Close-fitting polymeric jackets on silica optical fibres make the pulse transit time (delay) and attenuation of the fibre strongly time-, temperature- and history-dependent. Hartog, Conduit and Payne (1979) measured the bare-fibre calibrations (dτ/dσ = 53.7 ps km⁻¹ MPa⁻¹, dτ/dT = 35.7 ps km⁻¹ K⁻¹), showed that nylon jacketing induces a residual compressive stress of order 45 MPa that creeps and responds to humidity, observed an unexplained abrupt change of the delay-temperature slope near 30 °C with pronounced hysteresis, and found that immersion in salt solutions produces large, apparently irreversible microbending attenuation with little axial-stress change. The unresolved problem is to identify the controlling physical mechanisms and build a quantitative predictive model — coupling polymer viscoelasticity (creep), hygroscopy, and crystallinity evolution — for the delay and attenuation of jacketed fibre assemblies, replacing the non-existent single universal temperature coefficient of delay. Later literature has resolved the linear-elastic jacket-strain regime and the general lateral-pressure/buckling mechanism of microbending in dual-coated fibres, but the transition-region mechanism, the extrusion creep kinetics, the hygroscopic/osmotic irreversibility, solvent effects, and the integrative time-dependent model remain open, as shown by tight-buffered jumper-cable latency measurements reported in 2018 (arXiv:2204.05772) and by the explicitly conjectural status of the 2022 coating-transition explanation.

Background

An optical fibre delay line or sensor converts a change in the fibre's state into a change in the one-way pulse transit time τ\tau of a light pulse. Two calibrated sensitivities of a bare (unjacketed) silica fibre govern this conversion: dτ/dσ\mathrm{d}\tau/\mathrm{d}\sigma, the change of delay per unit applied axial tensile stress σ\sigma (a combination of the fibre's elastic lengthening and the stress-optic effect, i.e. the change of refractive index with stress), and dτ/dT\mathrm{d}\tau/\mathrm{d}T, the change of delay per unit temperature change (a combination of thermal expansion and the thermo-optic effect). For the graded-index germania/phosphosilicate fibres — graded index meaning that the core refractive index falls gradually from the axis outward — measured by Hartog, Conduit and Payne in 1979 these are 53.7 ps km−1 MPa−153.7\ \mathrm{ps\,km^{-1}\,MPa^{-1}} and 35.7 ps km−1 K−135.7\ \mathrm{ps\,km^{-1}\,K^{-1}} (about 7.5 ppm/K), values still cited for bare fibre in modern correlation-OTDR work (correlation optical time-domain reflectometry: an instrument that reads delay changes along a fibre by cross-correlating repeated probe traces).

Telecommunication fibres are rarely left bare: they carry a soft primary coating and, in tight-buffered constructions, a close-fitting hard secondary jacket (in 1979 typically nylon, extruded hot onto the coated fibre). Because nylon's thermal expansion coefficient is roughly two orders of magnitude larger than silica's, a simple linear-elastic model (used already in the 1979 paper) predicts that the jacketed assembly acquires a temperature coefficient of delay of the form

dτdT∣jacketed=APAFEPkP(dτdσ)F+(dτdT)F\left.\frac{\mathrm{d}\tau}{\mathrm{d}T}\right|_{\mathrm{jacketed}}=\frac{A_{P}}{A_{F}}E_{P}k_{P}\left(\frac{\mathrm{d}\tau}{\mathrm{d}\sigma}\right)_{F}+\left(\frac{\mathrm{d}\tau}{\mathrm{d}T}\right)_{F}

where AA, EE and kk denote cross-sectional area, Young's modulus and linear expansion coefficient, with subscripts PP (plastic jacket) and FF (fibre). With nylon parameters this predicts roughly 185 ps km−1 K−1185\ \mathrm{ps\,km^{-1}\,K^{-1}} — but the measured behaviour violates the elastic picture: the delay-temperature curve of the nylon-jacketed fibre is strongly nonlinear and hysteretic, shows an abrupt slope change near 30 ∘C30\,^{\circ}\mathrm{C} and a low-temperature slope of only about 170 ps km−1 K−1170\ \mathrm{ps\,km^{-1}\,K^{-1}}, depends on cycling history and pauses, and drifts in time — signatures of polymer creep (time-dependent viscoelastic deformation), nylon's hygroscopy (absorption and desorption of water, which swells and softens the polymer), and temperature-dependent modulus and expansion. Extruding the jacket hot and cooling it leaves a residual compressive stress of order 45 MPa in the glass, relieved only partially by creep, so the finished fibre's state depends on the manufacturing process. Chemically, immersion in water softens the nylon and relieves the residual stress, while immersion in a 20 wt% sodium chloride solution increases attenuation by several dB/km through microbending — light loss caused by tiny random lateral deflections of the fibre axis — and drying leaves an irreversible impairment. Microbending here arises from radial (lateral) pressure exerted by the jacket on the fibre, which, acting on an initially curved fibre free to buckle within its soft primary coating, kinks the axis without requiring a large change in axial stress; this buckling-based mechanism for dual-coated fibres has since been formalised (Shiue and co-workers, 1992–2010; Suhir 2007), and long-term viscoelastic treatments of tightly jacketed double-coated fibres exist (Lee, Yang and Chang 2006; Hsueh, Chiou and Shiue 2015, using measured coating viscoelastic data). What has not been achieved is a closed, mechanism-based account of the 1979 observations themselves and of their modern counterparts: tight-buffered jumper cables measured by correlation-OTDR (Azendorf, Dochhan and Eiselt, ITG Photonic Networks 2018, arXiv:2204.05772, posted 2022) still show large, nonlinear, heating/cooling-direction-dependent latency changes, and the low-temperature transition of the delay coefficient is explained in the recent literature only as an explicit conjecture — a soft/hard state change of the coating (Jiang, Su, Fan and Qiu, 2022).

Problem Statement

For a silica optical fibre enclosed in a close-fitting polymeric (nylon) jacket, identify the physical mechanisms responsible for — and construct a quantitative predictive model of — the changes in pulse transit time (delay) and attenuation produced by temperature history and chemical environment. In the formulation arising from Hartog, Conduit and Payne (1979), this comprises: (i) the mechanism of the abrupt change in the slope of the delay-temperature characteristic near 30 ∘C30\,^{\circ}\mathrm{C} in nylon-jacketed fibres, given that the non-linear variation of nylon's expansion coefficient and tensile modulus with temperature does not by itself identify the controlling process; (ii) the detailed kinetics and temperature-time dependence of stress relief by plastic creep (slow, permanent viscoelastic deformation of the polymer under stress) during and after jacket extrusion, so that the residual fibre stress can be predicted from process parameters (extrudate cooling conditions, creep-relaxation time) rather than measured after the fact; (iii) the mechanism by which immersion in water and salt solutions produces large, apparently irreversible increases in attenuation through microbending — light loss caused by tiny random lateral deflections of the fibre axis — while producing only modest axial-stress changes, i.e. how radial pressure changes can generate kinking of the fibre axis; (iv) the mapping from stress changes to pulse delay and attenuation under organic-solvent exposure, for which only the stress-level changes were observed; and (v) a formal predictive model of the effective temperature coefficient of delay of jacketed fibre assemblies that incorporates polymer viscoelasticity (creep), hygroscopy (absorption and desorption of water, which swells and softens the polymer), and crystallinity evolution (the slow change of the semi-crystalline polymer's ordered fraction with temperature history), since the measured coefficient is time-, temperature- and history-dependent and no single universally applicable dτ/dT\mathrm{d}\tau/\mathrm{d}T exists for such assemblies. The observed hysteresis, slope transitions, and irreversible environmental degradation of jacketed fibres should be derivable from jacket material properties and process history.

References

  1. A. H. Hartog, A. J. Conduit, D. N. Payne, "Variation of pulse delay with stress and temperature in jacketed and unjacketed optical fibres", Optical and Quantum Electronics 11, 265–273 (1979), DOI 10.1007/BF00620112, https://doi.org/10.1007/BF00620112
  2. N. Shibata, Y. Katsuyama, M. Tateda, S. Seikai, "Thermal characteristic measurement in jacketed and unjacketed fibres by a spatial technique", Electronics Letters 17, 345–347 (1981), DOI 10.1049/el:19810244, https://doi.org/10.1049/el:19810244
  3. Sham-Tsong Shiue, "Minimization of low-temperature microbending losses in initially curved tightly jacketed double-coated optical fibres", Optical and Quantum Electronics 28, 1379–1394 (1996), DOI 10.1007/BF00326211, https://doi.org/10.1007/BF00326211
  4. Sham-Tsong Shiue, Chen-Sheng Hsu, "Thermally induced microbending losses in double-coated optical fibers during temperature cycling", Journal of Applied Physics 88, 3840–3847 (2000), DOI 10.1063/1.1309033, https://doi.org/10.1063/1.1309033
  5. H.-L. Lee, Y.-C. Yang, W.-J. Chang, "Hydrostatic pressure and thermal loading induced long-term optical effects in tightly jacketed double-coated optical fibers", Applied Physics B 83, 601–607 (2006), DOI 10.1007/s00340-006-2175-4, https://doi.org/10.1007/s00340-006-2175-4
  6. E. Suhir, "Elastic stability of a dual coated optical fiber with a stripped off coating at its end", Journal of Applied Physics 102, 033109 (2007), DOI 10.1063/1.2756998, https://doi.org/10.1063/1.2756998
  7. Yu Lei, Xiumei Jin, Zhihong Zhao, "Influence of the Coating and Jacket on the Actual Strain of Fiber Under Temperature Variations", 2011 Third International Conference on Measuring Technology and Mechatronics Automation (ICMTMA), pp. 417–420 (2011), DOI 10.1109/ICMTMA.2011.391, https://doi.org/10.1109/ICMTMA.2011.391
  8. Yu-Chun Hsueh, Sheng-Che Chiou, Sham-Tsong Shiue, "Minimisation of thermally induced microbending losses in dual-coated optical fibres caused by viscoelastic behaviour of commercial polymeric coatings", IET Optoelectronics 9, 141–144 (2015), DOI 10.1049/iet-opt.2014.0025, https://doi.org/10.1049/iet-opt.2014.0025
  9. Florian Azendorf, Annika Dochhan, Michael Eiselt, "Temperature dependent latency of jumper cables", ITG Photonic Networks 2018, Leipzig, June 2018, arXiv:2204.05772 (posted February 2022), DOI 10.48550/arXiv.2204.05772, https://arxiv.org/abs/2204.05772
  10. Di Jiang, Jun Su, Zhiqiang Fan, Qi Qiu, "Transition of optical fiber delay coefficient under the action of fiber coating at low temperature", Proc. SPIE 12319, 2620405 (2022), DOI 10.1117/12.2620405, https://doi.org/10.1117/12.2620405
  11. Florian Azendorf, Annika Dochhan, Florian Spinty, Mirko Lawin, Bernhard Schmauss, Michael Eiselt, "Fiber as a temperature sensor with portable Correlation-OTDR as interrogator", 21st ITG-Symposium on Photonic Networks 2020, arXiv:2011.13175 (2020), DOI 10.48550/arXiv.2011.13175, https://arxiv.org/abs/2011.13175