Three-Dimentional Laser Writing of Mid-Infrared Waveguides Circuits in Lithium Niobate Crystal

Zusammenfassung

Three-dimensional ultrafast laser writing (3DLW) of mid-infrared (mid-IR) depressed-index cladding waveguides (CLWs) in crystals has received considerable attention in recent years, thanks to the ease with which mid-IR waveguiding can be achieved, and to the potential of the 3DLW technique for fabricating waveguide circuits which are free from the limitations of planar designs [1-4]. Among the commonly used crystals, Lithium Niobate (LiNbO3) has shown big advantages over the others due to its extensive characteristics including excellent electro-optic, piezo-electric, piezo-optic and non-linear optical properties, together with a broad transparency from the visible to mid-IR range. Implementation of mid-IR waveguides inside a volume of LiNbO3 crystal thereby has a wide range of uses: novel integrated nonlinear frequency conversion chips, electro-optic interferometric spectrometers, as well as applications in chemical and bio-medical fields, atmospheric research, high-resolution on-chip vibrational spectroscopy, and astrophotonic instrumentation [5-8]. Challenges of the field The tight focusing of ultrafast near-IR laser pulses inside LiNbO3 is known to induce a non-linear photo-ionisation process which ultimately results in a spatially localised lattice distortion at micron and nano-scale [9-11]. The irreversible changes involve lattice defects and amorphisation, typically leading to a modification in the refractive indices (RI) of the material. The modification can be judiciously exploited to construct 3D waveguide architectures. In general, two fundamental types of RI modification have been reported with the 3DLW technique in LiNbO3: (1) the so-called Type I modification which results in a RI increase within the laser focal volume; and (2) the Type II modification which involves negative-RI change occurred at the laser focal volume. In addition, strong stress-optic fields are always present at surrounds of the localised laser focal volume due to the lattice amorphisation and defect generation, involving collateral piezo-optic index changes [5, 12]. The stress-optic fields can expand several microns away from the laser-written structures and must be taken into account when computing RI profiles of optical waveguides. Depressed-index CLWs are typically fabricated by transversally writing tubular structures in the regime for producing Type II index changes, which therefore act as depressed-RI claddings for sustaining light propagating modes (leaky modes) within its un-modified core. Since the waveguiding core volume is un-modified/un-irradiated by the laser pulses, it preserves all the important optical properties of the pristine crystalline material. This is a clear advantage of this approach over the ion implantation of 3DLW in the Type I regime in which the laser-modified volume is used as a waveguiding core. The light guiding properties of CLWs depend on the magnitude of the negative RI change (-Δn) produced at the laser-written tracks, and on the size and spatial width of the cladding arrangement [13]. Full RI profiles of a cladding structure are the combination of (i) the local RI changes Δnlocal at the laser-modified tracks and (ii) the anisotropic piezo-optic RI fields Δnstress induced at their surrounds, which strongly depend on particular 3DLW parameters: pulse duration, pulse energy, repetition rate, scan speed, wavelength, polarisation, and focusing optics, as well as crystallographic orientation of the anisotropic LiNbO3 crystal. As a result of the high complexity of the laser-written structures which originates from the anisotropy of the material and the stress-induced effects, there exists a fundamental bottleneck for obtaining CLWs in LiNbO3 capable of single-mode low-loss guiding for both orthogonal polarisations at the mid-IR wavelengths. To the best of our knowledge, all LiNbO3 CLWs reported for operation at wavelengths longer than 3000 nm have been limited to guiding for mostly one polarisation, and propagation losses (PLs) of ~3 dB/cm [2, 3, 5, 10, 11, 14-21]. These high PLs are far from the value for practical applications, that is <0.5 dB/cm as required for high performance of mid-IR integrated photonic devices at cm’s scale. In pursuit of improving the optical performance of CLWs, thermal treatments have been additionally performed to reduce the anisotropic micro-stress fields induced during the laser writing process [5, 22, 23]. However, an underlying mechanism of the changes in mid-IR waveguide profiles under different thermal annealing conditions has not been fully understood, mainly due to the difficulty in measuring RI changes of sub-micron volume at this long wavelength range. To date, most experimental reports on CLWs were based on fabrication trial and error methodologies [1, 2, 4, 24-29], with no information on the RI profiles and therefore no proper understanding of the waveguide properties. Heuristic model for obtaining realistic information of the laser-written CLWs in LiNbO3. In this thesis, a comprehensive simulation model is built, taking into account both the depressed-RI changes and the anisotropic stress-induced RI fields. This efficient evaluation is important for developing high performance mid-IR cladding waveguides, and reliable laser manufacturing of photonic circuits. Full mid-IR refractive index profiles of LiNbO3 CLWs microstructured under a given 3DLW condition are reported for the first time to our knowledge. We developed a novel heuristic modelling approach based on the use of standard optical characterisation data (near-field mode diameters -MFDs and propagation losses -PLs), along with standard numerical methods (finite element method –FEM). The approach offers a satisfactory solution to the problem of designing realistic laser-written circuit building-blocks, such as straight waveguides, bends and evanescent splitters. Achievement of high performance LiNbO3 CLWs by low-repetition rate regime. In this section, the CLWs were fabricated by 3DLW technique in the low-repetition rate regime (1 kHz) in which full heat diffusion occurs between pulses and therefore the residual laser-induced stress as well as the population of lattice defects within the laser-written structures are significantly high. As evaluated in the simulation model, these high stress-induced fields anisotropically favour guiding for one polarised mode, while deteriorate guiding for the other. Based on the extracted values of the complex RI profiles, CLW designs were arranged towards single-guided mode low-loss performance for both transverse electric (TE) and transverse magnetic (TM) polarised lights. Cladding arrangements with core diameters of 40 - 100 µm, and cladding thicknesses of 10 - 20 µm were studied for given laser writing parameters, resulting in optical guiding of 3680 nm mid-IR light with lowest PLs of <0.5 dB/cm for the first time to our knowledge. Evidence of the inherent anisotropic behaviour is clearly observed, as the CLWs support single-mode guiding for TM polarised light, whereas they are multimode for TE light. In order to diminish the anisotropy of the RI profiles, the CLWs were further investigated with different thermal treatment conditions. We compared guiding properties of the CLWs both in as-fabricated samples and annealed samples. More importantly, we inferred how the sub-micron laser written tracks change in size and index of refraction, and also how the stress-fields reduce as a function of annealing temperature and time. Results reveal that by performing a sequential thermal annealing until a peak temperature of 773 K, single mode guiding at 3680 nm wavelength for both polarisations could be achieved, with a minimum obtained PL of 1.25 dB/cm and 1.79 dB/cm for TM and TE polarisations, respectively. Achievement of high performance LiNbO3 CLW devices in the critical repetition rate regime.In this set of studies, the fabrication of the CLWs was processed by 3DLW in the critical repetition rate regime (100 kHz). In this regime, the arrival time between pulses is equivalent to the time for the heat diffusion out of the focal volume, so that the heat begins to moderately accumulate, potentially erasing lattice defects and diminishing the generation of surrounding stress fields. Additionally, the use of a 100-fold higher repetition rate allows to fabricate waveguides at a 100-fold faster speed, thereby dramatically reducing the processing time. This is practically important for microstructuring complex waveguide circuits which involves thousands of written elements. The dependence of the CLW properties with respect to various 3DLW parameters, i.e. pulse duration and pulse energy was additionally investigated. In this regime, performance of CLWs with single-mode low-PLs of ~1 dB/cm for both TE and TM polarisation was directly obtained without the need of thermal annealing post-process, as it was needed for the low-repetition rate 3DLW waveguides. Based on the high-performance straight-CLWs, s-bend CLW structures were further developed under various laser parameters, also showing losses of as low as ~0.5 dB/cm configured with bending-radius of up to 200 mm. Following the success of the straight and s-bend CLWs, finite difference beam propagation method (FD-BPM) was used to numerically investigate the practical development of directional beam splitters. Configuration of the splitters was based on the concept of evanescent coupling waveguides combined with s-bend structures. The achieved directional beam splitters are the building blocks of a Mach-Zehnder (MZ) structure which was further numerically examined. This MZ structure in LiNbO3 demonstrates promising applications for light modulators and interferometers at the mid-IR range, and potentially for broader fields ranging from mid-IR sensing to astrophotonic instrumentation. Development of longitudinal laser writing scenario – a new approach for more symmetry of the laser-written LiNbO3 CLWs. It is pertinent to note that all the above CLWs were fabricated with the conventional transversal writing geometry in which the laser focus is kept at a constant depth (200 - 500 µm) inside the sample and transversally scanned with respect to its focusing/propagation direction. This approach allows laser-writing in the substrate with 3D arbitrary designs, which credits the capability of laser-modifications at different depths. However, the approach faces a critical limitation: since the laser focus has a typical Gaussian shape, the vertical cross-section of the laser-written volume has an elliptical shape which gives a strong source of anisotropic stress induced during the lattice amorphisation. This strong anisotropy is further added by a vertically elongated shape (~10 µm) of the laser-written tracks which are usually observed in this writing geometry. In order to overcome the anisotropic issue, we explored another approach which offers more symmetric profile of the laser-modified volume. In this approach, the sample is laser-inscribed along the longitudinal/parallel direction to the laser propagation. The 3DLW results in laser-written structures with circular cross-section which can minimise the anisotropic effect of the laser-induced stress. However, this requires a deep focusing of the laser beam inside the crystal (>2 mm), and thereby complicates the lattice modification process, due to strong refraction effects, aberrations, and non-linear processes. We found that the laser modification in this longitudinal writing scheme is extremely sensitive to the alignment between the laser beam and the sample, the surface quality of the crystal, the laser polarisation, and more importantly the pulse temporal duration and energy. Recognisable type II modification was achieved at depths of up to 5 mm inside the crystal once the fabrication conditions were finely adjusted. The LiNbO3 CLWs were demonstrated for the first time by this longitudinal writing geometry. Single-guided mode mid-IR performance at 3680 nm wavelength with lowest PLs of 0.5 dB/cm was achieved. The cladding structures were designed with several two-dimensional track arrangements: concentric circular rings, lattice-like hexagonal rings and with helical continuous tracks. Among these cladding geometries, the helical structure shows big advantages over the conventional structures, due to its fast processing time, three-fold faster than it is required for the others. The best waveguiding performance was however obtained by the hexagonal track arrangements. Future works High optical performance with PLs <0.5dB/cm was successfully achieved. However, the mid-IR LiNbO3 CWLs still can be improved by better adjustment of the designs, 3DLW conditions, and appropriate thermal treatments. Specifically, in the transversal laser writing scheme, the straight and bend-CLWs were not fabricated by identical conditions due to a technical problem: the laser operation was unstable at the time. Once the laser equipment is fully serviced, it is believed that the CLWs can be further improved. In the longitudinal writing scheme, the CLWs were still limited by lengths, as the difficulty in the laser modification at depths of longer than 5 mm inside the LiNbO3 crystal. This problem can be solved by using high power laser, and properly adjusting both energy and scan speed on the fly. Multi-scanning process is also worth to try for possible modification of higher refractive index contrast, which therefore improves the guiding performance of waveguides. Fabrication of mid-IR directional beam splitters in LiNbO3 is highly feasible with the fundamental foundation of the highly achieved CLWs in both straight and s-bend geometries. Implementation of the splitters can be done by properly designing the laser writing trajectories in a programed sub-micron precision motion stage. The next step is to fabricate fully functional MZ devices which basically consist of beam splitters and combiners. Electrodes can be embedded on the MZ platform for light modulators, and interferometers which are the fundamental base for applications in mid-IR sensing and astrophotonics. Other potential work is to fabricate the low-loss CLWs in periodically poled Lithium Niobate (PPLN) for on-chip mid-IR light generation/conversion. Reference 1. Y. Ren, G. Brown, A. Ródenas, S. Beecher, F. Chen, and A. K. Kar, Mid-infrared waveguide lasers in rare-earth-doped YAG. Opt. Lett., 2012. 37(16): p. 3339-3341. 2. R. He, Q. An, Y. Jia, G. R. Castillo-Vega, J. R. Vázquez de Aldana, and F. Chen, Femtosecond laser micromachining of lithium niobate depressed cladding waveguides. Optical Materials Express, 2013. 3(9): p. 1378-1384. 3. Q. An, Y. Ren, Y. Jia, J. Rodríguez Vázquez de Aldana, and F. Chen, Mid-infrared waveguides in zinc sulfide crystal. Opt. Mat. Express, 2013. 3: p. 466-471. 4. J. R. Macdonald, S. J. Beecher, P. A. Berry, G. Brown, K. L. Schepler, and A. K. Kar, Efficient mid-infrared Cr:ZnSe channel waveguide laser operating at 2486 nm. Opt. Lett., 2013. 38(13): p. 2194-2196. 5. R. He, Q. An, Y. Jia, G. R. Castillo-Vega, J. R. Vázquez de Aldana, and F. Chen, Femtosecond laser micromachining of lithium niobate depressed cladding waveguides. Opt. Mater. Express, 2013. 3: p. 1378-1384. 6. S. Heidmann, N. Courjal, and G. Martin, Double polarization active Y-junctions in the L-Band, based on Ti:LiNbO3 Waveguides: polarization and contrast performances. Opt. Lett., 2012. 37(16): p. 3318-3320. 7. G.Martin, S. Heidmann, J.-Y. Rauch, L. Jocou, and N. Courjal, Electro-optic fringe locking and photometric tuning using a two-stage Mach-Zehnder lithium niobate waveguide for high-contrast mid-infrared interferometry. Opt. Eng., 2014. 53(3): p. 034101. 8. F. Thomas, S. Heidmann, M. de Mengin, N. Courjal, G. Ulliac, A. Morand, P. Benech, E. Le Coarer, and G. Martin, First Results in Near and Mid IR Lithium Niobate-Based Integrated Optics Interferometer Based on SWIFTS-Lippmann Concept. J. Lightwave Technol., 2014. 32(22): p. 3736–3742. 9. A. Rodenas, G. Zhou, D. Jaque, and M. Gu, Direct laser writing of three-dimensional photonic structures in Nd:yttrium aluminum garnet laser ceramics. Appl. Phys. Lett., 2008. 93(15): p. 151104. 10. A. Rodenas, G. A. Torchia, G. Lifante, E. Cantelar, J. Lamela, F. Jaque, L. Roso, and D. Jaque, Refractive index change mechanisms in femtosecond laser written ceramic Nd:YAG waveguides: micro-spectroscopy experiments and beam propagation calculations. Appl. Phys. B, 2009. 95(1): p. 85–96. 11. A. Rodenas, Direct femtosecond laser writing of 3D photonic structures in rare-earth doped lithium niobate. 2009, Universidad Autónoma de Madrid. 12. H. Karakuzu, M. Dubov, and S. Boscolo, Control of the properties of micro-structured waveguides in lithium niobate crystal. Opt. Express, 2013. 21: p. 17122-17130. 13. J. Hu and C. R. Menyuk, Understanding leaky modes: slab waveguide revisited. Adv. Opt. Photonics, 2009. 1: p. 58–106. 14. A. Rodenas, G. A. Torchia, G. Lifante, E. Cantelar, J. Lamela, F. Jaque, L. Roso, and D. Jaque, Refractive index change mechanisms in femtosecond laser written ceramic Nd:YAG waveguides: micro-spectroscopy experiments and beam propagation calculations. Applied Physics B, 2009. 95(1): p. 85-96. 15. H. Karakuzu, M. Dubov, S. Boscolo, L. A. Melnikov, and Y. A. Mazhirina, Optimisation of microstructured waveguides in z-cut LiNbO3 crystals. Opt. Mat. Express, 2014. 4(3): p. 541-552. 16. H. D. Nguyen, A. Ródenas, J. R. Vázquez de Aldana, J. Martínez, F. Chen, M. Aguiló, M. C. Pujol, and F. Díaz, Heuristic modelling of laser written mid-infrared LiNbO3 stressed-cladding waveguides. Opt. Express, 2016. 24(7): p. 7777-7791. 17. Q. An, Y. Ren, Y. Jia, J. R. Vázquez de Aldana, and F. Chen, Mid-infrared waveguides in zinc sulfide crystal. Opt. Mater. Express, 2013. 3: p. 466-471. 18. A. Rodenas, L. M. Maestro, M. Ramirez, G. A. Torchia, L. Roso, F. Chen, and D. Jaque, Anisotropic lattice changes in femtosecond laser inscribed Nd3+:MgO:LiNbO3 optical waveguides. J. Appl. Phys., 2009. 106: p. 013110.