We demonstrate that the efficiency of CO2 laser writing of long-period fiber gratings in a solid-core photonic crystal fiber (PCF) can be enhanced greatly by applying tension to the fiber during the writing process through the mechanism of frozen-in viscoelasticity. Using this mechanism, we are able to write strong gratings in PCFs with a dosage of CO2 laser radiation low enough not to cause any significant fiber structure deformation.
©2009 Optical Society of America
Long-period fiber grating (LPFG), which provides an efficient means to couple light from the guided core mode to the cladding modes of a single-mode fiber at specific resonance wavelengths , has received much attention because of its wide applications in optical communications  and sensing . Apart from conventional single-mode fibers, LPFGs have been formed in various kinds of photonic crystal fibers (PCFs) [4–19] and such special gratings can offer many new applications as sensors [6, 10, 15, 19] and fiber components [7, 11]. Many methods are available for the fabrication of LPFGs in PCFs, which include the CO2-laser writing method [4–10], the mechanical method [11–13], the femtosecond-laser irradiation method [14, 15], and the electric-arc method [16–19]. The CO2-laser writing method is particularly flexible, as it is applicable to practically any fibers and the writing process can be computer-programmed to produce complicated grating profiles.
The mechanisms of CO2-laser writing of LPFGs in conventional fibers have been studied in detail [20, 21]. For fibers that contain a significant residual mechanical stress in the core, CO2-laser heating can relax the stress and thus lower the refractive index of the core . For heavily doped fibers, a low dosage of CO2-laser radiation is sufficient to change the glass structure in the core (by glass densification) and raise its refractive index . Thanks to these mechanisms, LPFGs can be written in those fibers with CO2-laser radiation weak enough not to deform the fiber. For single-material silica PCFs, however, there is little residual stress in the fiber (as the viscosity difference between the core and the cladding is small), so stress relaxation is not an effective mechanism for grating formation. Practically all the LPFGs written in PCFs by local heating rely on glass structure change and fiber deformation. Because of the high fictive temperature of pure silica, a high heating temperature (achieved with intense CO2-laser radiation [4–10] or an electric arc [16–19]) is needed to cause significant glass structure change, which also often results in fiber deformation (such as hole collapse and surface indentation) and hence a high insertion loss [5, 7, 8]. A recent study shows that by applying tension to a conventional lightly doped single-mode fiber, the CO2-laser radiation required for grating formation can be reduced significantly . The enhancement in the writing efficiency is due to the generation of inelastic frozen-in strains in the tensioned fiber by the CO2-laser radiation. The mechanism is known as frozen-in viscoelasticity, which is the effect of freezing tensile inelastic stresses into the glass network structure during the fiber draw process or heat treatment of a post-draw fiber under tension [23, 24].
In this paper, we explore frozen-in viscoelasticity as a new mechanism for the writing of LPFGs in PCFs and demonstrate the effect of reducing the CO2-laser dosage required for grating formation in a PCF by applying tension to the fiber. Our study opens up new possibilities on the fabrication of LPFGs in PCFs and the control of their characteristics. We should mention that applying tension to a fiber has been demonstrated as a means to increase the fiber’s ultra-violet photosensitivity , to fine-tune the strength of a CO2 laser written LPFG , or to reduce the polarization dependent loss of arc-induced  and mechanically induced LPFGs .
2. Grating fabrication
The PCF used in our experiments was a solid-core PCF (Crystal Fiber, LMA-10), as shown in Fig. 1, which was made of pure silica and had a 10-μm core. The writing laser was a high-frequency pulsed CO2 laser (CO2-H10, Han’s Laser), which could deliver a maximum average output power of 10 W. Light from a broadband source was launched into the PCF and the output spectrum was measured with an optical spectrum analyzer. The PCF (with the jacket along the section exposed to the CO2-laser beam removed) was mounted on a stage with one end fixed and the other end tensioned with a weight. The repetition rate and the width of the laser pulses were 5 kHz and 13 μs, respectively, which corresponded to an average output power of 0.65 W. The spot of the CO2-laser beam on the fiber was 90 μm in diameter. The beam was controlled by a computer to scan across the fiber in the transverse direction at a speed that could be set to control the energy density of the CO2-laser radiation on the fiber. Transverse scanning was advanced along the fiber at steps with each step equal to the grating period. A scanning cycle was completed when the number of periods required was reached. The scanning cycle could be repeated as many times as desired. The transmission characteristics of the fiber, in particular, the resonance wavelength and the grating strength (the contrast at the resonance wavelength), were recorded on the completion of every scanning cycle. We chose a grating period of 400 μm and a grating length of 20 mm (50 periods), which produced a clear rejection band within the spectral range of the broadband source.
3. Results and discussions
We first wrote gratings in the PCF with a low CO2-laser energy density of 1.8 J/mm2. We compare the gratings written in an untensioned fiber, a fiber tensioned with a 175-g weight, and a fiber tensioned with a 220-g weight. The variation of the contrast at the resonance wavelength with the number of scanning cycles for the three gratings is shown in Fig. 2(a). Practically no grating is formed in the untensioned fiber, while a strong grating is formed in the fiber tensioned with a 220-g weight. The grating contrast saturates after ~200 scanning cycles and subsequent irradiation cannot increase it much further. It is clear that the writing efficiency is enhanced greatly with the applied tension, which is consistent with the recent experiments using conventional lightly doped single-mode fibers .
As in the case of a conventional fiber, CO2-laser heating of a PCF under tension should lead to the generation of frozen-in strains across the fiber [23, 24], which tends to decrease the refractive index of the fiber with a stronger effect on the exposed side of the fiber . The shift in the resonance wavelength with the number of scanning cycles is shown in Fig. 2(b). With ~300 scanning cycles, the resonance wavelength shifts to a longer wavelength by ~7 nm for the less tensioned fiber and ~30 nm for the more tensioned fiber. The wavelength shift should be the combined result of the changes of the refractive indices in the core and the cladding with an index decrease in the core leading to a blue shift and an index decrease in the cladding leading to a red shift. The observed red shift shown in Fig. 2(b) suggests that the index decrease in the cladding is indeed more significant than that in the core. The amount of red shift should be larger for the more tensioned fiber because the effect of frozen-in viscoelasticity increases with the applied tension.
The transmission spectra of the gratings written in the two tensioned fibers measured at 255 scanning cycles are shown in Figs. 2(c) and 2(d), respectively. The grating in the more tensioned fiber has a contrast of ~16 dB and an insertion loss <0.5 dB. Because of the gradient of the heating temperature across the fiber, the effect of frozen-in viscoelasticity should produce a non-axially symmetric index distribution across the fiber, which may favor the coupling to non-axially symmetric cladding modes. As shown by the transmission spectra in Figs. 2(c) and 2(d), the most significant rejection dips of the two gratings do not seem to correspond to the same cladding mode, which can be attributed to the difference in the induced index distributions in the two fibers.
When the fiber was set free after the grating had been written, the grating written in the more tensioned fiber curved more severely with the exposed side bent outward, as shown in Fig. 3(a), which agree with the fact that the frozen-in strain was larger on the exposed side of the fiber. This observation confirms the presence of an uneven stress distribution across the PCF. To further verify the existence of frozen-in strain in a PCF, we performed an erasure experiment. After exposing the tensioned fiber (220 g) to CO2-laser radiation for 295 scanning cycles, we removed the weight and continued with the irradiation. The variations of the grating contrast and the resonance wavelength with the number of scanning cycles after removing the applied tension are shown in Fig. 3(b). The grating contrast decreases rapidly from ~16 dB to ~2.5 dB after 40 scanning cycles and the resonance wavelength undergoes a blue shift at the same time, which suggests that the grating can be erased with further CO2-laser irradiation after removing the tension. The mechanism responsible for the grating erasure is simply the relaxation of the frozen-in stresses in the fiber by CO2-laser heating. The erasure cannot be thorough in practice because of the small misalignment between the writing and erasing spots. Inspection of the fiber with a microscope showed negligible hole deformation and surface indentation in the PCF, which confirms that the writing mechanism is indeed due to frozen-in viscoelasticity - fiber structure deformation cannot be erased.
We next wrote gratings in an untensioned PCF and a PCF tensioned with a 60-g weight with a higher CO2-laser energy density of 2.9 J/mm2. Both gratings showed no obvious fiber deformation. Figure 4 shows the change of the grating contrast with the number of scanning cycles for the two gratings. The grating written in the tensioned fiber reaches a maximum contrast of ~16 dB with only 5 scanning cycles, while the grating written in the untensioned fiber reaches a maximum contrast of only ~10 dB with many more scanning cycles. A comparison of Fig. 4 and Fig. 2 shows that the effect of frozen-in viscoelasticity increases tremendously with the CO2-laser energy density. Using an energy density of 2.9 J/mm2 instead of 1.8 J/mm2, we can reduce the number of scanning cycle required to produce a contrast of ~16 dB from 200 to 5 with the applied tension reduced from 220 g to 60 g. The transmission spectra of the gratings written in the tensioned fiber with 5 scanning cycles and the untensioned fiber with 85 scanning cycles are also shown in Fig. 4. The rejection bands of the two gratings are similar, except that the one produced in the tensioned fiber is significantly stronger.
Finally, we wrote gratings in an untensioned PCF and a PCF tensioned with a 220-g weight with a much higher CO2-laser energy density of 5.1 J/mm2. At such a high energy density, it took 17 scanning cycles to reach the maximum contrast in the untensioned fiber. For the strongly tensioned fiber, the writing efficiency was so high that over-coupling occurred even with one scanning cycle, as confirmed by real-time monitoring of the growth of the grating during the scanning. Figure 5 shows the transmission spectra of the gratings written in the untensioned fiber with 17 scanning cycles and the tensioned fiber with one scanning cycle. The grating written in the tensioned fiber shows a complicated spectrum as a result of strong couplings to several orders of cladding modes. Figure 5 also shows the microscope images taken for the untensioned fiber after one scanning cycle (top), the tensioned fiber after one scanning cycle (middle), and the untensioned fiber after 17 scanning cycles (bottom). A comparison of the top and middle images indicates that the fiber deforms more easily under tension, which is expected because of the lower viscosity of the glass. The bottom image shows much more significant deformation along the untensioned fiber after 17 scanning cycles, which confirms the need to introduce fiber structure deformation for the realization of a strong grating (>20 dB) in an untensioned PCF with intense CO2-laser radiation [4–10]. Fiber structure deformation, however, usually leads to a large insertion loss, as shown in Fig. 5. It is important to note that we were not able to erase the grating written in the tensioned fiber after removing the 220-g weight. The grating formed in the tensioned fiber at such a high CO2-laser energy density should be the combined effect of frozen-in viscoelasticity and structure deformation.
We also measured the polarization dependence of the gratings. We observed a weak polarization splitting in the resonance wavelength (about 3–4 nm) for gratings written in untensioned and tensioned PCFs, which indicates that frozen-in viscoelasticity does not introduce significant polarization dependence.
We can enhance greatly the efficiency of CO2-laser writing of LPFGs in a solid-core PCF by applying tension to the fiber during the writing process through the mechanism of frozen-in viscoelasticity. This mechanism allows us to write gratings in a PCF with a CO2-laser dosage low enough not to cause any significant structure deformation along the fiber. It provides an efficient means to fabricate high-quality LPFGs in solid-core PCFs and could be further explored for the fabrication of other microstructured fibers, such as hollow-core bandgap PCFs.
The authors thank Y. Liu and C. Zhang for useful discussions. This research was supported by a research grant from the Research Grants Council of the Hong Kong Special Administrative Region, China [Project No. CityU 111907].
References and links
1. A. M. Vengsarkar, P. J. Lemaire, J. B. Judkins, V. Bhatia, T. Erdogan, and J. E. Sipe, “Long-period fiber gratings as band-rejection filters,” J. Lightwave Technol 14, 58–65, (1996). [CrossRef]
2. K. S. Chiang and Q. Liu, “Long-period gratings for application in optical communications,” Proc. 5th International Conference on Optical Communications and Networks and 2nd International Symposium on Advances and Trends in Fiber Optics and Applications (ICOCN/ATFO 2006) (Chengdu, China, Sept. 2006), 128–133 (2006).
3. S. W. James and R. P. Tatam, “Optical fibre long-period grating sensors: characteristics and application,” Meas. Sci. Technol . 14, R49–R61 (2003). [CrossRef]
4. G. Kakarantzas, T. A. Birks, and P. St. J. Russell, “Structural long period gratings in photonic crystal fibers,” Opt. Lett . 27, 1013–1015 (2002). [CrossRef]
5. Y. Zhu, P. Shum, H. J. Chong, M. K. Rao, and C. Lu, “Strong resonance and a highly compact long period grating in a large mode area photonic crystal fiber,” Opt. Express 11, 1900–1905 (2003). [CrossRef] [PubMed]
6. L. Rindorf, J. B. Jensen, M. Dufva, L. H. Pedersen, P. E. Hoiby, and O. Bang, “Photonic crystal fiber long-period gratings for biochemical sensing,” Opt. Express 14, 8224–8231 (2006). [CrossRef] [PubMed]
8. H. W. Lee, Y. Liu, and K. S. Chiang, “Writing of long-period gratings in conventional and photonic-crystal polarization-maintaining fibers by CO2-laser pulses,” IEEE Photon. Technol. Lett . 20, 132–134 (2008). [CrossRef]
10. Z. He, Y. Zhu, and H. Du, “Long-period gratings inscribed in air- and water-filled photonic crystal fiber for refractometric sensing of aqueous solution,” Appl. Phys. Lett . 92, 044105 (2008). [CrossRef]
14. G. Brambilla, A. A. Fotiadi, S. A. Slattery, and D. N. Nikogosyan, “Two-photon photochemical long-period grating fabrication in pure-fused-silica photonic crystal fiber,” Opt. Lett . 31, 2675–2677 (2006). [CrossRef] [PubMed]
15. A. A. Fotiadi, G. Brambilla, T. Ernst, S. A. Slattery, and D. N. Nikogosyan, “TPA-induced long-period gratings in photonic crystal fiber: inscription and temperature sensing properties,” J. Opt. Soc. Am . B 24, 1475–1481 (2007). [CrossRef]
16. G. Humbert, A. Malki, S. Fervrier, P. Roy, and D. Pafnous, “Electric arc-induced long period gratings in Ge-free air silica microstructure fibre,” Electron. Lett . 39, 349–350 (2003). [CrossRef]
17. K. Morishita and Y. Miyake, “Fabrication and resonance wavelengths of long period grating written in a pure silica photonic crystal fiber by the glass structure change,” J. Lightwave Technol . 22, 625–630 (2004). [CrossRef]
18. T. B. Iredale, P. Steinvurzel, and B. J. Eggleton, “Electric-arc-induced long-period gratings in fluid-filled photonic bandgap fibre,” Electron. Lett . 42, 739–740 (2006). [CrossRef]
19. H. Dobb, K. Kalli, and D. J. Webb, “Temperature-insensitive long period grating sensors in photonic crystal fibre,” Electron. Lett . 40, 657–658 (2004). [CrossRef]
20. C. S. Kim, Y. Han, B. H. Lee, W. T. Han, U. C. Paek, and Y. Chung, “Induction of the refractive index changes in B-doped optical fibers through relaxation of the mechanical stress,” Opt. Commun . 185, 337–342 (2000). [CrossRef]
21. Y. Liu, H. W. Lee, K. S. Chiang, T. Zhu, and Y. J. Rao, “Glass structure changes in CO2-laser writing of long-period fiber gratings in boron-doped single-mode fibers,” J. Lightwave Technol . 2008 (to appear).
23. A. D. Yablon, M. F. Yan, P. Wisk, F. V. DiMarcello, J. W. Fleming, W. A. Reed, E. M. Monberg, D. J. DiGiovanni, J. Jasapara, and M. E. Lines, “Refractive index perturbations in optical fibers resulting from frozen-in viscoelasticity,” Appl. Phys. Lett . 84, 19–21 (2004). [CrossRef]
24. A. D. Yablon, “Optical and mechanical effects of frozen-in stresses and strains in optical fibers,” IEEE J. Sel. Top. Quantum Electron . 10, 300–311 (2004). [CrossRef]
25. E. Salik, D. S. Starodubov, and J. Feinberg, “Increase of photosensitivity in Ge-doped fibers under strain,” Opt. Lett . 25, 1147–1149 (2000). [CrossRef]
26. R. Slavík, “Extremely deep long-period fiber grating made with CO2 laser,” IEEE Photon. Technol. Lett . 18, 1705–1707 (2006). [CrossRef]
27. G. M Rego, J. L. Santos, and H. M. Salgado, “Polarization dependent loss of arc-induced long-period fibre gratings,” Opt. Commun . 262, 152–156 (2006). [CrossRef]
28. G. Rego, “Polarization dependent loss of mechanically induced long-period fibre gratings,” Opt. Commun . 281, 255–259 (2008). [CrossRef]