A novel technique for splicing a small core Ge-doped photonic crystal fiber (PCF) was demonstrated using a commercial fusion splicer with default discharge parameters for the splicing of two standard single mode fibers (SMFs). Additional discharge parameter adjustments are not required to splice the PCF to several different SMFs. A low splice loss of 1.0~1.4 dB is achieved. Low or no light reflection is expected at the splice joint due to the complete fusion of the two fiber ends. The splice joint has a high bending strength and does not break when the bending radius is decreased to 4 mm.
© 2008 Optical Society of America
Photonic crystal fibers (PCFs) have been undergoing a rapid development over the past decade due to their unique microstructures and dispersive properties [1–4]. Recently, small core PCFs attracted a great deal of attention because of their potential sensing and communications applications [5, 6]. PCFs often have to be spliced to conventional single mode fibers (SMFs) to be utilized in the field of communications and sensors. However, it is a challenging task to splice an SMF and a PCF, especially for small core PCFs. This is a result of the mode field mismatch between the two types of fibers. Since Bennett, et al.  reported the first splicing of a PCF and a standard SMF, various methods have been demonstrated to connect the two types of fibers using an arc fusion splicer [8–10], a filament splicer , a CO2 laser , and gradient index fiber lenses . One of these methods, the arc fusion splice method, is a promising splicing technique. For all of these methods, however, splicing parameters such as current, arc duration, gap, overlap, and offset distance have to be delicately adjusted to reduce splice loss. Additionally, a clear interface usually occurs at the splice joint due to distinct fiber structures [8–10]. As a result, the splice joint between the PCF and the SMF is usually fragile. And such an interface may result in a high light reflection.
Ge-doping in the core of a PCF is of great interest when trying to achieve photosensitivity for the inscription of optical fiber gratings and when splicing a PCF to a conventional SMF. In this paper, we demonstrate a novel technique for splicing a small core Ge-doped PCF to different SMFs by using a commercial arc fusion splicer. This splicer is set to its default discharge parameters for the splicing of two standard SMFs. Our splice joint has a high bending strength. And no splice interface occurs at the splice joint.
2. Splicing small core PCFs and SMFs
A commercial Sumitomo type-36 arc fusion splicer was employed to splice a PCF and a conventional SMF. The PCFs employed are small core Ge-doped PCFs with a length of ~2m (PCF-252b5 and PCF-282b5), as shown in Fig. 1. Their parameters are listed in Table 1. Due to the complex structure of the core region, a numerical aperture value has been derived from a far field measurement (Fig. 1(c)). According to this measurement the numerical aperture has a value of around 0.3. Therefore, the PCF fiber is not a single mode fiber, but guides five higher order modes in addition to the fundamental mode.
Three conventional SMFs with a length of ~2.5m were employed to demonstrate our splicing technique (Table 1). They cover a range of different geometrical parameters for core diameter, mode field diameter and cladding diameter. Automatic and manual operation modes are available for the fusion splicer. In our experiments, the manual operation mode was applied to demonstrate in detail the splicing procedure. In the operation program of the splicer, the default parameters for the splicing of two standard SMFs are as follows: a fusion duration of 1.5 s, a prefusion duration of 0.1 s, an arc gap of 8.0 µm, an overlap of 15.0 µm, and an arc power of 23 steps. These default splicing parameters were applied to splice the PCF to various SMFs in our experiments; none of the additional splicing parameters had to be adjusted. The measurements were performed at a typical wavelength of 1550 nm. We employed PCF-252b5 to demonstrate our splice technique in the experiments as follows.
2.1 Splicing PCF to SMF-1
The small core Ge-doped PCF (PCF-252b5) was first spliced to a standard SMF (SMF-1, see Table 1). The 1550 nm single wavelength light from an Agilent 8164A tunable laser was transmitted from the SMF to the PCF. The splice loss was measured by a Newport 2832-C dual-channel power meter. In the first step, as shown in Fig. 2(a), the cleaved PCF and SMF-1 were manually aligned via motors to reduce the butt-coupling loss to as low as possible. Here a low butt-coupling loss of 3.9 dB was measured. Then a brief prefusion arc was discharged at a duration of 0.1 s to polish the end faces and remove any dust. In the second step, as shown in Fig. 2(b), the splicing arc was discharged at a fusion duration of 1.5 s, resulting in the fiber ends fusing into two spherical ends. Here the PCF end disappeared off the splicer screen due to the fiber fusion. Fig. 2(c) shows that, by moving the spherical end of the PCF toward the screen center, air holes in the fiber cladding were completely collapsed within a length of ~200 µm and gradually collapsed within a length of ~90 µm. In the third step, as shown in Fig. 2(d), the fibers were moved together via motors until the two spherical ends touched each other. Here a coupling loss of 9.8 dB was measured. In the fourth step, the splicing arc was again discharged at a fusion duration of 1.5 s, fusing the two spherical ends into a joint. This is depicted in Fig. 2(e). A splice loss of 2.7 dB was measured. Then, the splicing arc was repeatedly discharged at a fusion duration of 1.5 s to reduce the splice loss, as shown in Figs. 2(f), (g), and (h). Here the splice losses achieved were 1.5, 1.2, and 1.8 dB after the 3rd, 4th, and 5th arc discharge, respectively. We performed 10 spliced samples for the small core Ge-doped PCF and SMF-1 to investigate the repeatability of our splicing method. Their splice losses from SMF-1 to the PCF are listed on the left hand side of Table 2, showing that for each spliced sample the splice loss is usually reduced to a minimum value after the 3rd or 4th arc discharge.
Then, using the same splicing procedure, we produced another 10 spliced samples for the PCF (PCF-252b5) and SMF-1 to investigate the influence of the direction of light transmittance on splice loss. The splice losses were measured as light was transmitted from the PCF to the SMF-1. The results are listed on the right hand side of Table 2. The minimum splice losses achieved for each sample are illustrated in Fig. 3(a). Both Fig. 3(a) and Table 1 show that the minimum splice losses for coupling from SMF-1 to the PCF vary between 1.0 and 1.4 dB and that the values for coupling from the PCF to SMF-1 vary between 1.7 and 2.2dB.
2.2 Splicing PCF to SMF-2
The splicing procedure as described above was applied to splice the small core Ge-doped PCF (PCF-252b5) to another SMF (SMF-2, see Table 1). This SMF has a slightly smaller core diameter than the PCF. Again, we produced 10 spliced samples to investigate the splice loss from SMF-2 to the PCF and another 10 spliced samples to investigate the splice loss from the PCF to SMF-2. The minimum splice losses achieved for each sample are shown in Fig. 3(b). As can be seen from Fig. 3(b), the minimum splice losses of the former 10 samples for coupling from SMF-2 to PCF are 0.7~1.2 dB and the values of the latter 10 samples for coupling from PCF to SMF-2 are 2.0~2.8 dB.
2.3 Splicing PCF to SMF-3
The splicing procedure was finally applied to splice the small core Ge-doped PCF (PCF-252b5) to a third SMF (SMF-3, see Table 1). This SMF has a slightly larger core diameter than the PCF. Again, we produced 10 spliced samples to investigate the splice loss from SMF- 3 to the PCF and another 10 spliced samples to investigate the splice loss from the PCF to SMF-3. The minimum splice losses achieved for each sample are shown in Fig. 3(c). As can be seen from Fig. 3(c) the minimum splice losses of the former 10 samples for coupling from SMF-3 to PCF are 0.8~1.2 dB and the values of the latter 10 samples for coupling from PCF to SMF-3 are 1.0~1.8 dB.
2.4 Splicing another PCF to different SMFs
Using the same splicing procedure, we spliced another small core Ge-doped PCF (PCF-282b5, see Fig. 1(b) and Table 1) to different SMFs. The measured minimum splice loss from the PCF to SMF-1, SMF-2 and SMF-3 were approximately 0.6-1.0dB, 1.1-1.6dB and 0.7-1.1dB, respectively, after the 4th or 5th arc discharge. It is obvious that both the minimum splice loss and the optimal time of arc discharge depend strongly on the parameters of PCF and SMF, such as microstructure, Ge-doped region, germanium concentration, core/cladding diameters, and mode field diameter. Lower splice loss could be expected for PCFs with smaller NA and lower air-filling fraction in the cladding region, which however may be not desirable for some applications.
During each splicing process, the optimal time of arc discharge also depend strongly on the fiber alignment and the practical discharge current and fusion duration. But, if the minimum spliced loss is achieved after nth arc discharge, the difference between the splice losses after nth and (n±1)th arc discharge usually is a small value. During our splicing process, the next arc discharge can be performed immediately after the last discharge is finished and no latency time is required.
As discussed in previous publications on the splicing of PCFs and SMFs, the splicing parameters usually have to be delicately adjusted to achieve a low splice loss, e.g. 0.9 dB  and 0.7-1.0 dB  for small core PCFs with low air-filling fraction, and 2.53 dB for small core PCF with high air-filling fraction . Such improved splicing parameters have to be adjusted again as soon as a different SMF is spliced. In contrast to these results, the use of a Ge-doped PCF has been shown to strongly improve the splicing conditions with an SMF. The default arc discharge (for the splicing of two standard SMFs) achieved reproducible low splice losses for several different SMFs. No parameter adjustments were required during the splicing process. Low splice losses were achieved in all cases, as shown in Fig. 3, which is a prime advantage over other fusion splicing methods. As can be seen from Fig. 2(c), the PCF end is deformed into a ~200 µm length step-index optical fiber due to the complete collapse of air holes. Since the core of the PCF is Ge-doped, the refractive index in the core is still higher than that in the collapsed cladding. Hence, the light can be guided in the collapsed PCF. Moreover, the air holes are gradually collapsed within a length of ~90 µm along the fiber axis so that there is a gradual transformation from the un-collapsed to the collapsed section. As a result, the light wave in the PCF is gradually transited to a collapse-induced step-index fiber and then transmitted into an SMF. Therefore, as shown in Fig. 3, a low splice loss can be expected from our splicing samples.
We performed an experiment to evaluate the loss attributable to the light transition between the uncollapsed Ge-doped PCF and the collapsed Ge-doped PCF. Both ends of a Ge-doped PCF (PCF-252b5) with a length of 2 m were connected to a tunable laser and a power meter, respectively, by employing a standard SMF. Arc discharge with low current and short fusion duration occurred repeatedly to collapse air holes at the center of the PCF by use of a fusion splicer. The transmission loss measured was about 0.6 dB at 1550nm after air holes were completely collapsed at the center of the PCF. The transmission attenuation of the collapse-induced step-index fiber can be ignored due to the Ge-doped core and a short length of ~ 200 µm. Hence half of the loss, i.e. ~0.3 dB, was attributable to the light transition within one of the two symmetrical gradually collapsed PCF sections.
For the previously discussed splicing methods [8–10], a clear interface usually occurs at the splice joint due to the two distinct fiber structures. Such an interface may cause a high light reflection, which is disadvantageous for some sensing applications of PCFs. For example, when a PCF that has an FBG is spliced to an SMF, the light reflection from the splicing interface worsens the sensing resolution. In contrast, in our splicing method, since the two fiber ends were completely fused into an ideal waveguide, no splice interface was observed at our splice joints. Therefore, low or no light reflection is expected at our splice joint. Moreover, our splice joints have a high bending strength due to the complete fusion of the fiber ends. For instance, as the bending radius was decreased to 4 mm, the splice joint did not break, whereas the bare PCF would break near the joint.
It is a well-known fact that optical fibers should be carefully cleaved before splicing and the splice loss strongly depends on the quality of the fiber end face. In contrast, during our splicing process, even poorly prepared end faces were tolerable and low splice losses can be achieved. As shown in Fig. 4, a small core Ge-doped PCF (PCF-252b5) with a poor end face was spliced to a standard SMF (SMF-1) and a low splice loss of 1.2 dB was achieved after 4th arc discharge. This is due to the fact that during the splicing process the fiber end is firstly fused into a spherical end shape, as shown in Fig. 2.
As shown in both Fig. 3 and Table 2, the splice loss from SMF to PCF is lower than that from PCF to SMF in all cases (even when the core diameter of the PCF is smaller than that of the SMF). This effect can be explained by the fact that the PCF is not completely a single mode fiber. In the 1550 nm wavelength region, the PCF can transmit higher order modes. In the case of light coupling into an SMF, only the fundamental mode can be coupled efficiently. Higher order modes will not be guided in the SMF; therefore, the observed coupling efficiency will be limited. Due to this limitation, the non-reciprocal losses shown in Fig. 3 are understandable. The non-reciprocal loss phenomenon was also observed when a photonic bandgap fiber was spliced to a standard SMF .
It has been shown that Ge-doping in the core of PCFs not only offers some advantages, such as photosensitivity for the grating inscription, but also improves the possibility of splicing with conventional SMFs. A small core Ge-doped PCF was spliced to different conventional SMFs using a commercial arc fusion splicer with default discharge parameters. These parameters were set for the splicing of two standard SMFs. Additional parameter adjustments were not required during our splicing process and a low splice loss of 1.0~1.4 dB was achieved from a standard SMF to a PCF. Low or no light reflection is expected at the splice joint. Compared with the splicing techniques described previously, our method has the advantages of simple operation, great flexibility, and good reproducibility. Moreover, the splice joint exhibits a high bending strength.
This work was supported by the Alexander von Humboldt Foundation, the National Science Foundation of China and the Thuringian Ministry of Education and Cultural Affairs.
References and links
3. Y. Wang, L. Xiao, D. N. Wang, and W. Jin, “In-fiber polarizer based on a long-period fiber grating written on photonic crystal fiber,” Opt. Letters 32, 1035–1037 (2007). [CrossRef]
5. A. Efimov, A. J. Taylor, F. G. Omenetto, A. V. Yulin, N. Y. Joly, F. Biancalana, D. V. Skryabin, J. C. Knight, and P. S. J. Russell, “Time-spectrally-resolved ultrafast nonlinear dynamics in small-core photonic crystal fibers: Experiment and modelling,” Opt. Express 12, 6498–6507 (2004). [CrossRef] [PubMed]
6. J. Zagari, A. Argyros, N. A. Issa, G. Barton, G. Henry, M. C. J. Large, L. Poladian, and M. A. van Eijkelenborg, “Small-core single-mode microstructured polymer optical fiber with large external diameter,” Opt. Lett. 29, 818–820 (2004). [CrossRef] [PubMed]
7. P. J. Bennett, T. M. Monro, and D. J. Richardson, “Toward practical holey fiber technology: fabrication, splicing, modeling, and characterization,” Opt. Lett. 24, 1203–1205 (1999). [CrossRef]
8. L. Xiao, W. Jin, and M. S. Demokan, “Fusion splicing small-core photonic crystal fibers and single-mode fibers by repeated arc discharges,” Opt. Lett. 32, 115–117 (2007). [CrossRef]
9. L. Xiao, M. S. Demokan, W. Jin, Y. Wang, and C.-L. Zhao, “Fusion splicing photonic crystal fibers and conventional single-mode fibers: microhole collapse effect,” J. Lightwave Technol. 25, 3563–3574 (2007). [CrossRef]
10. B. Bourliaguet, C. Paré, F. Émond, A. Croteau, A. Proulx, and R. Vallée, “Microstructured fiber splicing,” Opt. Express 11, 3412–3417 (2003). [PubMed]
11. J. H. Chong, M. K. Rao, Y. Zhu, and P. Shum, “An effective splicing method on photonic crystal fiber using CO2 laser,” IEEE Photon. Technol. Lett. 15, 942–944 (2003). [CrossRef]
12. A. D. Yablon and R. T. Bise, “Low-loss high-strength microstructured fiber fusion splices using GRIN fiber lenses,” IEEE Photon. Technol. Lett. 17, 118–120 (2005). [CrossRef]
13. R. Thapa, K. Knabe, K. L. Corwin, and B. R. Washburn, “Arc fusion splicing of hollow-core photonic bandgap fibers for gas-filled fiber cells,” Opt. Express 14, 9576–9583 (2006). [CrossRef] [PubMed]