OSA/OFC/NFOEC 2011
OWS6.pdf OWS6.pdf
Low loss microstructured Polymer Optical Fibre (mPOF) Richard Lwin, Alexander Argyros, Sergio Leon-Saval and Maryanne C. J Large Institute of Photonics and Optical Sciences, University of Sydney, Australia
[email protected]
Abstract: Using a high tension draw we produced a fibre with a loss of 0.16 dB/m at 650nm, making mPOF competitive for the first time with conventional POF. Annealing increases thermal stability. OCIS codes: 060.2310, 060.4005, 060.2270
1.
Introduction
Microstructured polymer optical fibres (mPOF) have attracted considerable attention for a range of applications from ultra-high bandwidth fibres to sensing and transmission of difficult wavelengths in hollow-cores. These applications have exploited the powerful combination of material properties, range of possible structures and possibility of large cores that is made available in mPOF [1,2]. However, in the past the losses of fibres made them uncompetitive with conventional POF, with large core fibres often having losses that were twice that of their conventional counterparts. Previous studies of mPOF losses [3,4] identified a number of contributing factors to the loss including material contamination, surface and material scattering, structural defects, microbending and confinement losses. These insights led to changes in the processing of the fibres, to make the processes cleaner, and to the use of fibres with a larger diameter to reduce microbending. Extensive study of losses in solid core silica microstructured fibres identified surface quality inside the holes as the ultimate limitation to loss [5]. In general the loss performance of a microstructured fibre will include contributions from the several sources identified, but the relative importance of each will depend on the fibre structure. For example surface quality is less significant in large core fibres than in small core fibres because in the latter case the intensity is much higher at the core cladding boundary. For the purposes of this study we have used a large “suspended core” fibre (shown in Fig. 1) to minimize the confinement losses [6] and surface scattering. Previous work [3] had identified the outer diameter as an important parameter, leading to the fibres being sleeved. However more recently we have noted some fibres showing “spot” defects which we believe may be due to trapped air between the fibre and sleeve. For this reason, the test fibre was not sleeved, rather the preform was prepared so that the desired core and outer dimensions could be obtained directly.
Fig. 1. The fibre used in these experiments. The external diameter of the fibre was 500µm, the core diameter was 140 µm. The fibre was not sleeved.
2.
The role of draw tension
The fabrication parameter that was varied in this study was the draw tension. The results are shown in Fig. 2. They clearly show that high tension draws are preferable, with the best loss result, 0.16 dB/m at 650 nm, being comparable with that of commercial POF (0.15 dB/m at 650 nm). The result is also consistent with surface scattering measurements from the external drawn polymer surfaces [7], which showed higher tension draws lead to smoother surfaces.
OSA/OFC/NFOEC 2011
OWS6.pdf OWS6.pdf
1
0.9 20g
0.8
Loss (dB/m)
0.7
0.6
0.5 60g 0.4
0.3 80g 0.2
0.1 500
550
600
650
700
Wavelength (nm)
Fig. 2. Loss spectra for fibres drawn at different tensions. Increasing the draw tension significantly improves the loss. At 80 g the minimum loss is 0.16 dB/m at 650nm.
Fibres drawn at high tensions however are not thermally stable, because subsequent heating causes the polymer to relax to a more energetically favorable configuration. We found that subsequent heating to 90 °C for periods of 25 hours can cause changes to the length of the fibre of up to 2% for the 80 g draw, as well as increased loss. To address this issue, we studied the effect of annealing on the fibre. 3.
The role of annealing
The fibre samples were tested using annealing temperatures ranging from 90 °C-50 °C. Longer times were used for lower temperatures, in accordance with the Arrhenius equation. The results (Fig. 3) clearly show lower annealing temperatures give improved loss performance. While each gives a higher loss than that of the unannealed fibre, this can be reduced by longer lower temperature annealing times (these results will be presented at the conference). 1.4
1.2 90deg Anneal 1 Day
Loss (dB/m)
1
70deg Anneal 3 Days
0.8
0.6 50deg Anneal 7 Days 0.4
0.2 Unannealed 0 500
550
600
650
700
Wavelength (nm)
Fig. 3. Loss spectra of fibres after annealing at 50°C, 70°C and 90°C. Annealing can relieve the effects of a high tension draw, though there is an increase in loss post-annealing for these temperatures. Low temperature produces the best loss result but require longer annealing times.
We subsequently tested the annealed fibre performance after it was heated for an additional period of a day at 90°C (Fig. 4) and found it was quite stable.
OSA/OFC/NFOEC 2011
OWS6.pdf OWS6.pdf
0.8
0.7
Loss (dB/m)
0.6
0.5 C 0.4
B
0.3
0.2
0.1 500
A
550
600
650
700
Wavelength (nm)
Fig. 4. Temperature stability of fibres after annealing. A is the unannealed fibre, B is a fibre annealed at 50°C for 7 days and fibre is heated for an additional 1 day to 90°C.
We therefore believe that the combination of a high tension draw, with no sleeving, but with low temperature annealing offers a route to low loss mPOF that are also thermally stable.
4. References [1] M C. J. Large, L Poladian, G W. Barton and M A. van Eijkelenborg, Microstructured Polymer Optical Fibres, (Springer-Verlag, 2007). [2] A. Argyros, “Microstructured polymer optical fibres,” Journal of Lightwave Technology, 27(11), pp. 1571-1579 (2009). [3] M.A. van Eijkelenborg, et al, “Bandwidth and loss measurements of graded-index microstructured polymer optical fibre”, Electronics Letters, Volume: 40 Issue: 10 Pages: 592-593 (2004). [4] M C J Large et al, “Microstructured polymer optical fibres: New opportunities and challenges”, in Crystals and Liquid Crystals Journal, Special issue, Proceedings of the 8th international conference on frontiers of polymers and advanced materials, volume 446, pages 219–31. Taylor & Francis (2006) [5] Roberts, P. J. et al, “Loss in solid-core photonic crystal fibers due to interface roughness scattering” Optics Express, 13(1), 236 (2005) [6] N. A Issa, “High numerical aperture in multimode microstructured optical fibers” Applied Optics 43 (33), pp. 6191-6197 (2004) [7] J Poulin et al, unpublished results.
Acknowledgement This work was performed in part at the OptoFab node of the Australian National Fabrication Facility, a company established under the National Collaborative Research Infrastructure Strategy to provide nanofabrication and microfabrication facilities for Australian researchers.
