mode. The strength of the LPG is around 5dB with an FWHM bandwidth of ∼28 nm, and its strength evolution is less than 0.5 dB per period compared to 3 dB for the other dip. While the operating wavelength is close to the fiber’s cutoff wavelength (∼2300 nm), the reduced coupling strength may be more accurately attributed to the lower overlap between the fundamental core mode and the specific cladding mode involved, rather than to the fiber’s multimode nature. Further investigation of the cladding modes involved could clarify the observed differences in dip strength. The inscription parameters with higher or lower levels of deformation diameter can be used for the LPG inscription; however, a higher level of deformation can potentially result in higher transmission losses, while low deformation may lead to weak periodic changes in the core and, as a result, a higher number of periods may require success to develop a strong LPG. It is worth noting that in this work, we present a methodology to achieve strong LPGs in fluoride fiber; however, different combinations of the heat duration, tension, and temperature using power and argon flow can be used and potentially control better the micro-tapering characteristics of the inscription and material modifications. Moreover, the LPG structure was further analyzed using an Interferometric fiber profiler (IFA-100). We performed refractive index measurements at 975 nm with an index-matching oil with a refractive index of 1.503. We measured the refractive index difference between the waist region, the unmodified region, and the transition region as depicted in Fig.5(a). Regarding the waist region, we took a measurement of three points with a
Letter Vol. 50, No. 18/15 September 2025/Optics Letters 5681 Fig. 6.Transmission losses of a 16.4-mm ZBLAN LPG for the range between 2000 nm and 2200 nm, with the red color as the average fitting for easier reading of the results. difference of 20 μm to ensure that all the region is modified uniformly, and we repeat our measurement for three consecutive periods. We can see that during the thermal procedure using the filament, a negative refractive index is induced on the fiber. Particularly, the refractive index difference between the unmodified and waist region as measured using IFA is found to be on average −0.6 ×10−3for the cladding and −0.33 ×10−3for the core. Moreover, we mapped the refractive index dependence of the ZBLAN fiber with respect to the position scanning the whole transition region for one period with a scanning step of 20 μm (Fig.5(b)). The results show an identical refractive index change rate for the core and cladding for the whole scanning region. In terms of transmission loss, we measured the spectrum before any fiber modification and after the final inscription, and the inscription loss is less than 0.8 dB or 0.049 dB/mm at 2180 nm. Figure6presents the loss of the LPG using the difference between the reference and final LPG spectrum from 2000 nm to 2200 nm. The red plot is the average loss difference between the spectra for better comparison. Conclusions. In conclusion, in this Letter, we present the inscription of an LPG in fluoride ZBLAN single-mode fiber using the pull-and-heat technique. We perform an analysis of fiber deformation when subjected to a combination of temperature, time, and tension, and we suggest a methodology for performing controlled periodic micro-tapering fiber regions with a resolution of <180 μm of a waist length and waist diameter control for a low-loss LPG inscription. Moreover, we developed an LPG with a notch strength >25 dB, bandwidth at FWHM 1.5 nm, and losses ∼0.049 dB/mm. Finally, we show that the proposed methodology has the potential for even stronger LPGs since it offers a high degree of uniformity between the periods and controllability of the micro-tapering regions. Funding. Ministerstvo Školství, Mládeže a Tělovýchovy (CZ.02.01.01/ 00/22_010/0001912, CZ.02.01.01/00/22_008/0004573). Acknowledgments. The work was co-funded by European Union and the state budget of the Czech Republic under the projects LEGEND CZ.02.01.01/00/22_010/0001912, and LasApp CZ.02.01.01/00/22_008/0004573. Disclosures. The authors declare no conflicts of interest. Data availability. No data were generated or analyzed in the presented research. REFERENCES 1. S. De Bruyne, M. M. Speeckaert, and J. R. Delanghe, Crit. Rev. Clin. Lab. Sci. 55, 1 (2018). 2. J. Haas and B. Mizaikoff, Annu. Rev. Anal. Chem. 9, 45 (2016). 3. B. M. Walsh, H. R. Lee, and N. P. Barnes, J. Lumin. 169, 400 (2016). 4. J. H. V. Price, X. Feng, A. M. Heidt, et al.,Opt. Fiber Technol. 18, 327 (2012). 5. Y. Wang and S. Dai, PhotoniX 2, 9 (2021). 6. O. Henderson-Sapir and D. J. Ottaway, APL Photonics 9, 100901 (2024). 7. T. Sylvestre, E. Genier, A. N. Ghosh, et al.,J. Opt. Soc. Am. B 38, F90 (2021). 8. T. S. Saini and R. K. Sinha, Prog. Quantum Electron. 78, 100342 (2021). 9. S. D. Jackson, APL Photonics 9, 070904 (2024). 10. I. Chiamenti, T. Elsmann, A. Reupert, et al.,Opt. Lett. 46, 1816 (2021). 11. A. Theodosiou, O. Henderson-Sapir, K. Kalli, et al.,Proc. SPIE 12573, 1257303 (2023). 12. F. Anelli, A. Annunziato, A. M. Loconsole, et al.,J. Lightwave Technol. 42, 2457 (2024). 13. M. Rezaei, G. T. Zeweldi, M. H. M. Shamim, et al.,Opt. Express 31, 27183 (2023). 14. B. Perminov, K. Grebnev, U. Hübner, et al.,Light Adv. Manuf. 5, 1 (2024). 15. S. Magnan-Saucier, S. Duval, C. Matte-Breton, et al.,Opt. Lett. 45, 5828 (2020). 16. K. Goya, Y. Koyama, Y. Nishijima, et al.,Sens. Actuators B Chem. 351, 130904 (2022). 17. F. Anelli, S. Venck, S. Cozic, et al.,J. Lightwave Technol. 43, 5982 (2025). 18. S. W. James and R. P. Tatam, Meas. Sci. Technol. 14, R49 (2003). 19. F. Esposito, Res. Opt. 5, 100196 (2021). 20. A. Theodosiou, R. Min, A. G. Leal-Junior, et al.,Opt. Lett. 44, 5346 (2019). 21. R. Falciai, A. G. Mignani, and A. Vannini, Sens. Actuators B Chem. 74, 74 (2001). 22. C. Du, Q. Wang, S. Zhao, et al.,Opt. Laser Technol. 158, 108936 (2023). 23. L. Qi, C.-L. Zhao, J. Yuan, et al.,Sens. Actuators B Chem. 193, 185 (2014). 24. A. Martinez, I. Y. Khrushchev, and I. Bennion, Opt. Lett. 31, 1603 (2006). 25. F. Tian, J. Kanka, B. Zou, et al.,Opt. Express 21, 13208 (2013). 26. R. Ranjan, F. Esposito, A. Iadicicco, et al.,IEEE Sens. J. 16, 4265 (2016). 27. M. A. van Eijkelenborg, W. Padden, and J. A. Besley, Opt. Commun. 236, 75 (2004). 28. M. Heck, S. Nolte, A. Tünnermann, et al.,Opt. Lett. 43, 1994 (2018). 29. M. Heck, J.-C. Gauthier, A. Tünnermann, et al.,Opt. Express 27, 21347 (2019). 30. L. She, Q. Qi, P. Zhang, et al.,Infrared Phys. Technol. 116, 103808 (2021). 31. A. Theodosiou, Y. Baravets, K. V. Grebnev, et al., in Advanced Photonics Congress 2024 (Optica Publishing Group, 2024), paper JTu1A.10. 32. F. Anelli, A. Annunziato, A. M. Loconsole, et al.,J. Lightwave Technol. 43, 2803 (2025). 33. A. Theodosiou, O. Henderson-Sapir, Y. Baravets, et al.,Adv. Photonics Res. 6, 2400150 (2025).