We propose a novel technique for the fabrication of long-period fiber gratings (LPFGs) using ultrafast laser pulses coupled into opposite ends of an optical fiber. The pulses counter-propagate in the fiber and induce a permanent refractive index change where they overlap. Based on absorption rate measurements, it was confirmed that the pulses were partially absorbed at the overlapping position, with a maximum absorption rate of 6%. Furthermore, a LPFG with a smooth transmission spectrum and low background loss was successfully fabricated using this technique.
© 2017 Optical Society of America
Long-period fiber gratings (LPFGs), which act as wavelength-dependent attenuators, are important devices in optical communication and sensing . LPFGs are structures periodically inscribed along the inside of an optical fiber that modulate the refractive index of the fiber core with a typical period of hundreds of microns. Various LPFG fabrication techniques have been reported, such as the use of ultraviolet (UV) lasers [1–3], CO2 lasers [4,5], and near-infrared (NIR) ultrafast lasers [3,6–13].
An NIR ultrafast laser tightly focused inside a transparent material can induce a permanent refractive index change at the focal point through nonlinear processes that occur under high optical intensity [14,15], and this phenomenon is utilized in LPFG fabrication. In general, the fiber is irradiated from the direction perpendicular to its axis, and the laser beam is focused onto the fiber core through the cladding. LPFGs fabricated using this technique have high temperature resistance and better aging stability compared with LPFGs fabricated using UV lasers . Furthermore, this technique is applicable to both photosensitive and photo-insensitive fibers. However, it requires highly precise alignment between the focal point and the fiber core [3,6–8]. A slight misalignment can cause degradation of LPFG transmission spectrum with large background loss. Thus, complex alignment systems and processes have been required to avoid misalignment.
In this paper, we propose a novel NIR ultrafast laser-based LPFG fabrication technique that overcomes the misalignment issue, in which two beams are coupled into each end of a fiber. In addition, this technique does not require a fiber jacket removal process. We demonstrate the formation of LPFGs through laser pulse absorption measurements, and then fabricate and characterize a LPFG. To our knowledge, this is the first time that LPFGs have been prepared using such a technique.
2. LPFG fabrication using end-coupled pulses
Figure 1(a) shows schematic illustrations of the proposed technique, in which NIR ultrafast laser pulses are coupled into both ends of an optical fiber. The pulses counter-propagate [Fig. 1(b)] and overlap in the fiber. The high optical intensity in the pulse overlap region induces energy absorption [Fig. 1(c)], which results in a permanent change in the refractive index [Fig. 1(d)] of the fiber core. The overlapping position of the pulses can be shifted by changing the time delay between the pulses. Hence LPFGs can be fabricated by continuously sweeping the overlapping position [Fig. 1(e)] over hundreds of microns, and forming such regions repeatedly and periodically. Using this technique, even if the pulses coupled into the ends of the fiber are slightly misaligned, only pulses successfully coupled into the fiber core can overlap. Consequently, the misalignment issue discussed above that leads to degraded LPFG transmission spectra is avoided.
3. Experimental setup
The experimental setup used to measure the laser pulse absorption and fabricate the LPFG is shown in Fig. 2(a). An 100-kHz ultrafast laser system (Hamamatsu L11590) that generates pulses with a 1030-nm center wavelength and a 3-ps pulse width was used. The pulses were passed through a computer-controlled variable optical attenuator (VOA), a computer-controlled mechanical shutter, and a beam expander (3 × ), and were then split into two optical paths (upper and lower paths) with a half-mirror. In each path, a VOA was used for fine adjustment of the pulse energies, which were adjusted to be equal in the two paths. All of the three VOAs consisted of a half-wave plate and a polarizing beamsplitter. Quarter-wave plates were used to convert the polarization state of the pulses to circular polarization. The pulses were coupled from the ends of a single-mode fiber (Thorlabs SM980-5.8-125, operating wavelength 980–1550 nm, fiber length 0.5 m) through objective lenses (NA = 0.1, 5 × ). In the coupling process, the energy of the pulses was sufficiently low to prevent damage to the fiber ends, and the ends were aligned to maximize the transmittance of the pulses. In order to eliminate the influence of fiber bending during the experiment, a holder made of glass plates was attached to the center of the fiber, as shown in Fig. 2(b). This kept a 8-cm region of the fiber containing the pulse overlapping position straight. The overlapping position could be confirmed by visible light emission which occurs when the energy of the pulses were increased moderately. In the upper path, an optical delay line consisting of a retroreflector on a computer-controlled linear stage was used to change the optical path length. Moving the retroreflector using the linear stage changes the delay between the pulses and thus the overlapping position of the pulses in the fiber. Here, the relationship between the translation distance of the retroreflector d and the overlapping position L is represented by L = d/ne, where ne is the effective refractive index of the fiber core, which is approximately 1.46. A non-polarizing beamsplitter (NPBS) and a photodiode (PD) in the lower path were used to measure the laser pulse absorption, which will be described in detail in the next section. The average optical power was measured by inserting an optical power meter prior to the objective lenses.
4. Measurement of the laser pulse absorption
4.1 Experimental procedure
To measure the absorption of the pulses at the overlapping position, the output values of the PD that receives the pulses reflected by the NPBS were measured using a digital multimeter (Agilent Technologies 34410A) under the following three conditions. The first value Pu was measured under the condition that a beam blocking device was inserted in the lower path, i.e., only the pulses in the upper path were coupled into the fiber and received at the PD after passing through the fiber. The second value Pb was measured under the condition that the beam blocking device was removed, i.e., the pulses in both paths were coupled into the fiber and overlapped in the fiber. Roughly speaking, a difference between Pb and Pu indicates that absorption occurred at the overlapping position, and 1 – Pb/Pu represents the absorption rate of the pulses. However, Pb contains background noise due to the fraction of the pulses of the lower path back-reflected by the objective lens and the end of the fiber. Therefore, to determine the absorption rate with high accuracy, it is necessary to eliminate the background noise. The third value Pl, which corresponds to the background noise, was measured with the beam blocking device inserted in the upper path, i.e., only the back-reflected pulses from the lower path were measured. In terms of these three quantities, the absorption rate can be expressed as 1 – (Pb – Pl)/Pu. In this experiment, the absorption rate was repeatedly measured while gradually increasing the energy of the pulses in one fiber sample, and the overlapping position was shifted by 200 μm for each new measurement so that a fresh position of the sample was used. The polarization state of the pulses was set to circular polarization to eliminate any influence of the rotation of the polarization plane during propagation in the fiber. Two sets of circular polarization states were used in this experiment: right circular polarization in both paths, and right circular polarization in the upper path and left circular polarization in the lower path.
Figure 3 shows the absorption rate of the pulses as a function of pulse energy when right circular polarization was used in both paths. Three fiber samples were used to confirm the reproducibility of the measurements, and similar curves were obtained for all the samples. Based on the curves, the threshold energy for absorption of the pulses was about 0.5 μJ/pulse, and the maximum absorption rate of 6% occurred at an energy of 0.7 μJ/pulse. When the energy exceeded 0.7 μJ/pulse, the absorption rate could not be reliably measured because the ends of the fiber were damaged.
After the measurement, we examined the inside of the fibers using a digital microscope (Keyence VHX-100), but found that no change in the refractive index at the overlapping positions of the pulses could be observed. Therefore, it is thought that the refractive index change of the fiber was very small.
Figure 4 shows the absorption rate of the pulses as a function of pulse energy when right circular polarization in the upper path and left circular polarization in the lower path were used. As is clear from the figure, the results differ significantly from those shown in Fig. 3. The variation among the three samples was large, and the absorption rate could not be measured beyond about 1%. The reason was not damage to the fiber ends but rather the occurrence of a crack in the fiber, as shown in Fig. 5. The width of the crack was about 7 μm, and similar cracks were observed in all three samples. Although the mechanism is unknown, it is thought that the absorption of energy increased rapidly after the absorption rate exceeded 1%, and the large amount of absorbed energy induced the cracks. It is noteworthy that the crack was inside the fiber core. That is, if it is possible to form a region of refractive index change, that region is considered to be only inside the core. These results show that the absorption characteristics were strongly dependent on the polarization state of the pulses.
5. Fabrication of a LPFG
5.1 Experimental procedure
To demonstrate our LPFG fabrication technique, we attempted to inscribe a LPFG with length, a period and duty cycle of 6 cm, 600 μm and 0.5, respectively, in a fiber, using the experimental setup shown in Fig. 2(a). Right-circularly polarized pulses in both paths were coupled into the fiber, and the pulse energy was set to 0.66 μJ. This pulse energy corresponds to an absorption rate of about 4%, as shown in Fig. 3. The pulses were intermittently irradiated by repeatedly (100 times) opening and closing a mechanical shutter for 3 s each time. During that time, the overlapping position of the pulses was continuously shifted at a speed of 100 μm/s by translating the retroreflector.
The transmission spectrum of the processed fiber sample is shown in Fig. 6. The spectrum corresponds to the difference between the spectra before and after processing, which were measured using an optical spectrum analyzer (ANDO AQ-6315B) and a white light source (ANDO AQ-4303B). From the figure, several rejection bands due to the LPFG were observed. The spectrum had a smooth shape and the background loss in the wavelength range of 1100 nm to 1200 nm was only about 0.1 dB. Based on these results, it is considered that the periodic refractive index change regions were successfully formed in the sample. The maximum depth of the rejection band of 2.1 dB at 1448 nm was smaller than that in previous studies (e.g., ), which is thought to be because the refractive index change in the fiber was very small. To improve the performance of the fabricated LPFGs, it is necessary to know the relationship between the processing conditions and the degree of the refractive index change.
We have proposed a novel LPFG fabrication technique using end-coupled NIR ultrafast laser pulses, in which the pulses counter-propagate in the fiber and induce a permanent change in the refractive index of the core at their overlapping position. The primary advantage of this technique is that it avoids degradation of the LPFG transmission characteristics, without requiring highly precise alignment during the fabrication process. In order to demonstrate this technique, we first measured the absorption rate of the pulses at their overlapping position, and then fabricated a LPFG. The results of the absorption rate measurements showed that the pulses were absorbed up to 6% before the fiber was damaged, and the absorption characteristics were strongly dependent on the polarization state of the light. Furthermore, the fabricated LPFG had a smooth transmission spectrum with low background loss. By further clarifying the processing mechanism underlying this technique, it will be possible to fabricate LPFGs with better transmission characteristics. Moreover, we believe that other than LPFGs, such as fiber Bragg gratings, can be fabricated using this technique in the future.
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(1), 58–65 (1996). [CrossRef]
2. A. I. Kalachev, V. Pureur, and D. N. Nikogosyan, “Investigation of long-period fiber gratings induced by high-intensity femtosecond UV laser pulses,” Opt. Commun. 246(1–3), 107–115 (2005). [CrossRef]
3. D. N. Nikogosyan, “Multi-photon high-excitation-energy approach to fiber grating inscription,” Meas. Sci. Technol. 18(1), R1–R29 (2007). [CrossRef]
4. D. D. Davis, T. K. Gaylord, E. N. Glytsis, S. G. Kosinski, S. C. Mettler, and A. M. Vengsarkar, “Long-period fibre grating fabrication with focused CO2 laser pulses,” Electron. Lett. 34(3), 302–303 (1998). [CrossRef]
5. Y. Wang, “Review of long period fiber gratings written by CO2 laser,” J. Appl. Phys. 108(8), 081101 (2010). [CrossRef]
6. Y. Kondo, K. Nouchi, T. Mitsuyu, M. Watanabe, P. G. Kazansky, and K. Hirao, “Fabrication of long-period fiber gratings by focused irradiation of infrared femtosecond laser pulses,” Opt. Lett. 24(10), 646–648 (1999). [CrossRef] [PubMed]
7. F. Hindle, E. Fertein, C. Przygodzki, F. Dürr, L. Paccou, R. Bocquet, P. Niay, H. G. Limberger, and M. Douay, “Inscription of long-period gratings in pure silica and germano-silicate fiber cores by femtosecond laser irradiation,” IEEE Photonics Technol. Lett. 16(8), 1861–1863 (2004). [CrossRef]
8. N. Zhang, J. J. Yang, M. W. Wang, and X. N. Zhu, “Fabrication of long-period fibre gratings using 800 nm femtosecond laser pulses,” Chin. Phys. Lett. 23(12), 3281–3284 (2006). [CrossRef]
9. T. Allsop, K. Kalli, K. Zhou, Y. Lai, G. Smith, M. Dubov, D. J. Webb, and I. Bennion, “Long period gratings written into a photonic crystal fibre by a femtosecond laser as directional bend sensors,” Opt. Commun. 281(20), 5092–5096 (2008). [CrossRef]
10. B. Li, L. Jiang, S. Wang, H. L. Tsai, and H. Xiao, “Femtosecond laser fabrication of long period fiber gratings and applications in refractive index sensing,” Opt. Laser Technol. 43(8), 1420–1423 (2011). [CrossRef]
11. X. Sun, P. Huang, J. Zhao, L. Wei, N. Zhang, D. Kuang, and X. Zhu, “Characteristic control of long period fiber grating (LPFG) fabricated by infrared femtosecond laser,” Front. Optoelectron. 5(3), 334–340 (2012). [CrossRef]
12. J. C. Guo, Y. S. Yu, X. L. Zhang, C. Chen, R. Yang, C. Wang, R. Z. Yang, Q. D. Chen, and H. B. Sun, “Compact long-period fiber gratings with resonance at second-order diffraction,” IEEE Photonics Technol. Lett. 24(16), 1393–1395 (2012). [CrossRef]
13. F. Ahmed, H. E. Joe, B. K. Min, and M. B. G. Jun, “Characterization of refractive index change and fabrication of long period gratings in pure silica fiber by femtosecond laser radiation,” Opt. Laser Technol. 74, 119–124 (2015). [CrossRef]