Open Access
Add to BibSonomy
Abstract
We demonstrate a new design for active fluoride glass fibres in which a metal coating instead of a polymer coating is applied. A 2-μm-thick silver layer is fabricated chemically to the periphery of the fiber, that in this case is doped with Dy3+ ions, and coating lengths of up to 20 cm were produced. The temperature rise in the core of the fiber was accurately measured using a sensitive fiber Mach-Zenhder interferometer (MZI) employing a stabilised He-Ne laser. We use this setup to show that the metal coating can achieve a reduction in the fibre core temperature by up to 22% for low launched pump power levels. A basic two-dimensional model shows reasonable agreement with the experiment. This work paves the way for the development of active fluoride fiber applications in which reducing and stabilising the temperature of the fibre is critical.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
When compared to active silicate glass fibre, active fluoride glass fibre is significantly more fragile and prone to failure under much lower heat loads. But, for a number of mid-infrared and visible light rare earth ion laser transitions, the use of soft glass fibre such as fluoride glass is unavoidable because these glasses support the lower phonon energies required to maintain sufficient luminescent lifetimes of the upper laser levels. To increase the output power from active fluoride glass fiber, efforts have been directed towards strengthening fluoride fibre and making it more robust to varying environmental conditions. Endcapping the fibre [1], maintaining the confinement of the light to the fibre using fusion splices [2] and the use of fibre Bragg gratings [3,4] have all assisted in improving, as they do for their silicate glass fibre counterparts, the performance of fluoride fibre-based lasers and amplifiers.
For active fiber applications, thermal management of the internal heat load created from pump light absorption becomes an additional concern prompting the need to investigate fiber coatings in particular metal coatings because of the improved heat transfer potential. The addition of a metallic (gold or aluminum) layer onto silicate glass fibre [5,6] has been shown to enhance the robustness of silicate fibre by reducing the core temperature leading to the stable performance from kW-class near-infrared fibre lasers. For silica glass, the value of the thermal conductivity is moderate compared to other oxide glasses (1.38 W/m$\cdot$K [7]) but it has a relatively high melting temperature of 1,710 $^{\circ }$C making it robust to high thermal loads. The fluoride glass family, on the other hand, has not only comparatively lower thermal conductivities e.g., ZBLAN glass has a value of 0.628 W/m$\cdot$K [8] but the glass transition temperatures are much lower, e.g., ZBLAN glass has a value of 260 $^{\circ }$C. This means the fluorides can only sustain thermal loads that are orders of magnitude lower compared to the silicates. Fluoride glass fiber therefore is a particularly good candidate for the benefits provided by metal-coatings given the potential for temperature-related failures of fluoride glass fiber at lower power levels. In particular, the heat load along the fiber can be an accute problem for mid-infrared systems because of the larger quantum defects involved. Water ingress and the runaway failure of doped fluorozirconate glass are also temperature dependent [9] meaning any reduction in the core temperature of active fluoride glass fibre is beneficial.
Coating bulk fluoride glass with a metal coating was first reported decades ago [10]. In this pioneering study, rf-sputtered layers of metal and ceramic materials in 27.5 nm to 100 nm thick layers were applied; it was established that good coating adhesion was achieved when the coating had a thermal expansion coefficient similar to fluoride glass. In a later study [11], low-energy oxygen ions were implanted into the surface of fluoride bulk glass samples leading to the creation of an oxyfluoride layer that was resistant to water ingress. The ion implantation process was expanded to the incorporation of aluminum fluoride [12]. A phosphate glass layer can also be used to overclad fluoride glass for water resistance [13] whereby the glass transition temperature, as well as the thermal expansion coefficient, were important design parameters. In each of these studies, the overall aim was to make fluoride glass more resistant to the environment, especially to water vapor which is known to attack zirconium-based fluoride glasses, the most widely used fluoride glass for optical applications.
In this Letter, we report the fabrication of silver-coated active fluoride glass fibres. To quantify the effect of metal coating on the fiber, we experimentally studied the thermal sensitivity of two different fibres, i.e., polymer-coated and metal-coated Dy$^{3+}$-doped ZBLAN fibres, using a sensitive Mach-Zenhder interferometer (MZI) setup. We finally discuss the possibilities for metal-coated active fibre as a reliable thermal management tool for active fiber systems.
2. Metallic coating fabrication process
The silver coating was applied to the fiber using electroless plating in the form of the Tollens’ reaction, leading to the creation of a thin film of pure silver. The most common ways to coat non-conductive substrates are electroless plating [14], electroplating [15], and vacuum deposition [16]. The Tollen’s reaction was chosen because of the sensitivity of fluoride glass to an acid environment. Moreover, this way is the easiest to perform compared with the other methods. In the first step, aqueous silver nitrate is mixed with ammonia solution leading to the formation of Tollens’ reagent Ag(NH$_{3}$)$_{2}$OH. In a second step Tollens’ reagent oxidizes (aldehyde group) into the corresponding carboxylic acid [17].
The reaction is accompanied by the reduction of silver ions into a metallic silver thin film on the fiber [18]. Figure 1 shows the fabrication process of silver coating applied on a stripped fiber. All reagents, i.e., AgNO$_{3}$, C$_6$H$_{12}$O$_6$ and NH$_4$OH, for the reaction were used without additional purification. The original polymer coating was removed at the fluoride fibre extremity by a chemical process involving dichloromethane and rinsed with ethyl alcohol to remove all organic impurities. The cleaned fiber is then centered in a plastic tube. Aqueous ammonium hydroxide water solution is added to an aqueous silver nitrate solution (0.3 M) until complete dissolution of the silver oxide, Ag$_{2}$O. The plastic tube is then filled with the Tollens’ reagent and a freshly prepared 1.2 mL of glucose solution in water (0.4 M) was added. The solution is heated at 60 $^{\circ }$C for 15 minutes. The silver coated fiber was removed from the plating bath, washed with ethylic alcohol, and dried at room temperature. The thickness of the coating is approximately 2 $\mu$m. Figure 2 shows microscopic end-view and plan-view images of the silver-coated active fluoride glass fibre.
Fig. 1. Schematic diagram of the fabrication process of applying the silver coating to a stripped section fluoride glass fiber using the Tollens’ reaction.
Fig. 2. Measured microscopic images of the metal-coated active fluoride glass fibre. (a) Endface picture, (b) surface picture.
3. Experimental setup and results
It is well known that a fibre temperature change causes an optical phase change in optical fibre and a change in the refractive index of fibre core, (thermo-optic effect). This relation is expressed by [19]:
To measure the core temperature of the active fibres directly, we used an MZI arrangement, and the experimental setup is shown in Fig. 3. A He-Ne laser (0.633 $\mu$m, Melles Griot) was used as a light source. Light from the He-Ne laser was divided by a 50:50 beam splitter. One beam was focused into the active fibre with an aspheric objective lens (OL1, f = 6 mm, Innovation Photonics), and the other propagated through free space. Light from the active fibre was collimated with another aspheric objective lens (OL2, f = 6 mm, Innovation Photonics) onto the second beam splitter. The probe beams were combined using the second beam splitter to make an interference pattern. The fringe pattern was measured with a photodetector and recorded using an oscilloscope and a PC. An in-house-fabricated CW Er-doped ZBLAN fibre laser was employed as the pump source. The pump laser beam and probe beam were combined by a custom dichroic mirror (Rocky Mountain Instrument Co.) that had $\sim$77% reflectivity at 2.8 $\mu$m and $\sim$80% transmission at 0.633 $\mu$m. The pump beam was focused into the core of the active fibre by objective lens (OL2). The active fibres that were tested with either polymer and metal coating was a single-clad Dy$^{3+}$-doped ZBLAN fibre (2000 ppm, Le Verre Fluoré) with a core diameter of 12.5 $\mu$m and numerical aperture (NA) of 0.16. Each fibre had a total length of $\sim$1 m and the silver-coated fiber had a thin film coating that was $\sim$8 cm long.
Fig. 3. Schematic of the Mach-Zehnder interferometer setup for the core temperature excursion measurement. BS: Beam splitter, OL: Objective lens, DM: Dichroic mirror, FM: Flip mirror.
In this experiment, we used commercially available tapered V-groove fiber holders (HVF002, Thorlabs) to support the active fibers. We have placed the holders at each end of the active fiber i.e., the metal-coated pump input side and polymer-coated output side. The dimension of each V-groove holder was 82.5 mm x 32.0 mm x 12.5 mm and composed of stainless steel. We used thin magnetic sheets wrapped in aluminum foil to fix the fiber to the holder. A total of 165 mm of fiber was therefore in contact with a high thermal conductivity fibre support with the rest of the fiber placed onto a metal plate. We fully stabilized our measurement setup, because the MZI was sensitive and the core temperature excursions were small (1$\sim$2 K) by removing vibration, removing any airflow using shielding and maintaining a constant ambient temperature. The He-Ne laser was stabilized using a 1 hour warm-up time before starting a measurement. We checked the stability of the interference fringe pattern before pumping the fiber.
With increasing incident pump power, the interferometer patterns change. Figure 4 shows the number of measured fringes as a function of the elapsed time after pump switch-on until the fringe count became stable. The operating wavelength of pump beam was 2825 nm and the incident pump power was $\sim$230 mW. The coated length in the used Dy$^{3+}$-doped ZBLAN fibre aligned with the portion of the fibre absorbing the most pump power, i.e., the coated length can absorb most of the used pump power ($\sim$230 mW). Also, the coated part heats up the most because the 2.8 $\mu$m pump beam is launched through the metal-coated part. This means that cooling is most important here.
Fig. 4. The measured number of the fringes v.s. elapsed time after optical pumping for the polymer- and silver-coated active fibres at the pump power of $\sim$230 mW.
The total number of fringes for the polymer-coated active fibre was 32 (red dots), while the number of fringes for the metal-coated active fibre was 27 (black dots), for the same pump conditions. According to the Eq. (2), the change in the core temperature excursion above ambient with an incident pump power of $\sim$230 mW for both cases were $\sim$1.79 K and $\sim$1.39 K, respectively. Whilst the difference between the polymer and silver coated fibres is small, given the low pump power and the fact that 8% of the fiber was silver coated, we clearly observed a lower core temperature excursion with the silver-coated Dy-doped ZBLAN fibre. Note that these results were consistent and highly reproducible because the laboratory temperature was stable and we allowed sufficient time for the He-Ne probe laser to stabilise both in terms of intensity and phase. As far as we are aware, this is the first direct measurement of the core temperature excursion of an optically-pumped active fiber as a function of cladding coating material.
4. Discussion
To aid in the understanding of the experiment, we conducted simple numerical simulations of the metal-coated and polymer-coated active fluoride glass fibers using COMSOL for the incident pump power of 230 mW. In this simulation, the following equations were used for a stationary two-dimensional heat transfer model:
where T is the temperature, $\rho$ is the density, C$_{p}$ is the heat capacity, u is the velocity field of fluid in the model, k is the thermal conductivity of solid material and Q is the heat source.In the simple model, the polymer-coated active fibre had core/cladding/coating diameters of 12.5/125/200 $\mu$m, and the silver-coated active fibre had core/cladding/coating diameters of 12.5/125/129 $\mu$m. The thermal conductivity values of the polymer (Polyacrylate) and silver were 0.17 W/m$\cdot$K and 429 W/m$\cdot$K, respectively. Table 1 shows the parameters used in the simulation. The launch efficiency was estimated to be 80% using measurements from passive fibres and the power absorbed to heat generated efficiency was 99% given the low quantum efficiency of the 6H$_{13/2}$ -> 6H$_{15/2}$ transition [21]. And we set the boundary temperature of the outer surface as 20 $^{o}$C for both cases. In this simplified simulation, we didn’t consider heat radiation and air convection at the metal-air or polymer-air boundaries. A more detailed numerical model would also require modelling in three dimensions (through various boundary conditions) to account for the complex mounting, heat flow in the tangential direction in the metallic layer and the limited metallic coating relative to the whole fibre; this is well beyond the scope of the current study. The physics-controlled mesh with normal element size was applied. The simulation results are shown in Fig. 5. The calculated core temperatures of two cases are 21.73 $^{o}$C for the polymer coating and 21.07 $^{o}$C for the metal coating, respectively and the values of the temperature excursion for the experiment and simulation are displayed in Table 2. Despite of simplicity of the numerical model a reasonable agreement between the experiment and simulation is achieved.
Fig. 5. Calculated temperature profiles for (a) polymer-coated and (b) metal-coated active fluoride glass fibers for a 230 mW incident pump power.
Table 1. The used thermal parameters for this simulation
Table 2. Core temperature change comparison between experiment and simulation with a launched pump power of 230 mW
We have conducted simple simulations of the active fluoride fibres to estimate the cooling effect of the silver coating at 10 W pump power. The calculated core temperature excursions were 102.16 $^{o}$C (polymer coating) and 70.73 $^{o}$C (silver coating), respectively. Although the core temperature of real active fibre systems is influenced by many parameters, for example, the doping ion concentration, fibre losses, and absorption cross-section, this results clearly shows the benefits of the metal coating to reduce the core temperature rise.
We investigated the effect of the thickness of the cladding coating on the fiber temperature profiles. In the case of metal-coated fibre, we have conducted simulations with various metal coating thicknesses of 2, 10, 40 $\mu$m but as expected, the temperature profiles are largely unaffected because the large thermal conductivity of silver results in a small temperature gradient across the metallic coating. This means that the absolute temperature difference between each side of the silver coating is small. On the other hand, in the case of polymer-coating fibre, whereby the much lower thermal conductivity of the polymer results in steep temperature gradients across the polymer means the polymer thickness is critical and it is advantageous, from a thermal point of view, to use thinner polymer coatings.
In our experiments, we used commercial metal-based V-groove fibre holders to fix the position of the fibre tips. As shown in Fig. 6, there are two dominant contacts between the metallic (polymer) coating and the fiber mount. The generated heat from the core will flow to the contact point along the metal coating, but it will be more restricted with a polymer coating. Given the high thermal conductivity of metallic coatings compared to polymer coatings, the degree of contact with the surrounding heat sink is comparatively more critical for polymer coatings. Convective heat flow conditions aside, the metallic coating would provide more axially-symmetric temperature profiles compared to polymer coatings under the conditions of a highly asymmetric heat sink arrangement as shown in Fig. 6. Thus stress-induced birefringence and the potential polarisability of the optical fibre may also be correspondingly reduced. More measurements and calculations are required to further understand the benefits and potential drawbacks of metallic coatings that are applied to passive and active soft glass optical fibers.
Fig. 6. Schematic diagram of heat flow (Yellow arrow) in a metal-coated fibre with V-grooved fibre mount.
The metal coating itself can absorb both stray pump light and fluorescence which can be trapped by reflection at the interface with the metallic layer. This will cause additional heating but given the high thermal conductivity of the metal, the additional heat will quickly dissipate through conduction. As shown in Ref. [5] and by numerous laser and amplifier demonstrations, heating caused by additional light absorption will not result in a significant contribution to the core temperature excursion. For mid-infrared lasers, the fluorescence can be strongly absorbed in the polymer in standard arrangements causing fiber failure; metallic coatings will negate this effect.
5. Conclusion
In conclusion, we have presented the first metal-coated soft glass fibre. A sensitive MZI setup was used to measure the core temperature excursion for both polymer-coated and partially silver-coated Dy$^{3+}$-doped ZBLAN fibre. We confirmed the beneficial cooling effect on the core of the fibre provided by the silver-coating, despite the low pump power levels and short coating lengths (relative to the total fibre length) that were involved. Our results show that metallic coating of soft glass fibres is a viable methodology towards the better thermal management of active and passive soft glass fiber systems.
Funding
Air Force Office of Scientific Research (FA2386-19-1-4049); Australian Research Council (DP210102442).
Disclosures
The authors declare no conflicts of interest.
Data availability
The data that supports the findings of this study is available from the corresponding author upon reasonable request.
References
1. Y. O. Aydin, F. Maes, V. Fortin, S. T. Bah, R. Vallée, and M. Bernier, “Endcapping of high-power 3 μm fiber lasers,” Opt. Express 27(15), 20659–20669 (2019). [CrossRef]
2. D. Faucher, M. Bernier, G. Androz, N. Caron, and R. Vallée, “20 w passively cooled single-mode all-fiber laser at 2.8 μm,” Opt. Lett. 36(7), 1104–1106 (2011). [CrossRef]
3. G. Bharathan, R. I. Woodward, M. Ams, D. D. Hudson, S. D. Jackson, and A. Fuerbach, “Direct inscription of bragg gratings into coated fluoride fibers for widely tunable and robust mid-infrared lasers,” Opt. Express 25(24), 30013–30019 (2017). [CrossRef]
4. M. Bernier, D. Faucher, R. Vallée, A. Saliminia, G. Androz, Y. Sheng, and S. Chin, “Bragg gratings photoinduced in zblan fibers by femtosecond pulses at 800 nm,” Opt. Lett. 32(5), 454–456 (2007). [CrossRef]
5. J. M. Daniel, N. Simakov, A. Hemming, W. A. Clarkson, and J. Haub, “Metal clad active fibres for power scaling and thermal management at kw power levels,” Opt. Express 24(16), 18592–18606 (2016). [CrossRef]
6. X. Y. Charles, O. Shatrovoy, T. Fan, and T. F. Taunay, “Diode-pumped narrow linewidth multi-kilowatt metalized yb fiber amplifier,” Opt. Lett. 41(22), 5202–5205 (2016). [CrossRef]
7. D. C. Brown and H. J. Hoffman, “Thermal, stress, and thermo-optic effects in high average power double-clad silica fiber lasers,” IEEE J. Quantum Electron. 37(2), 207–217 (2001). [CrossRef]
8. X. Zhu and N. Peyghambarian, “High-power zblan glass fiber lasers: review and prospect,” Adv. OptoElectron. 2010, 1–23 (2010). [CrossRef]
9. N. Caron, M. Bernier, D. Faucher, and R. Vallée, “Understanding the fiber tip thermal runaway present in 3μm fluoride glass fiber lasers,” Opt. Express 20(20), 22188–22194 (2012). [CrossRef]
10. P. Schultz, L. Vacha, C. T. Moynihan, B. Harbison, K. Cadien, and R. Mossadegh, “Hermetic coatings for bulk fluoride glasses and fibers,” in Materials Science Forum, vol. 19 (Trans Tech Publ, 1987), pp. 343–352.
11. Y. Dai, T. Kawaguchi, K. Suzuki, S. Suzuki, K. Yamamoto, and A. Masui, “Water protective surface layer on fluorozirconate glass produced by low-energy oxygen ion implantation,” Appl. Phys. Lett. 58(4), 422–424 (1991). [CrossRef]
12. Y. Dai, T. Kawaguchi, T. Osuka, M. Tada, K. Suzuki, S. Suzuki, and A. M. A. Masui, “Ion-assisted deposition of aluminum nitride on the fluorozirconate glass surface,” Jpn. J. Appl. Phys. 30(7B), L1314 (1991). [CrossRef]
13. M. Hartmann, G. Frischat, K. Högerl, and G. West, “Resistant oxide coatings for heavy metal fluoride glasses,” J. Non-Cryst. Solids 184, 209–212 (1995). [CrossRef]
14. Y. Shacham-Diamand, T. Osaka, Y. Okinaka, A. Sugiyama, and V. Dubin, “30 years of electroless plating for semiconductor and polymer micro-systems,” Microelectron. Eng. 132, 35–45 (2015). [CrossRef]
15. B. Satpathy, S. Jena, S. Das, and K. Das, “A comparative study of electrodeposition routes for obtaining silver coatings from a novel and environment-friendly thiosulphate-based cyanide-free electroplating bath,” Surf. Coat. Technol. 424, 127680 (2021). [CrossRef]
16. L. Quintino, “Overview of coating technologies,” in Surface Modification by Solid State Processing, (Elsevier, 2014), pp. 1–24.
17. W. E. Benet, G. S. Lewis, L. Z. Yang, and D. P. Hughes, “The mechanism of the reaction of the tollens reagent,” J. Chem. Res. 35(12), 675–677 (2011). [CrossRef]
18. S. Sandlin, T. Kinnunen, J. Rämö, and M. Sillanpää, “A simple method for metal re-coating of optical fibre bragg gratings,” Surf. Coat. Technol. 201(6), 3061–3065 (2006). [CrossRef]
19. M. Silva-López, A. Fender, W. N. MacPherson, J. S. Barton, J. D. Jones, D. Zhao, H. Dobb, D. J. Webb, L. Zhang, and I. Bennion, “Strain and temperature sensitivity of a single-mode polymer optical fiber,” Opt. Lett. 30(23), 3129–3131 (2005). [CrossRef]
20. B. J. White, J. P. Davis, L. C. Bobb, H. D. Krumboltz, and D. Larson, “Optical-fiber thermal modulator,” in Optical Fiber Communication Conference, (Optica Publishing Group, 1987), p. TUJ1.
21. L. Gomes, A. F. H. Librantz, and S. D. Jackson, “Energy level decay and excited state absorption processes in dysprosium-doped fluoride glass,” J. Appl. Phys. 107(5), 053103 (2010). [CrossRef]
