We report on fs-laser micromachining of active waveguides in a new erbium-doped phospho-tellurite glass by means of a compact cavity-dumped Yb-based writing system. The spectroscopic properties of the glass were investigated, and the fs-laser written waveguides were characterized in terms of passive as well as active performance. In particular, internal gain was demonstrated in the whole C+L band of optical communications (1530–1610 nm).
©2008 Optical Society of America
Since its first demonstration by Hirao’s group , fs-laser micromachining of optical waveguides has been extensively developed and several devices, both passive and active, have been so far demonstrated in glasses, crystals and polymers (see  and references therein). Actually, the fast prototyping of devices makes this fabrication method particularly suitable for research laboratories, allowing a wide experimentation of the writing parameters (i.e. pulse energy, repetition rate, writing speed, single- and multi-scan writing) on the same substrate. In particular, glass substrates have attracted much interest for their high mechanical and thermal stability and the potential of superior compatibility with silica-based fiber-optic circuits. For the past years femtosecond inscription of waveguide channels on various bulk glass substrates like silica and multi-component silicate glasses, borates, phosphates, fluorozirconates (ZBLAN), chalcogenides has been reported [1–9].
Recently, tellurite glasses doped with erbium ions have attracted much interest in view of a broad erbium emission bandwidth. Since the first demonstration of a broadband Er-doped tellurite fiber amplifier , a strong effort has been done to fabricate channel waveguide amplifiers for broad-band amplification in the C+L telecom bands. Unfortunately, the surface damage due to photolithographic processing disables the fabrication of ion-exchanged waveguide channels in tellurite glasses . An attempt of waveguide fabrication by micromachining with a continuous-wave doubled frequency Ar-ion laser operating at 244-nm was reported in a modified tellurite glass . A local increase in the refractive index up to a maximum value of 1.5 × 10-3 was achieved; anyway, the 10-mm long waveguides exhibited insertion loss as high as 8 dB. Femtosecond laser writing technique was also experimented in pure tellurite glass and resulted unsuccessful, because ultrafast laser irradiation could induce only a negative refractive index change in this material, making waveguiding impossible [11, 13]. Meanwhile it has been reported that modifying the tellurite glasses with phosphate or niobium allows positive refractive index change by micromachining with femtosecond infrared pulses and channel waveguides have been demonstrated [14, 15]. But till date no gain was reported for tellurite or modified tellurite channel waveguides.
In this paper we report on the demonstration of fs-laser micromachining of active channel waveguides in a new erbium-doped phospho-tellurite glass by means of a compact Yb-based writing system. A spectroscopic investigation of the glass base is reported as well as a complete characterization of the fs-laser written active waveguides. In particular, a gain measurement was carried out in the C (1530–1565) and L (1565–1610 nm) telecom bands.
2. Fabrication and characterization of the glass sample
The glass sample for the femtosecond laser writing experiment was prepared in a two-step process. In the first step, a glass containing TeO2, ZnO and Na2O in the molar ratio 80:10:10 and doped with 0.5 wt% Er2O3 was prepared. A 10g precursor batch was melted at 850°C for 3 hours under flowing dry oxygen atmosphere and it was quenched at 280°C casting in a preheated mould. It was annealed for 3 hours at that temperature and then slowly cooled down to room temperature. The glass thus formed was crushed and mixed with 20 mol% of P2O5 and heated in a furnace kept at 300°C for 3 hours. It was then quickly transferred to another furnace with temperature kept at 900°C. The mixture was melted for 3 hours before casting the glass at 300°C and then annealed at this temperature for 3 hours. This procedure was followed in order to decrease the sublimation of P2O5 while melting it.
A glass density as high as 4.36 g/cm3 was measured and erbium ion concentration was calculated from the glass density and the initial batch compositions and it was found to be 7.0 × 1019 ion/cm3.
After fabrication the glass was polished to optical quality for spectroscopic measurements and waveguide writing.
The refractive index was measured using a prism coupler (Metricon 2010) at 633 nm and resulted to be about 1.95, thus significantly higher than in phosphate glasses.
The absorption spectrum [Fig. 1(a)] was recorded in the range 300–1800 nm using a dual beam spectrophotometer (mod. JASCO V570). A blue shift was observed for the UV absorption edge in our new composition of phosphate modified tellurite glass compared to the glass reported in . Erbium absorption peaks are labeled with respect to transitions from the ground level (4I15/2) to various upper levels of erbium. From the measured absorption and the calculated erbium concentration the absorption cross-section was computed [see Fig. 1(b)], resulting in a peak value of 6.99 × 10-21 cm2 at 1533.5 nm.
The emission cross section [see Fig. 1(b)] was calculated according to McCumber approach  starting from the absorption cross section and metastable state lifetime. Note that the emission cross section exhibits a linewidth as large as 46.3 nm (against 41 nm value reported for a similar glass ) and a relatively high shoulder in the long wavelength range (1560–1620 nm), which are features belonging to the tellurite nature of the glass composition (see as example  and references therein).
3. Waveguide writing
Waveguides have been fabricated in the erbium-doped phospho-tellurite glass by a diode-pumped cavity-dumped Yb:KYW femtosecond laser oscillator, operated at 600 kHz [18, 19]. An average power of 800 mW was measured in the dumped beam, yielding a pulse energy of 1.3 µJ. The pulse width is about 350 fs at a wavelength of 1040 nm.
Previous results on waveguide writing in a similar phospho-tellurite glass employed a 1 kHz Ti:sapphire amplified laser . The laser system used in this work, however, presents the following main advantages: 1) the lack of amplification stages, making it simpler and more cost effective; 2) the use of diode pumping, making the system more compact, efficient, and reliable; 3) the range of pulse energies (hundreds of nanojoules) and repetition rates (up to 1 MHz), more suited to waveguide writing. Note that this laser operates in an intermediate repetition rate range between the low- and high-frequency regimes [20, 21].
The setup used in the experiments is the same as that reported in . The writing geometry is transversal, since it allows for writing waveguides with arbitrary length and minimizes the detrimental aberration effects. The laser beam is focused into the glass sample by means of a microscope objective at a depth from the sample surface of 150 µm. Two objectives are alternatively used: (i) 50× long working distance (N.A. 0.6), and (ii) 100× oil immersion (N.A. 1.4). The sample is mounted on a motorized translation stage (mod. Physik Instrumente, M-511.DD).
Several waveguides have been fabricated using the two microscope objectives at different writing speeds, ranging from 0.1 mm/s to 1 mm/s, and pulse energies, from 50 nJ to 1 µJ. With the 100× objective we could only produce damages at any combination of speed and pulse energy. With the 50× objective we found a parameter range, i.e. translation speed of 0.1 mm/s and pulse energies ranging from 330 nJ to 620 nJ, where good waveguides could be created. Differential Interference Contrast (DIC) microscope images of such waveguides are shown in Fig. 2(a).
The waveguide cross-sections are asymmetric due to the transverse writing geometry. A solution for the waveguide cross-section asymmetry has already been demonstrated , but cannot be applied in this case since the maximum laser pulse energy is too low and the use of an astigmatic beam would decrease the intensity in the focal region below the modification threshold. The waveguides clearly show a double structure, a brighter core with a droplet-like surrounding volume. This kind of structure is rather typical for this irradiation conditions [21, 23] and can be interpreted by considering the inner core as the directly irradiated region and the outer structure as the region modified by thermal diffusion of the energy stored in the inner core.
Figure 2(b) shows the comparison between two waveguide cross-sections written at 420 nJ, one at 0.1 mm/s and the other at 1 mm/s. It is worth noting that by changing the pulse energy at a constant translation speed the overall modified volume increases in size [Fig. 2(a)], while increasing the translation speed with the same pulse energy the outer structure maintains the same size and the inner core tends to vanish [Fig. 2(b)].
This effect is related to the repetition rate of this laser, 600 kHz, which induces an intermediate situation between single pulse and cumulative regimes . In fact, the outer structure can be attributed to thermal diffusion of the single pulse and is thus affected by the pulse energy but not by the scan speed, while the inner core cumulates the modifications from subsequent pulses hitting the same volume and therefore is smaller for the higher writing speed. As usual in this kind of waveguides, the main guiding region is the bright inner core ; for this reason good confinement is expected only in structures written at 0.1 mm/s.
4. Waveguide passive and active characterization
4.1 Passive characterization
A preliminary investigation of the guiding properties was carried out in the visible by focusing the 633 nm radiation of an He-Ne laser on the input facet of the fs-laser written waveguides. As expected, only the waveguides fabricated at 0.1 mm/s writing speed and using the 50×microscope objective exhibited good confinement. Among them, the waveguide fabricated at 420 nJ pulse energy resulted to be the most promising and was selected for a complete passive and active characterization at telecom wavelength.
By means of a refractive near-field profilometer (Rinck Elektronik) we estimated a maximum refractive index change Δn of about 1 × 10-3 at 633 nm, but an accurate measurement of the refractive index profile was prevented by the lack of an index matching fluid with the proper refractive index value (i.e. of about 1.95, namely, close to the bulk refractive index of the phospho-tellurite glass).
The waveguide was butt-coupled to a fiber-pigtailed infra-red laser diode at 1600 nm and the near-field image profile at the output facet was recorded with a microscope objective and a vidicon camera (mod. Hamamatsu C2400). An almost circularly symmetric transverse mode was obtained [see Fig. 3(a)], resulting to be highly compatible with standard telecom fiber (which is an essential requirement to minimize the device insertion loss in telecom optical circuits), as ascertained by comparison to the near-field mode of a standard telecom fiber [see Fig. 3(b)]. Note that a side lobe, carrying almost negligible power, is nevertheless present (see the near-field profile along y-axis), thus indicating a slight deviation from single-transverse mode behavior, as expected in view of the asymmetry exhibited by the cross-section of the waveguides (see Fig. 2). From the overlap integral between waveguide and fiber near-field modes and also taking into account Fresnel losses we estimated coupling losses to standard single-mode fiber as low as 0.5 dB/facet.
4.2 Active measurements
The active waveguide was then characterized in terms of active performance by means of a standard set-up for waveguide amplifier testing (see Fig. 4). The waveguide was butt-coupled to single mode optical fibers (Corning SMF-28) by means of precision 5-axes micro-positioning stages. A fiber-pigtailed AlGaAs laser diode provided pump radiation at 980 nm (220 mW maximum pump power), and a co-propagating pumping scheme was adopted. The probe signal was provided by a broad-band source (mod. IPG Fibertech ASE-100-C + ASE- 100-L) or by means of a tunable laser diode (mod. Agilent 8164B). The broad-band source is an ASE erbium doped fiber source, delivering a broad-band signal in the C+L band (1530–1565 nm+1565–1610 nm). The tunable laser is a low noise narrow bandwidth source operating in the wavelength range 1460–1640 nm. The probe signal was properly attenuated to -26 dB/nm input power level (monitored by a high precision power meter, mod. Ando AQ2140) by means of a variable attenuator. Pump and signal radiations were coupled at the input of the waveguide by means of a wavelength division multiplexer (WDM). An index matching fluid able to support high power density at the pump wavelength was also inserted between waveguide and fiber ends.
By exploiting this set-up we measured absorption spectrum, insertion loss and internal gain of the waveguide. At first, the total (passive and active) insertion loss spectrum of the waveguide was recorded by comparing the output power through the test bed with and without the presence of the waveguide. Note that in this phosphate-modified tellurite glass erbium absorption cross section at 1640 nm is almost negligible [see Fig. 1(b)], therefore total insertion loss at 1640 nm accounts only for the passive insertion loss (IL), namely scattering loss plus coupling loss to single mode fiber. By using the tunable laser diode a passive insertion loss of about 4.4 dB was measured at 1640 nm. Propagation loss (PL) can thus be calculated as the difference between passive insertion loss and coupling loss, resulting in <1.35 dB/cm, a slight improvement with respect to previous result of 2 dB/cm reported in . Absorption spectrum (in log scale) was then computed as the difference between total insertion loss spectrum and passive insertion loss. A maximum absorption of 5.2 dB was measured at 1533.5 nm [see Fig. 5(a)], corresponding to an absorption per unit length of ~2.08 dB/cm, in good agreement with the theoretical value of 2.13 dB/cm expected from bulk spectroscopic measurement [see section 2].
For gain measurement we adopted the ON-OFF technique : at first, the enhancement spectrum of the waveguide, namely the ratio between signal power level at the output with pump on and signal power level at the output with pump off, was recorded as a function of signal wavelength; then the internal gain was computed as the difference (in log scale) between the enhancement and the absorption spectrum previously recorded. The measurement was carried out by means of the broad-band ASE source (blue solid line) and confirmed by some spot measurements provided by means of the tunable laser diode (blue circles) [see Fig. 5(a)]. A maximum internal gain of 1.25 dB at 1555 nm was demonstrated, as well as internal gain in the whole C+L band.
The small-signal internal gain as a function of incident pump power at 1534 nm, 1555 nm and 1590 nm signal wavelengths is shown in Fig. 5(b). The pump power at transparency is 130 mW at 1534 nm, 30 mW at 1555 nm and 13 mW at 1590 nm. The very limited internal gain achieved at 1534 nm indicates a low average inversion along the waveguide. Actually at this wavelength the emission cross section of erbium is comparable to the absorption cross section [see Fig. 1(b)], thus a saturated internal gain of about 5 dB (i.e. comparable to the peak absorption value) would be expected for a complete population inversion. An improved performance under 980 nm pumping wavelength is expected by codoping with ytterbium ions. Furthermore, a higher refractive index change should reduce propagation losses, because micro-bending losses (induced by mechanical vibrations during the writing process) typically scale with Δn-3  and this is expected to result in a higher pumping efficiency and, most importantly, in the demonstration of a real amplifier, exhibiting net gain. To address this point, several phospho-tellurite glasses of different composition should be experimented and operated by the Yb-based fs-oscillator under an extended range of writing parameters.
Femtosecond laser written active waveguides were fabricated in an erbium-doped phospho-tellurite glass by means of a compact cavity-dumped Yb-based writing system. The broadband emission provided by tellurite base composition and the high phonon energies induced by phosphate modification resulted in a broad-band internal gain of the fs-laser written waveguides, spanning the whole C+L band of optical communications (1530–1610 nm). This preliminary result demonstrates the potential of fs-laser written phospho-tellurite waveguides for the development of ultra-broad-band optical amplifiers.
Authors T. T. Fernandez and P. Laporta would like to acknowledge the fellowship from the Italian Ministry of University and Research (Prot. n.179, 29.01.2007) in the framework of the research co-operation between Politecnico di Milano and Mahatma Gandhi University, Kottayam (India). The authors would also like to thank N. V. Unnikrishnan, School of Pure & Applied Physics, Mahatama Gandhi University. The authors A. Jha and G. Jose acknowledge the EPSRC-UK Basic Technology Grant (EP/D048672/1).
References and links
2. R. R. Gattass and E. Mazur, “Femtosecond laser micromachining in transparent materials,” Nat. Photonics 2, 219–225 (2008). [CrossRef]
3. 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, 3329–3331 (1997). [CrossRef]
4. A. Streltsov and N. F. Borrelli, “Study of femtosecond-laser-written waveguides in glasses,” J. Opt. Soc. Am. B 19, 2496–2504 (2002). [CrossRef]
5. K. Minoshima, A. M. Kowalevicz, I. Hartl, E. P. Ippen, and J. G. Fujimoto, “Photonic device fabrication in glass by use of nonlinear materials processing with a femtosecond laser oscillator,” Opt. Lett. 26 , 1516–1518 (2001). [CrossRef]
6. V. R. Bhardwaj, E. Simova, P. B. Corkum, D. M. Rayner, C. Hnatovsky, R. S. Taylor, B. Schreder, M. Kluge, and J. Zimmer, “Femtosecond laser-induced refractive index modification in multicomponent glasses,” J. Appl. Phys. 97, 083102-1-8 (2005). [CrossRef]
7. J. W. Chan, T. R. Huser, S. H. Risbud, J. S. Hayden, and D. M. Krol, “Waveguide fabrication in phosphate glasses using femtosecond laser pulses,” Appl. Phys. Lett. 82, 2371–2373 (2003). [CrossRef]
8. K. Miura, Jianrong Qiu, T. Mitsuyu, and K. Hirao, “Preparation and optical properties of fluoride glass waveguides induced by laser pulses,” J. Non-Cryst. Solids 256&257, 212–219 (1999). [CrossRef]
9. M. Hughes, W. Yang, and D. Hewak, Appl. Phys. Lett. 90, 131113-1-3 (2007).
10. A. Mori, T. Sakamoto, K. Kobayashi, K. Shikano, K. Oikawa, K. Hoshino, T. Kanamori, Y. Ohishi, and M. Shimizu, “1.58-µm Broad-Band Erbium-Doped Tellurite Fiber Amplifier,” J. Lightwave Technol. 20, 794–799 (2002). [CrossRef]
11. G. C. Righini, I. Banyasz, S. Barneschi, M. Brenci, M. Cremona, D. Ehrt, M. Ferrari, R. M. Montereali, G. N. Conti, S. Pelli, S. Sebastiani, and C. Tosello, “Laser irradiation, ion implantation and e-beam writing of integrated optical structures,” in Photonic Materials, Devices, and Applications, G. Badenes, D. Abbott, and A. Serpenguzel, eds., Proc. SPIE 5840, 649–657 (2005).
12. E. Chierici, M. C. Didavide, A. Moro, O. Rossotto, L. Tallone, and E. Monchiero, “Direct writing of channel waveguide on a tellurite glass using a focused ultraviolet laser beam,” Proc. IEEE/LEOS Workshop on Fibre and Passive Components , 24–28 (2002). [CrossRef]
13. S. K. Sundaram, C. B. Schaffer, and E. Mazur, “Microexplosions in tellurite glasses,” Appl. Phys. A 76, 379–384 (2003). [CrossRef]
14. Y. Tokuda, M. Saito, M. Takahashi, K. Yamada, W. Watanabe, K. Itoh, and T. Yoko, “Waveguide formation in niobium tellurite glases by pico- and femtosecond laser pulses,” J. Non-Cryst. Solids 326, 472–475 (2003). [CrossRef]
15. P. Nandi, G. Jose, C. Jayakrishnan, S. Debbarma, K. Chalapathi, K. Alti, A. K. Dharmadhikari, J. A. Dharmadhikari, and D. Mathur, “Femtosecond laser written channel waveguides in tellurite glass,” Opt. Express 14, 12145–12150 (2006). [CrossRef] [PubMed]
17. P. C. Becker, N. A. Olsson, and J. R. Simpson, Erbium-Doped Fiber Amplifiers (Academic Press,1998).
18. A. Killi, A. Steinmann, J. Dörring, U. Morgner, M. J. Lederer, D. Kopf, and C. Fallnich, “High peak power pulses from a cavity-dumped Yb:KY(WO4)2 oscillator,” Opt. Lett. 30, 1891–1893 (2005). [CrossRef] [PubMed]
19. R. Osellame, N. Chiodo, G. Della Valle, G. Cerullo, R. Ramponi, P. Laporta, A. Killi, U. Morgner, and O. Svelto, “Waveguide lasers in the C-band fabricated by laser inscription with a compact femtosecond oscillator,” IEEE J. Sel. Top. Quantum Electron. 12, 277–285 (2006). [CrossRef]
20. R. Osellame, N. Chiodo, G. Della Valle, S. Taccheo, R. Ramponi, G. Cerullo, A. Killi, U. Morgner, M. Lederer, and D. Kopf, “Optical waveguide writing with a diode-pumped femtosecond oscillator,” Opt. Lett. 29, 1900–1902 (2004). [CrossRef] [PubMed]
21. S. Eaton, H. Zhang, P. Herman, F. Yoshino, L. Shah, J. Bovatsek, and A. Arai, “Heat accumulation effects in femtosecond laser-written waveguides with variable repetition rate,” Opt. Express 13, 4708–4716 (2005). [CrossRef] [PubMed]
22. 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]
23. N. D. Psaila, R. R. Thomson, H. T. Bookey, S. Shen, N. Chiodo, R. Osellame, G. Cerullo, A. Jha, and A. K. Kar, “Supercontinuum generation in an ultrafast laser inscribed chalcogenide glass waveguide,” Opt. Express 15, 15776–15781 (2007). [CrossRef] [PubMed]
24. J.-I. Sakai and T. Kimura, “Practical Microbending Loss Formula for Single-Mode Optical Fibers,” J. Quantum Electron. 15, 497–500 (1979). [CrossRef]