Various types of pre- and post-treatments to optical fibers are typically used to improve and/or change the properties of fiber Bragg grating (FBG). Here, we investigate experimentally the effects of NIR femtosecond (fs) laser pulses used for pre- and post-treatment on the refractive index of the fiber, and the resulting center wavelength shift of an fs inscribed FBG. We observe “red”-shift when applying a suitable pre-treatment, and both “blue”- and “red”-shifts when applying post-treatment. We characterize the photo-treatment parameters and compare to an FBG inscribed on a fresh fiber without any treatment. We also show that when the photo-treatment is saturated the result is a phase-shifted grating. Our results give further insight into the process of fs photo-treatments and quantify the effects in the case of fs FBG inscription.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Femtosecond induced refractive index change in transparent materials has been under extensive research in the past two decades. Numerous devices and applications such as waveguides, couplers and fiber Bragg gratings (FBGs) were reported . Specifically, FBGs were widely explored and grating inscription in various types of fibers and with various types of focusing conditions have been demonstrated. Generally, FBGs were fabricated by using either the point-by-point (PbP) technique in which grating planes are formed one by one, or by the phase mask (PM) technique where the grating is inscribed as one whole section [2,3].
Two types of NIR femtosecond induced index changes in SMF-28 were identified; type I IR, and type II IR . They differ mainly within their inscription intensity threshold and annealing behavior. The intensity threshold for type I IR and type II IR modification is , below the damage threshold and , above the damage threshold, respectively . The reported index change for type I gratings is isotropic and on the order of , while the reported index change for type II gratings is higher and usually anisotropic. Type I gratings are also characterized by their ability to be thermally erased (annealed) at ~, while type II gratings can withstand annealing of up to a temperature of without any serious degradation .
In addition, various types of treatments to the fiber before and after the FBG inscription process have been investigated as well. For example, hydrogen loading is well known to enhance the fiber sensitivity for inscription [5,6] and it is typically used in traditional UV FBG inscription . Post-fabrication thermal annealing was used to narrow the grating bandwidth . Wavelength tuning and spectral modification of the Bragg gratings made with a traditional UV laser source, have been investigated with pre-treatment at UV wavelengths and post-treatment at NIR wavelengths [9,10]. Photo-annealing of gratings fabricated with fs pulses in doped fibers and glasses were reported as well. Rapid annealing of type II fiber Bragg gratings inscribed with fs pulses in Yb-doped fibers was demonstrated when employed as high reflectors in pulsed Q-switched fiber lasers . Photo-annealing of a waveguide Bragg structure in phosphate glass doped with Yb was investigated as well . Recently, in our group full “erasure” of an FBG and complete “immunity” to FBG inscription has been demonstrated with NIR fs laser photo-treatment .
In this paper, we experimentally investigate the effect of pre- and post-treatment with fs 800 nm pulses on the properties of the fs-inscribed FBG in silica SMFs. We investigate the effect of different pulse energies and different time durations, both with pre- and post- treatments, on the FBG central wavelength shift and we compare the experimental results with coupled-mode theory. We further investigate index-change saturation effects due to the Gaussian profile of the inscription beam.
2. Experimental setup and procedure
Our experimental setup (see Fig. 1) includes an amplified Ti-sapphire laser system (Coherent Legend Elite) producing pulses with energy of 3.5 mJ and 35 fs duration at 1 kHz repetition rate and a center wavelength of 800 nm. The pulse energy is adjusted by using a half wave plate and a polarizer, and the average power is measured with a power meter. The Gaussian beam diameter is ∼8 mm and it is focused, on the fiber core, through a PM with a 40 mm cylindrical lens. The PM, made by Ibsen Ltd., has a pitch of 2140 nm, corresponding to a second-order Bragg grating at a wavelength of 1550 nm. Less than 5% of the energy goes to the zero order, while ~75% of the energy diffracts to the ± 1 orders. The mask is positioned ~2 mm in front of the fiber to ensure pure two-beam interference . The fiber is held in two grooves, each on a three-axis stage and under controlled tension, which provides high repeatability of the inscription system. The grating size in the fiber is calculated to be ~5.1 µm x 6 mm, covering only part of the fiber mode field diameter, which is ~9 µm in SMF fibers. Therefore, during the fs FBG inscription, we move the cylindrical focusing lens with a linear DC actuator (PI M-230.25) and scan ± 10 µm around the core, perpendicular to both the fiber and the inscription beam axes, in order to achieve proper coverage and hence maximum reflectivity. In the other dimension (inscription beam axis), the extent of the Rayleigh range is significantly larger than the mode field diameter, so we do not move the cylindrical lens in the focusing direction during the inscription. Before the inscription, the fiber is stripped and cleaned from its polymer coating. The fiber is connected to an amplified spontaneous emission (ASE) source on one end and to an optical spectrum analyzer (Yokogawa AQ6370D OSA) on the other end. We monitor the transmission spectra of the inscribed Bragg grating with 20 pm resolution during the inscription process.
The central Bragg wavelength in an optical fiber is given by:
3. Results and discussion
3.1 Pre-treatment with NIR femtosecond laser pulses
First, we inscribe a grating on a fresh fiber (no pre-treatment) with a pulse energy of 370 µJ for 3 min. The result is an FBG with a −54 dB transmission dip at a center wavelength of ~1548.70 nm [see Fig. 2(a)]. We then apply different pre-treatments to fresh fibers by scanning the fibers with fs pulses. The scanning is performed ± 18 µm around the fiber core with a line pattern for ~24 min (~1.4M pulses). We performed five different pre-treatments, each on a separate fresh fiber, with pulse energies of 400 µJ, 450 µJ, 500 µJ, 550 µJ, and 600 µJ, in order to achieve different increased levels of the effective refractive index of the fiber, . Then, we inscribe FBGs with the PM in the same area of the pre-treatment. The resulting FBGs have an increasing “red”-shifted Bragg wavelength (1548.95 nm, 1549.10 nm, 1549.25 nm, 1549.50 nm, and 1549.70 nm); [see Figs. 2(b)-2(f)]. On the other hand, the transmission dip is reduced when increasing the pre-treatment pulse energy due to a smaller refractive index modulation of the inscribed FBG. In addition, we used coupled-mode theory [15,16] to calculate the theoretical transmission spectra of the FBGs and we compared them to our experimental results [see Fig. 2]. It is easy to calculate from Eq. (1) that an increase of to the effective refractive index should result in a ~1 nm shift of the center Bragg wavelength. In Fig. 3 we can see the Bragg grating wavelength and the increase of the effective refractive index of the fiber as a function of the pre-treatment pulse energy. As evident, there is a nearly linear dependence between the wavelength change and the pre-treatment pulse energy.
The refractive index change of the fiber due to the inscription is described by :16]:16]. The bandwidth of the Bragg grating is described by :15,16]) and the bandwidth, of the inscribed FBG. The results are summarized in Table 1.
3.2 Post-treatment with NIR femtosecond laser pulses
To investigate the effect of post-treatment on the spectral transmission of the gratings we first inscribe an FBG with a pulse energy of 370 µJ for 2 min. The result is an FGB with a center Bragg wavelength of 1548.74 nm and a transmission dip of −44 dB [see Fig. 4(a)]. Then, we remove the PM from the inscription beam path, and scan with a line pattern the previously inscribed FBG. We scan with a pulse energy of 400 µJ, ± 18 µm around the fiber core and record the FBG transmission spectra as a function of time (number of pulses). Surprisingly, after 200K pulses the FBG is “blue”-shifted to a central Bragg wavelength of 1548.62 nm and the transmission dip reduces to −35 dB [see bottom graph in Fig. 4(b)]. Then, with increasing the post-treatment duration the FBG is increasingly “red”-shifted while simultaneously the transmission dip is decreased. After one hour of post-treatment (3.6M pulses), the center of the transmission dip seems more flat and the FBG wavelength is shifted to ~1549 nm [see upper graph in Fig. 4(b)]. The reason for the initial “blue” shift is not clear. It may result from an initial annealing process. The subsequent “red”-shift and decrease of the transmission dip is well explained by the increase of the effective refractive index and the decrease in the grating contrast. The flattening of the transmission dip after long treatment durations may be explained by saturation of the central part of the grating due to the Gaussian profile of the inscription beam. The Bragg wavelength and the transmission dip as a function of the number of post-treatment pulses is shown in Fig. 5.
During the femtosecond post-treatment procedure, the effective refractive index of the fiber is increased, while the modulation (contrast) of the refractive index is deceased. This modification to the refractive index causes the Bragg wavelength to “red”-shift on one hand and to the transmission dip to decrease on the other hand. This modification of the refractive index ignoring the Gaussian-apodized shape is illustrated schematically in Fig. 6.
3.3 Saturation of pre- and post-treatment with NIR femtosecond laser pulses
Next, we investigate the regime of highly saturated pre- and post-treatments. First we perform a pre-treatment of the fiber with a line pattern (similar to the prior experiments), but with pulse energy of 650 µJ. Then, when we inscribe an FBG with pulse energy of 370 µJ for 3 min, we observe a phase-shifted grating as can be seen in Fig. 7(a). This result can be explained as follows: because of the Gaussian beam profile, the center part of the FBG is now saturated, and there is no refractive index modulation. The result is a very short mid section with no modulation between two sections of refractive index modulation (two FBGs). This results in a phase-shifted grating. Next, we applied a post-treatment with pulse energy of 450 µJ to an FBG previously inscribed on a fresh fiber. After 45 min (~2.7M pulses) of post-treatment, we get a phase-shifted grating as can be seen in Fig. 7(b). In this case, we already have an FBG in the fiber, before the post-treatment. As we mentioned previously during the post-treatment we increase the refractive index and decrease the contrast (modulation). Again, since the intensity has a Gaussian shape, the center part of the FBG saturates (there is no more refractive index modulation in the center part of the FBG) while on both sides of the center there is still a modulation. Now, like in the case of the pre-treatment there is a short mid section with no modulation between two sections of modulation (two FBGs). We also found that by increasing the post-treatment pulse energy a phase-shifted grating occurs after a shorter period of time.
We have demonstrated that precise “engineering” the effective refractive index of the fiber is feasible with NIR fs pulses. We characterize different pulse energies and treatment durations on the central wavelength shift of fs inscribed FBGs in silica SMF. A maximum DC increase modification of the refractive index of the order of was achieved, which results in a ~1 nm “red”-shift. The FBG wavelength shift is a result of an increase of the effective refractive index of the fiber. While increasing the pre-treatment pulse energy, the refractive index modulation (grating contrast) of the FBG decreases, resulting in a smaller transmission dip. During post-treatment, we observe first a small “blue”-shift of the center FBG wavelength, followed by a continuous “red”-shift, while the FBG transmission dip is reduced with time due to a decrease in the modulation of the refractive index. We also found that the transmission dip decrease and the FBG wavelength shift are both nearly linear to the duration of the post-treatment (number of pulses). In addition, we found that there is a saturation point for both pre- and post-treatment, which results in no modulation of the center part of the FBG. This saturation results in a phase-shifted grating.
Israel Ministry of Industry, Trade, and Labor, ALTIA Magnet program (60882); Israel Ministry of Science, Technology and Space (3-14289).
References and links
1. G. Della Valle, R. Osellame, and P. Laporta, “Micromachining of photonic devices by femtosecond laser pulses,” J. Opt. A, Pure Appl. Opt. 11(1), 013001 (2009). [CrossRef]
2. J. Thomas, C. Voigtlaender, R. G. Becker, D. Richter, A. Tuennermann, and S. Nolte, “Femtosecond pulse written fiber gratings: a new avenue to integrated fiber technology,” Laser Photonics Rev. 6(6), 709–723 (2012). [CrossRef]
3. S. J. Mihailov, D. Grobnic, C. W. Smelser, P. Lu, R. B. Walker, and H. Ding, “Bragg grating inscription in various optical fibers with femtosecond infrared lasers and a phase mask,” Opt. Mater. Express 1(4), 754–765 (2011). [CrossRef]
5. K. A. Zagorul’ko, P. G. Kryukov, Y. V. Larionov, A. A. Rybaltovskii, E. M. Dianov, N. S. Vorob’ev, A. V. Smirnov, M. Y. Shchelev, and A. M. Prokhorov, “Fabrication of a long-period grating in a fiber by second harmonic radiation from a femtosecond Ti:sapphire laser,” Quantum Electron. 31(11), 999–1002 (2001). [CrossRef]
6. P. G. Kryukov, Y. V. Larionov, A. A. Rybaltovskii, K. A. Zagorul’ko, A. Dragomir, D. N. Nikogosyan, and A. A. Ruth, “Long-period fibre grating fabrication with femtosecond pulse radiation at different wavelengths,” Microelectron. Eng. 69(2–4), 248–255 (2003). [CrossRef]
7. P. J. Lemaire, R. M. Atkins, V. Mizrahi, K. L. Walker, K. S. Kranz, and W. A. Reed, “High pressure H2 loading as a technique for achieving ultrahigh UV photosensitivity and thermal sensitivity in GeO2 doped optical fibres,” Electron. Lett. 29(13), 1191–1193 (1993). [CrossRef]
8. J. Shi, Y. Li, S. Liu, H. Wang, N. Liu, and P. Lu, “Bandwidth-narrowed Bragg gratings inscribed in double-cladding fiber by femtosecond laser,” Opt. Express 19(3), 1734–1742 (2011). [CrossRef] [PubMed]
9. G. Simpson, K. Kalli, K. Zhou, L. Zhang, and I. Bennion, “Blank beam fabrication of regenerated type IA gratings,” Meas. Sci. Technol. 15(8), 1665–1669 (2004). [CrossRef]
10. K. Kalli, A. G. Simpson, K. Zhou, L. Zhang, D. Birkin, T. Ellingham, and I. Bennion, “Spectral modification of type IA fibre Bragg gratings by high-power near-infrared lasers,” Meas. Sci. Technol. 17(5), 968–974 (2006). [CrossRef]
11. M. L. Åslund, N. Jovanovic, J. Canning, S. D. Jackson, G. D. Marshall, A. Fuerbach, and M. J. Withford, “Rapid Decay of Type-II Femtosecond Laser Inscribed Gratings Within-switched Yb-Doped Fiber Lasers,” IEEE Photonics Technol. Lett. 22(7), 504–506 (2010). [CrossRef]
12. P. Dekker, M. Ams, G. D. Marshall, D. J. Little, and M. J. Withford, “Annealing dynamics of waveguide Bragg gratings: evidence of femtosecond laser induced colour centres,” Opt. Express 18(4), 3274–3283 (2010). [CrossRef] [PubMed]
14. C. W. Smelser, D. Grobnic, and S. J. Mihailov, “Generation of pure two-beam interference grating structures in an optical fiber with a femtosecond infrared source and a phase mask,” Opt. Lett. 29(15), 1730–1732 (2004). [CrossRef] [PubMed]
15. H. Venghaus, Wavelength Filters in Fibre Optics (Springer, 2006), Chap. 5.
16. T. Erdogan, “Fiber Grating Spectra,” J. Lightwave Technol. 15(8), 1277–1294 (1997). [CrossRef]