A one-step type II photosensitivity process has been optimized for inscribing strong >30–dB first-order Bragg-gratings during laser formation of buried waveguides in borosilicate glass. Mode profiles, propagation losses, waveguide birefringence and transmission and reflection spectra were systematically studied in the 1550-nm telecom band over a wide range of laser exposure conditions. Low-loss and birefringence-free waveguides are reported for a narrow laser processing window of 1.0 ± 0.2 ps pulse duration, yielding highly stable Bragg resonances to temperatures up to 500°C.
©2007 Optical Society of America
Ultrashort duration laser pulses have been widely studied [1–3] as a new means for photowriting of optical waveguides in flexible three-dimensional (3-D) geometry inside the bulk of various transparent materials. While numerous optical circuit devices such as couplers [4, 5], waveguide amplifiers , and multi-mode interference waveguides  have been demonstrated, the formation of strong Bragg gratings directly within such waveguides have eluded many investigators due possibly to a saturated photosensitivity response. Nevertheless, such gratings are highly desirable for providing narrow-spectrum filters, mirrors, couplers, multiplexers, taps and sensing elements that define the basic functions required, for example, in passive optical networks (PON).
Ultraviolet  and ultrafast lasers  readily generate Bragg gratings in optical fibers by holographic, phase mask, or point-by-point writing means [10, 11], where the pre-existing waveguide core often facilitates strong photosensitivity enhancement from rare-earth element dopants and hydrogen soaking. In contrast, only weak or high-order Bragg gratings have been reported in waveguides first formed in bulk transparent materials with ultrashort pulse lasers. Kamata and Obara  proposed using a second laser exposure step to overlay gratings and generated weak 10th order gratings. Marshall et al.  used a different 2-step exposure method to produce weak 2nd order Bragg gratings at ~1550-nm telecom wavelength.
We reported recently a single-step laser process for fabricating buried optical waveguides and Bragg grating structures simultaneously . These segmented waveguides were created in boro-aluminosilicate glass with weakly focused [0.25 numeric aperture (NA)] laser pulses of 200 to 300-fs duration. Isolated refractive-index voxels were defined by single laser pulses through a type II photosensitivity response, and formed into periodic arrays to yield 11-dB transmission and 36% reflection peaks of 0.08-nm linewidth (3 dB) for a 50-mm long grating. These Bragg grating waveguides (BGW) were easily formed in the 1300 – 1600 nm band.
In this paper, we present a three order of magnitude improvement on BGW grating strength fabricated by a single laser exposure scan. Propagation loss and spectral response were systematically characterized while exploring tighter focusing and longer pulse duration (2 ps) exposure conditions. Optimum laser processing conditions for fabricating low-loss and high-strength BGWs are provided together with means for tuning or eliminating BGW birefringence. Thermal annealing studies reveal high stability of both optical guiding and Bragg wavelength resonance.
The laser beam delivery system is shown in Fig. 1(a). A 1-kHz Ti:Sapphire ultrafast laser system (Spectra Physics Spitfire Pro) delivered 2.4-mJ maximum energy at 800-nm wavelength. The laser pulse duration was adjusted from 100 fs to 2 ps by tuning the compressor grating in the amplifier and confirming duration with an autocorrelator (APE Pulse Check). An aspheric lens of 0.55 NA was mounted on a vertically positioned linear stage (Aerotech ALS130) and placed to focus the laser beam to ~1-μm diameter (1/e2 intensity) at a position 200 μm below the surface of a boro-aluminosilicate glass sample (Corning EAGLE2000, 50 mm × 10 mm × 1 mm). Glass samples were mounted on 2-D air-bearing motion stages (Aerotech ABL1000, 2-nm resolution and 50-nm repeatability) and scanned perpendicularly to the laser direction to yield 10-mm long waveguides. The scan speed was fixed at 0.52 mm/s to yield 1st-order Bragg reflection at 1550 nm for the 1-kHz laser repetition rate. The laser energy was adjusted from 0.5 to 10 μJ with a half-wave plate and polarizer. Laser polarization was perpendicular to the sample scan direction.
After laser exposure, the glass sample was ground and polished at both end facets. The BGWs were then characterized in terms of guided modes, propagation losses, transmission spectra, and reflection spectra using the fiber-based arrangement shown in Fig. 1(b). Light from a broadband source (Thorlabs ASE-FL7002, 1530 nm to 1610 nm) or a tunable laser (Photonetics BT, 1500 nm to 1620 nm) was routed through a fiber circulator and butt coupled into the BGW with a single mode fiber for purposes of spectral analysis and mode profiling, respectively. At the exit facet, guided modes of the BGWs at 1560 nm were magnified with a 60X objective lens (not shown in Fig. 1(b) and captured by a phosphor-coated CCD camera (Spiricon SP-1550M). Alternatively, another single-mode fiber was butt coupled to the exit facet to measure transmitted power and record transmission spectra. Reflected light from the BGW was guided via the circulator to the OSA as shown in Fig. 1(b) to record reflection responses. Transmission spectra were normalized relative to the peak transmitted power on the long-wavelength side of the Bragg resonance at 1560 nm, and the BGW insertion loss at 1560 nm was assessed separately by comparison with direct fiber-to-fiber coupling as described in Ref. . The reflection spectra were normalized against the power returned by a high reflectivity fiber reflector (R = 96%). The lens-firing arrangement of Fig. 1(c) provided polarization control of the launch light into the BGWs to assess the waveguide birefringence. Unpolarized light from the broadband source was collimated in free space, then focused into the BGWs by a 30X objective lens. A linear polarizer then excited TM (electric field 90° to sample surface), TE (electric field parallel to surface), and mixed TE/TM modes (45° linear polarization) in the BGWs, while the half-wave plate adjusted the intensity of the input light source. All spectra were recorded with an optical spectrum analyzer (Ando AQ6317B) with 0.01-nm resolution. Index matching fluid was applied at glass-fiber interfaces for all spectral recordings and insertion loss measurements. To test the thermal stability of the BGWs, samples were heated in a tube furnace in several cycles of increasing temperature and optically characterized after cooling to room temperature for changes to mode profile, Bragg wavelength, and grating strength.
3. Results and discussion
3.1 Low-loss waveguide writing
The present BGWs comprised of a periodic array of refractive index voxels with 0.52-μm centre-to-centre separation, as defined by 0.52-mm/s scan speed and 1-kHz laser repetition rate. Each voxel was formed by a single laser pulse driving a type II photosensitivity response . Over the wide 100-fs to 2-ps pulse duration range examined here, continuous and homogenous modification tracks were observed with an optical microscope only for pulse durations in the range of 100 fs to 1.5 ps and pulse energy in the range of 2 to 7 μJ. The laser tracks appeared faint, discontinuous, or invisible for lower pulse energy of < 2 μJ, or appeared inhomogeneous and damaged above 7 μJ. For all pulse durations tested, the lowest propagation losses were found at pulse energies only near 3 μJ, slightly above the ~2-μJ threshold for generating guiding tracks.
The top row of Fig. 2 shows overhead optical microscope images of the laser exposed tracks inscribed with 3-μJ pulse energy, 0.52-mm/s scan speed, and various pulse durations from 100 fs to 2 ps. All waveguide tracks appear ~2-μm wide. The 0.52-μm voxel separation was not resolvable with the 40X microscope objective. Under similar back-lit conditions, the laser tracks appear smooth and brighter than the surrounding unexposed area for pulse durations up to 1.25 ps. Inhomogeneous damage tracks were observed for longer duration. A similar trend is apparent in the cross-sectional end-view of the waveguides shown in the middle row of Fig. 2. The bright near-circular shapes at the top mark the region of positive refractive index change responsible for infrared waveguiding. However, the laser-modified region extends vertically for ~30 μm further below this spot, far exceeding the 3-μm theoretical value for the depth of focus. Nonlinear filamentation and self-focusing effects  as well as spherical aberration at the air-glass interface may account for this elongation. For waveguides written with longer than 1-ps pulse duration, the cross-section was defined by a dark elliptical shape that is indicative of negative refractive index change or formation of inhomogeneous scattering centers. This dark volume increased in size and opacity with increasing pulse duration, and was associated with higher observed waveguide loss.
Guided mode profiles at 1560-nm wavelength are shown in the bottom row of Fig. 2. With the exception of 200 fs, all the modes shown in Fig. 2 could be well-approximated by Gaussian intensity profiles along both horizontal and vertical axes that show only a slightly elongated vertical dimension (aspect ratio ~ 1.1). For 1-ps pulse duration, the 10 μm × 11 μm mode-field diameter suggests a small <0.1-dB coupling loss with standard single mode fiber according to modal overlap calculation . Such facet loss calculation was repeated for each BGW and subtracted from the measured insertion loss to estimate the propagation loss discussed below. The ~2-μm waveguide width here (middle row in Fig. 2) is similar to widths reported with weaker focusing (NA = 0.25) , but provide larger refractive index contrast given the smaller ~68% mode diameter generated with the larger 0.55-NA objective. MODE Solutions software (Lumerical) was used to match this profile to a step-index cylindrical waveguide with 2-μm diameter, and provide an estimate of ΔnDC = ~0.01 for average refractive index change, matching the high end of other reported values [19–21] of ~10-3 to 10-2 for kHz-rate waveguide writing. All waveguides in Fig. 2 were single mode at wavelengths from 1500 nm to 1600 nm.
Figure 3(a) plots the measured propagation loss as a function of pulse duration for 1560-nm wavelength guiding in the BGWs shown in Fig. 2. Two windows for low-loss waveguide writing of ~0.5 dB/cm are apparent at 100-fs and 1-ps pulse duration, between which losses rise significantly to ~3 dB/cm. Losses increased more sharply above 1.0 ps and guiding was no longer observable above 2 ps. The low-loss observations near 100 fs are consistent with several reports [2, 19, 22] of optimum waveguide writing for pulse duration below 200 fs, while the low-loss 1-ps window has only recently been identified in several other materials. Our group noted 0.2 dB/cm losses for 633-nm light in fused silica in a narrow 1-ps window when using a 1-kHz Ti:Sapphire laser . Nejadmalayeri and Herman  reported that pulse duration above 850 fs together with circular polarization were necessary for low-loss waveguide writing in lithium niobate with the same 1-kHz laser system. Watanabe and coworkers  observed optimum waveguide formation with 870-fs duration pulses from a high repetition rate (173-kHz) fiber laser system. One can expect other low-loss laser processing windows to exist at writing speeds outside the present 0.52-mm/s value, which was fixed here to provide a ~1550-nm Bragg resonance.
Figure 3(b) shows the propagation losses of waveguides written with various pulse energies, at pulse durations of 100 fs, 300 fs, and 1 ps. For energies lower than 3 μJ, the refractive index change was not sufficient for strong waveguiding, leading to high insertion losses. For 1-ps pulses, the waveguides become increasingly inhomogeneous for energies greater than 7 μJ, yielding high >3-dB/cm propagation loss. At 100 fs, the losses are less than 1 dB/cm for all the energies above 1 μJ, revealing a wide processing window for low-loss waveguide fabrication. However, mode-field diameters of 12 μm × 14 μm are ~20% larger than observed at 1 ps duration.
3.2 High-strength gratings
Bragg grating responses were systematically characterized with an unpolarized broadband light source for BGWs formed with 1 to 7 μJ laser pulse energy and 100- to 1500-fs pulse duration. Optimized Bragg transmission responses varied from relatively weak (< 5 dB) dual-peaked lines at short 100-fs pulse duration to strong >35-dB single-peaked resonances at long 1.0-ps pulse duration as shown in Fig. 4. In Fig. 4(a) two sharply resolved (0.1-nm wide FWHM) peaks at 1550.9 nm and 1551.1 nm wavelength indicate waveguide birefringence, and present only weak 1.9 dB and 2.6 dB transmission and 31% and 38% reflection resonances, respectively, for 100-fs duration and 3-μJ energy pulses. In contrast, Fig. 4(b) shows that much stronger responses of 35 dB in transmission and 95% in reflection is available in similar 0.2-nm bandwidth from a BGW written with 1.0-ps pulse duration and 3-μJ pulse energy. Considering the short 10-mm sample length, this Bragg response is three orders of magnitude stronger than the 10-dB transmission and 40% reflection responses in 50-mm long BGWs written with weaker focusing condition (NA = 0.25) . Radiation mode losses of ~0.3 dB and 5 dB are apparent on the short wavelength side of the Bragg resonance in the 100-fs and 1-ps cases, respectively.
The refractive index modulation ΔnAC of the BGWs can be inferred from 
where λ is the Bragg wavelength, L is the grating length, and R= 1 - T is the reflectance derived from the grating transmittance, T, at the Bragg resonance. As an approximation, η was taken as the modal overlap factor defined by the fraction of modal power inside a 2-μm diameter waveguide core, yielding η = 0.06 for the ~11-μm diameter BGW mode. For the 10-mm long BGW in Fig. 4(b), one estimates a strong ΔnAC of ~4 × 10-3, which represents a large ~40% component of the average refractive index ΔnDC = ~0.01 inferred above. Such index modulation is an order of magnitude larger than typically found in strong fiber Bragg gratings (FBG), and is consistent with formation of optically isolated index voxels during the type II waveguide writing process.
The BGWs were classified by grating strength, propagation loss, and birefringence and mapped according to laser pulse energy and duration in Fig. 5. The symbol size represents the grating strength in transmission, with small, medium, and large corresponding to <10 dB, 10 to 20 dB, and >20 dB response, respectively. Half-filled squares represent birefringent BGWs and solid squares correspond to non-birefingent BGWs with single peaks. Open squares represent BGWs with large propagation loss (>3 dB/cm). For pulse duration less than 500 fs, Fig. 5 shows effective guiding above a 1 to 2 μJ energy threshold, but only producing birefringent and weak gratings (<10 dB). An optimum window for generating strong (>20 dB) and low-loss (~0.5 dB/cm) BGWs is identified for 0.8 to 1.2 ps duration and 3 to 6 μJ pulse energy. Higher pulse energy yielded weaker and double-peaked gratings while longer pulse duration (1.5 ps) required higher pulse energy (5 μJ) for guiding and provided only weak and birefringent BGWs.
3.3 Waveguide birefringence
To assess the waveguide birefringence, the BGWs were excited with distinct polarization modes using the free space end-firing arrangement of Fig. 1(c). Figure 6(a) shows the TE and TM transmission spectra (bottom) recorded for the BGW in Fig. 4(a), together with a spectrum excited by 45o linearly polarized light (top). The double-peaked spectrum for 45° polarization closely matches that of Fig. 4(a), which was excited with unpolarized light. The spectra separated into two distinct Bragg resonances for pure TE and TM mode excitation, yielding stronger resonances of 6.5 dB and 9.2 dB at λTE = 1550.9 and λTM = 1551.1 nm, respectively, compared with the respective ~2 dB and 3.2 dB peaks for the unpolarized light in Fig. 4(a).
The waveguide birefringence in BGWs written with 6-μJ pulse energy was inferred from ΔnB = nTM - nTE = (λTM - λTE) / 2Λ, where Λ = 0.52 μm is the grating period, and plotted in Fig. 6(b) as a function of pulse duration. The birefringence strongly correlates with the propagation loss data in Fig. 2(a), suggesting that laser damage also induces asymmetric waveguide stresses. At ~0.1-nm spectral separation, the TE and TM resonances merged into unresolved lines, setting an upper bound of ΔnB = ~1×10-4, for the measurable birefringence in waveguides formed at lower pulse energy, particularly those in the optimum processing window of 1.0-ps duration and 3-μJ pulse energy. The present results suggest new possibilities for fabricating polarization-dependent components in 3-D optical circuits or for trimming birefringence-free waveguides for telecom systems.
3.4 BGW thermal stability
Low-loss BGWs written with 3-μJ pulses of 100-fs and 1.0-ps duration were characterized in several heating cycles beginning at 250°C for 1, 1, 2, and 4 hours (i.e. total accumulated 1, 2, 4, 8 hours), then for 1 hour at 500°C, and finally for one 1 hour at 750°C. After each bake, the samples were cooled to room temperature, observed under an optical microscope, and characterized for mode profile, Bragg wavelength, birefringence, and transmission strength. Results are summarized in Fig. 7 for both the weak birefringence BGWs formed at 100 fs duration and the strong birefringence-free BGWs formed at 1 ps.
The microscope images in Figs. 7(a) and 7(c) show no change in both the 100-fs and 1-ps waveguides for the 250 and 500°C heating steps, but strong fading is apparent after the 750°C annealing step which exceeds the 666°C strain point for the EAGLE2000 glass. Waveguiding was no longer observable after this 750°C cycle. Inspection of mode profiles in Fig. 7(a) and 7(c) reveal strong degradation after the 500°C heat cycle, with mode profile diameter increasing 80% from 11 μm to 20 μm for the 100-fs BGW sample and from 10 μm to 18 μm for the 1-ps case. These changes were commensurate with weakened Bragg resonances as shown respectively in Figs. 7(b) and 7(d) by respective 0.45-nm and 0.3-nm wavelength shifts and diminished transmission peaks from ~3 dB to ~2 dB and from ~35 dB to ~29 dB, respectively. Waveguide birefringence was also no longer detectable in the weaker 100-fs BGWs. These wavelength shifts represent a reduction in the effective index by ~4 × 10-4 in the 100-fs waveguides and 3 × 10-4 in the 1-ps waveguides, with results overall pointing to moderate structural changes and stress reduction at this high 500°C temperature.
The 100-fs and 1-ps BGWs remained highly stable with no degradation of grating strength following each of the four annealing steps at 250°C. A small 0.015-nm shift of Bragg wavelength occurred only for the first 250°C heat cycle for the 1-ps BGW, representing a small ~1 × 10-5 decrease in refractive index that may easily be a thermo-optic response due to ±1°C room temperature fluctuation. For the 100-fs BGWs, both TE and TM resonances shifted ~0.06 nm during the first hour of 250°C baking, while also remaining stable thereafter for the additional 7 hours of annealing. This larger wavelength shift corresponds to an effective index decrease of ~6 × 10-5 that represents a modest 0.6% of the total ΔnDC change induced by laser writing. The 1-ps BGWs appear most stable, and generate very little overall degradation especially compared with FBGs fabricated with traditional ultraviolet lasers , suggesting good prospects for meeting Telecordia standards for 20-year product lifetime.
The results of Fig. 3(a) and 5 together show that high NA (0.55) focusing of 1-ps laser pulses offers both low-loss and strong-grating BGW formation, while 100-fs is suited to only producing low-loss waveguides due to weak and dual-peaked Bragg resonances (<10 dB). In contrast, BGW writing with low-NA (0.25) focusing was previously found to trade increasing waveguide propagation loss against increasing grating strength  limiting the scope for low-NA applications in BGW fabrication. The present high NA method provides significantly stronger gratings in a single process step than in waveguides and gratings formed separately in two laser exposure steps [12, 13]. Such single-step and maskless fabrication offers facile means for writing strong and low-loss Bragg gratings in novel 3-D geometries that extend Bragg gratings beyond traditional optical fibers or planar lightwave circuits. While present results were focused only on boro-aluminosilicate glass, the BGW writing process is expected to be extensible to a broader range of transparent glasses, and also crystals, where waveguide writing recipes are well established. For example, we have used similar exposure conditions to write low-loss waveguides for visible light at 633nm inside fused silica glass , but the BGW processing window will vary widely with material types and laser systems.
In summary, high-strength Bragg grating waveguides were fabricated in a single process step by scanning a tightly focused picosecond laser pulse beneath the surface of boro-aluminosilicate glass. Spectra analysis showed three orders of magnitude improvement in grating strength compared with our previously reported values. Two processing windows for low-loss waveguides were identified at short (100 fs) and long (1 ps) pulse windows, but only long pulses were suitable for generating strong Bragg resonances and low-birefingence waveguides. The BGWs showed high thermal stability at 250°C, with degradation beginning at 500°C that surpasses the performance of fiber Bragg gratings used in Telecom applications. These results are significant in extending the commercial success of FBGs to planar and 3-D optical circuits where 1-ps laser writing now promises facile integration of Bragg filtering and sensing functions in bulk glasses, and possibly other optical materials.
We thank Rajiv Iyer, Dr. Abbas Hosseini, Mi Li Ng and Stephen Ho for helpful discussions. Technical support from Sergey Reznik is greatly appreciated. Haibin Zhang is supported by Alcan and University of Toronto scholarships. Support from the Canadian Institute for Photonics Innovation and the Natural Sciences and Engineering Research Council of Canada is gratefully acknowledged.
References and Links
2. K. Miura, J. Qiu, H. Inouye, T. Mitsuyu, and K. Hirao, “Photowritten optical waveguides in various glasses with ultrashort pulse laser,” Appl. Phys. Lett. 71,3329 (1997). [CrossRef]
3. K. Itoh, W. Watanabe, S. Nolte, and C. B. Schaffer, “Ultrafast processes for bulk modification of transparent materials,” MRS Bulletin 31,620–625 (2006). [CrossRef]
4. A. M. Streltsov and N. F. Borrelli, “Fabrication and analysis of a directional coupler written in glass by nanojoule femtosecond laser pulses,” Opt. Lett. 26,42–43 (2001). [CrossRef]
5. S. M. Eaton, W. Chen, L. Zhang, H. Zhang, R. Iyer, J. S. Aitchison, and P. R. Herman, “Telecom-Band Directional Coupler written with Femtosecond Fiber Laser,” IEEE Photon. Technol. Lett. 18,2174–2176 (2006). [CrossRef]
6. R. Osellame, S. Taccheo, M. Marangoni, R. Ramponi, P. Laporta, D. Polli, S. De Silvestri, and G. Cerullo, “Femtosecond writing of active optical waveguides with astigmatically shaped beams,” J. Opt. Soc. Am. B 20,1559–1567 (2003). [CrossRef]
8. K. O. Hill, B. Malo, F. Bilodeau, D. C. Johnson, and J. Albert, “Bragg gratings fabricated in monomode photosensitive optical fiber by UV exposure through a phase mask,” Appl. Phys. Lett. 62,1035–1037 (1993). [CrossRef]
9. S. J. Mihailov, C. W. Smelser, P. Lu, R. B. Walker, D. Grobnic, H. Ding, G. Henderson, and J. Unruh, “Fiber bragg gratings made with a phase mask and 800-nm femtosecond radiation,“ Opt. Lett. 28,995–997 (2003). [CrossRef] [PubMed]
10. E. Wikszak, J. Burghoff, M. Will, S. Nolte, A. Tunnermann, and T. Gabler, “Recording of fiber Bragg gratings with femtosecond pulses using a “point by point” technique,” in Conference on Lasers and Electro-Optics (CLEO) (IEEE, San Francisco, CA, USA, 2004), p. 2 pp. Vol.2.
11. A. Martinez, M. Dubov, I. Khrushchev, and I. Bennion, “Direct writing of fibre Bragg gratings by femtosecond laser,” Electron. Lett. 40,1170–1172 (2004). [CrossRef]
12. M. Kamata and M. Obara, “Waveguide-based Bragg filters inside bulk glasses integrated by femtosecond laser processing,” in Conference on Lasers and Electro-Optics Europe (CLEO Europe)(IEEE, Munich, Germany, 2005), p. 492.
17. W. Liu, S. L. Chin, O. Kosareva, I. S. Golubtsov, and V. P. Kandidov, “Multiple refocusing of a femtosecond laser pulse in a dispersive liquid (methanol),” Opt. Commun. 225,193–209 (2003). [CrossRef]
18. L. McCaughan and E. J. Murphy, “Influence of temperature and initial Titanium dimensions on fiber-Ti:LiNbO3 waveguide insertion loss at lambda equals 1. 3 mm,” J. Quanum Electron., IEEE QE-19,131–136 (1983). [CrossRef]
19. S. Nolte, M. Will, J. Burghoff, and A. Tuennermann, “Femtosecond waveguide writing: a new avenue to three-dimensional integrated optics,” Appl. Phys. A: Mater 77,109–111 (2003). [CrossRef]
20. R. Osellame, N. Chiodo, V. Maselli, A. Yin, M. Zavelani-Rossi, G. Cerullo, P. Laporta, L. Aiello, S. De Nicola, P. Ferraro, A. Finizio, and G. Pierattini, “Optical properties of waveguides written by a 26 MHz stretched cavity Ti : sapphire femtosecond oscillator,” Opt. Express 13,612–620 (2005). [CrossRef] [PubMed]
21. R. S. Taylor, C. Hnatovsky, E. Simova, D. M. Rayner, M. Mehandale, V. R. Bhardwaj, and P. B. Corkum, “Ultra-high resolution index of refraction profiles of femtosecond laser modified silica structures,” Opt. Express 11,775–781 (2003). [CrossRef] [PubMed]
22. C. B. Schaffer, J. F. Garcia, and E. Mazur, “Bulk heating of transparent materials using a high-repetition-rate femtosecond laser,” Appl. Phys. A-Mater 76,351–354 (2003). [CrossRef]
23. A. H. Nejadmalayeri and P. R. Herman, “Ultrafast laser waveguide writing: Lithium Niobate and the role of circular polarization and picosecond pulse width,” Opt. Lett. 31,2987–2989 (2006). [CrossRef] [PubMed]
24. T. Tamaki, W. Watanabe, H. Nagai, M. Yoshida, J. Nishii, and K. Itoh, “Structural modification in fused silica by a femtosecond fiber laser at 1558 nm,” Opt. Express 14,6971–6980 (2006). [CrossRef] [PubMed]
25. K. O. Hill and G. Meltz, “Fiber Bragg grating technology fundamentals and overview,” J. Lightwave Technol. 15,1263–1276 (1997). [CrossRef]
26. T. Erdogan, V. Mizrahi, P. J. Lemaire, and D. Monroe, “Decay of ultraviolet-induced fiber Bragg gratings,” J. Appl. Phys. 76,73–80 (1994). [CrossRef]