Fusion splicing of plastic optical fibers using a mid-IR fiber laser

This work demonstrated the fusion splicing of plastic optical fibers (POFs) using a 2.8 μm continuous-wave fiber laser. This mid-IR laser-based fusion splicing technique does not require the use of adhesives or any other treatments. The performance of the proposed method was investigated by assessing the optical transmission, tensile strength and bending strength values of POF specimens after fusion bonding while employing various splicing conditions. An optical transmission of 0.76 was obtained by splicing under appropriate conditions. A minimum bending radius and tensile strength of the POF samples were found to be 1.5 mm and 13.5 N, respectively.

M id-infrared (IR) laser sources [1][2][3][4][5][6] are gaining recognition as a key technology for many applications in spectroscopy, 1) surgery 7,8) and chemical sensing. 9,10) Over the last decade, there was significant progress in terms of increasing the power outputs of mid-IR fiber lasers emitting at approximately 3 μm. [11][12][13][14][15][16][17] Such devices have potential industrial uses, such as in systems involving free-space transmission over large distances in the atmosphere and selective absorption by specific materials or molecules. Our own group previously developed a stable, high-power, industrial-grade 2.8 μm erbium-doped fiber laser 6) and has since focused on exploring applications for this technology. One possible use for this laser is the fusion processing of plastic optical fibers (POFs), which are likely to be incorporated into future communication systems. POFs provide various advantages, including cost-efficient connections, improved safety, high flexibility and the ability to fabricate sensing/monitoring devices. [18][19][20] To date, the technologies investigated for the fusion splicing of POFs have included electric arc splicing and thermal splicing with poly (ether ether ketone) tubes. 21) However, although electric arc splicing has been used to process silica glass fibers, the high temperatures associated with this technique can thermally degrade POFs. A CO 2 laser 22) could potentially be employed as an alternative means of POF splicing, although some low melting point polymers such as poly (methyl methacrylate) (PMMA) exhibit overly high absorption of CO 2 laser light. To achieve uniform heating along the radial direction of the fiber, the laser beam has to be split using optical separators to provide multi-pass irradiation because the amount of energy imparted by a 10.6 μm CO 2 laser is too high (Fig. 1). For more uniform heating, lasers with wavelengths of relatively low absorbance are suitable. In contrast to commercially available CO 2 lasers, the Er fiber laser constructed in our laboratory that operates at a wavelength of 2.8 μm provides high beam quality with an M 2 value of approximately 1.0 and stable output power (with a root mean square value of 0.25%) at an output power of 30 W. 6) The present work demonstrated the fusion splicing of POFs using a 2.8 μm continuous-wave Er fiber laser operating in the 5 W range. This device is able to process POFs without adhesives or any other treatments. In addition, there is no need to split the laser beam irradiating the POFs because the resins in these fibers exhibit moderate energy absorption at 2.8 μm. The performance of the proposed method was evaluated by assessing the optical transmission, tensile strength and bending strength of POFs after fusion bonding.
The work reported herein employed commercially available POFs (ESKA-GCK-20E, Mitsubishi Chemicals) made from PMMA and having a diameter of 0.5 mm. Figure 2 presents a drawing of the experimental setup used for POF splicing. Here, Figs. 1(b) and 1(c) correspond to views along the Y and Z axes, respectively, as defined in the diagram. In this apparatus, a CaF 2 lens with a focal length of 20 mm focused a 2.8 μm laser beam to provide a spot size of 15 μm at the focal point. Based on the results of preliminary experiments, the laser focal point was positioned at a distance of 1.0 mm far from the opposite surface of the POF relative to the position of the laser, as can be seen from Fig. 2(b). This configuration was found to provide uniform heating of the fiber. Note that when the focal point was closer to the laser source than the POF, the fiber end bent toward the laser source and the fusion splicing failed. Figure 3 provides diagrams and photographic images acquired using a microscope showing the steps involved in the splicing. In this process, two fibers (POF1 and POF2) were mounted on individual three-axis, manually-operated translation stages [ Fig. 3(a)]. The POFs were positioned so that their ends faced one another and the optical axes of the two fibers were aligned while observing the POFs along the Y and Z axes using a microscope. As shown in Fig. 3(b), the POFs were subsequently moved so that the tips almost came into contact and the fiber tips were simultaneously pre-heated by applying laser light at a power of 80 mW. Following this pre-heating, the laser power was increased to 120 mW and the fiber tips were simultaneously forced together by moving the translation stage over a set distance (Δx). At this time, the optical axes of the two fibers [ Fig. 3(c)] can be re-aligned after bonding, and the splice loss can be reduced by stopping laser irradiation when the transmission is at its maximum. In such splicing process, no hot push delay is given and heating time is determined arbitrarily. Trials were performed in which Δx was varied and the light transmittance, tensile strength and bending strength of the resulting specimen in each case were determined. Figure 4 plots the allowable bending radius values of the spliced POF specimens as a function of x.
D The allowable bending radius is defined here as the maximum bending radius at which the fiber does not break. POFs were bent by being forced onto resin rods having radii ⩾1.5 mm, as shown in Fig. 4. These values provided an indication of the bending strength of the spliced POFs, with a smaller bending radius showing greater strength. As can be seen from Fig. 4, the allowable bending radius was decreased with increases in Δx and the minimum radius was found to be less than 1.5 mm for Δx ⩾ 0.375 mm. These results confirm that the bending strength was correlated with Δx, presumably because the fiber diameter at the fusion point, D, increased with increases in Δx, as shown in Fig. 5.
The tensile strengths of the fusion points in the various specimens were evaluated using a force gauge while fixing each POF sample with two clamps. In these trials, three spliced POF specimens were produced and tested for each value of Δx. Each spliced fiber was installed in the test device and tension was applied to the fusion point by manually rotating a wheel as indicated in Fig. 6(a). As shown in Fig. 6(b), the tensile strength increased up to 13.5 N with increases in Δx. This was the maximum force that could be measured because higher loads caused the fiber itself to elongate, so the values plateaued at 13.5 N for Δx ⩾ 0.375 mm. Based on the above data and assuming a tensile strength of 13.5 N for the original POF, values of 4.92 and 17.2 N m −2 were calculated for the fusion points and the original POFs, respectively. This result demonstrates that the tensile strength at the fusion points was sufficiently high. However, it should be noted that the higher value obtained for the fusion points can be at least partly attributed to the increased diameter at that point compared with the original POF diameter.
The optical transmission properties of five spliced POFs at the fusion point were evaluated. In these trials, a laser diode (LD, Civil Laser, LSR660-NL) and a photodiode detector (PD, Ophir, PD300-1W) were positioned on either side of the splicing point between two joined fibers, as shown in Fig. 7(a). It should be noted that the length of the fiber labeled POF2 in this diagram was at least 3 m to avoid guiding cladding modes to the PD. To assess the net splice loss, the propagation loss and the coupling loss to POFs were eliminated by comparing the transmitted power before and after splicing. In contrast to the allowable bending radius and tensile strength values, both of which increased with increases in Δx, the optical transmission underwent a gradual and intermittent decrease, as shown in Fig. 7(b). Figure 7(c) provides a photographic image of a fusion point during irradiation with the LD as acquired through a microscope. The decrease in transmittance is attributed to an increase in the effect of mode mismatch at the fusion point as the diameter at this point increased. That is, the transmission loss at the fusion point was reduced by decreasing Δx. The maximum transmission was found to be 0.76 at Δx = 0.063 mm. This value of Δx was the minimum that could be obtained from manual operation of the stages used to manipulate the POFs during splicing. The difference between maximum and minimum transmission values was reduced by increasing Δx (with the exception of the data for Δx = 0.875 mm), indicating that the process became more reproducible.
In conclusion, this work demonstrated the fusion splicing of POFs using a 2.8 μm continuous-wave fiber laser without any requirement for adhesives or other processes. The optical transmission, tensile strength and bending strength values of POFs after fusion splicing were determined. Although the transmission properties of the present spliced specimens would be insufficient for applications requiring high-quality transmission, the tensile strength and bending strength were suitable for practical use. The proposed technology could therefore contribute to the future development of POF networks. Furthermore, it is expected that the optical transmission of these spliced POFs could be improved by reducing Δx based on employing an automatic stage. Fiber properties satisfying the requirements for practical applications could be obtained by optimizing both Δx and the heating process including laser irradiation conditions in future work.