Open Access
Add to BibSonomyAbstract
The fabrication of arrayed waveguide gratings (AWGs) using the femtosecond laser direct-write technique is investigated. We successfully demonstrate the fabrication of large planar waveguides that act as 2D free propagation zones. These slabs were found to have a highly uniform refractive index with a standard deviation of 1.97% relative to the total index contrast. The incorporation of low loss linear adiabatic tapers resulted in an increase of transmission by 90%. Strategies for manufacturing integrated laser written AWGs using continuous contouring to avoid lossy defects are discussed and demonstrated.
© 2015 Optical Society of America
1. Introduction
Arrayed waveguide gratings (AWGs) are typically used by the telecommunications industry as wavelength multiplexers/demultiplexers since they are highly stable, compact, and low loss [1]. AWGs have recently been used for applications such as: spectral domain optical coherence tomography [2], miniaturized Raman spectroscopy [3], integrated optical tweezers [4], and as compact spectroscopic sensors for biomedical [5] and astronomical applications [6].
Currently AWGs are mass produced by photolithography. However, when developing AWGs for new applications, small-scale fabrication runs for prototyping and specific devices are expensive and time consuming. A more recent alternative for rapidly fabricating integrated photonic circuitry is the femtosecond (fs) laser direct-write technique [7]. This technique allows for fast and cost effective manufacturing of devices such as: three dimensional splitters [8], waveguide lasers [9], photonic lanterns [10, 11], stellar interferometers [12], and integrated multi-mode filters [13].
All of these devices rely on single or multi-mode waveguides. To laser fabricate an AWG, both single-mode waveguides and planar slab waveguides termed free propagation zones (FPZ) are required. To date there have been four reports on the fabrication of slab waveguides namely Watanabe et al. [14], Da-Yong et al. [15], Szameit et al. [16] and, Ghosh et al. [17]. Watanabe et al. laser fabricated a multi-mode interference device (MMI) in fused silica using an amplified Ti: sapphire laser centered at 800 nm, 85 fs pulses, 1 kHz repetition rate, 1.5 µJ pulse energy, 0.3 numerical aperture (NA) objective and a writing speed of 60 µm/min. Using longitudinal geometry a 100 µm long filament was induced that was scanned to fabricate a 30 µm wide slab. The length of the slab was increased to 870 µm by stitching these sections together at 50 µm intervals. The slab’s refractive index contrast was estimated to be 2.0 10−3. However variations in the slab’s refractive index contrast were observed and attributed×to the head and tail regions of the filament inducing a different index change [14]. Da-Yong et al. also fabricated MMI waveguides in fused silica using an amplified Ti: sapphire laser centered at 800 nm, 120 fs pulses, 1 kHz repetition rate, 0.53 µJ pulse energy, 0.2 NA objective and a writing speed of 3 mm/min. Using a transverse writing geometry a filament was induced and scanned to write an 870 µm long, 48 µm wide slab waveguide. The index contrast of the modification was estimated to be 2.5 × 10−3 with non-uniformities in the filament’s index profile elongating the MMI patterns [15]. Szameit et al. used a Ti: sapphire (Coherent RegA) centered at 800 nm, 160 fs pulses, 100 kHz repetition rate, 250 nJ pulse energy, 20× objective, 90 mm/min writing velocity and a 1 µm multi-scan spacing to fabricate a sinusoidally undulating slab waveguide in fused silica to examine discrete light propagation [16, 18]. Finally Ghosh et al. used a Yb-doped fibre laser centered at 1064 nm, 350 fs pulses, 500 kHz repetition rate, 86 nJ pulse energy, 20× objective, 480 mm/min writing speed and a 0.4 µm multi-scan spacing to form 500 µm wide slab waveguides in erbium-doped bismuthate glass. The slab had single mode guidance in the vertical direction while the horizontal direction was highly multimode [17]. Also noteworthy is Thomson et al. demonstration of a 35 µm × 35 µm multi-mode waveguide [10]. The waveguide was fabricated in a borosilicate glass (Corning Eagle 2000) using 350 fs pulses, 500 kHz repetition rate, 165 nJ pulse energy, 0.4 NA objective, 480 mm/min writing speed and a 0.4 µm multi-scan spacing in the horizontal axis and 8 µm in the vertical axis. The maximum index contrast of the multi-mode waveguide was recorded as 1.76 × 10−3.
The first two approaches are limited because the width or length of the slab’s modification is always restricted by the maximum focal depth and additionally the focusing setup would have to be altered to integrate single-mode waveguides. The final three approaches enable larger slabs to be manufactured. However none of these examples characterise propagation or index profilometry for slab waveguides at the scale required for FPZs. FPZs for AWGs require a width at the end-facet that is wide enough to accommodate numerous single mode waveguides at sufficient separation to avoid cross talk. Additionally the injected light needs sufficient propagation length to diffract in order to illuminate these single-mode waveguides without interacting with the side walls of the slab and thus necessitates FPZ widths of several millimetres.
As the manufacturing speeds in these previous reports are low, the ability to scale up these inscription techniques are highly dependent upon the stability of the laser system over long time periods. If the laser writing power fluctuates during slab fabrication the refractive index across the slab region will be non-uniform. Since AWG operation requires FPZs to have a uniform refractive index change across the slab region, new high speed slab fabrication techniques are necessary to reduce the fabrication time of large slab waveguides, thus reducing laser stability requirements and enabling rapid AWG prototyping.
High speed fabrication of AWGs however is challenging due to a non-tangential transition between the FPZ and the waveguide array arising from the Rowland curvature, meaning specific manufacturing techniques are required to fabricate AWGs quickly while maintaining low loss and slab uniformity. High speed fabrication requires the modifications to be written using a megahertz repetition rate where the sample can be translated at speeds of tens of millimetres per second [19]; However, modifications written in this regime predominately have complex refractive index profiles that contain positive and negative index regions. Index profiles of these types are problematic for slab manufacturing as the modified regions need to be stitched together to form a uniform slab. Therefore a processing technique that produces a prominently positive index change in the megahertz regime is necessary for uniform slab fabrication. Writing modifications in this regime is also beneficial as each individual modification acts as a single-mode waveguide and thus does not require multi-scanning.
In this paper we report a systematic study of the rapid fabrication of each component required to build an AWG. We show that FPZs with propagation lengths ranging from 110 – 710 µm and 200 µm wide can be fabricated at speeds 4 times faster than previously demonstrated using a tightly controlled fs laser operating at megahertz repetition rate. The manufactured slabs have a refractive index contrast of ∆n = (1.5 ± 0.1)×10−3 which is comparable to previous results. The transverse index standard deviation across the slabs was found to be 1.97% relative to the slab index change. Laser fabricated transition tapers are demonstrated to improve throughput by 90%. Finally, the engineering requirements to fabricate complete AWGs ‘on-the-fly’ while maintaining low losses are discussed.
2. AWG operation
A basic schematic of an AWG is shown in Fig. 1. AWGs consist of four key components: input taper, input/output FPZ, FPZ/waveguide array transition taper, and a waveguide array. The basic principle of operation is as follows: light injected into the first FPZ is no longer laterally confined and is able to freely diffract within the uniform refractive index slab. At the end of the first FPZ the divergent light is captured and tapered into an array of individual single mode waveguides. Light is then individually guided in each waveguide of this array. The waveguide array output is delivered into a second FPZ where each wavelength of light constructively interferes into a distinct focal point at the output. Since each waveguide in the waveguide array is incrementally longer, each wavelength has a different phase front tilt resulting in a dispersed spectrum at the output FPZ. Input tapers are not considered in this paper as they have been shown to reduce the spectral resolution [20]. A detailed description of AWG operation can be found in [21].
AWGs in this paper have been designed for a central wavelength of 633 nm. Compact sensors in the visible are of interest to the sensing community as many materials and analytes have specific markers in this region [1], and as compact high resolution spectrograph’s for exoplanet detection and asteroseismology [22]. Furthermore, this wavelength also allows for easy visualisation of scatter at fabrication defects within the devices.
3. Fabrication and characterisation
All structures were inscribed using an ultrafast Ti: sapphire oscillator (FEMTOSOURCE XL 500, Femtolasers GmbH), which emits pulses centered at 800 nm, with a 5.1 MHz repetition rate and a pulse duration of 50 fs. The laser was focused 170 µm deep into an alkaline earth boro-aluminosilicate glass sample (Corning Eagle2000) using a 40×, 0.65 NA focusing objective. The sample was placed on a set of Aerotech 3-axis air-bearing translation stages. By moving the sample, the focal point within the sample is changed generating an associated modification in refractive index.
To characterise the structures, 633 nm light from a HeNe laser was launched via a single-mode fiber (SM600) into the devices. A microscope objective imaged the output near-field profile onto a camera (Pulnix TM-745E). The Spiricon LBA-PC software was used to capture the output mode profiles for post processing. Losses were measured by integrating the intensity of the captured mode profiles.
4. Results
4.1. Slab waveguides
Circular single-mode waveguides at 633 nm with a mode-field diameter 7.3 × 8.1µm (1/e2) were written with a pulse energy of 55 nJ at a translation speed of 2000 mm/min. Using the near-field profile of the waveguide and applying the inverse Helmholtz technique [23] the peak refractive change was determined to be (1.5 ± 0.2) × 10−3. Similarly, using the beam propagation modelling software RSoft BeamPROP the near-field profile was found to match that of a 4 µm step index waveguide with ∆n = 1.5 × 10−3. As this refractive index contrast is lower than that of typical lithographically fabricated waveguides, laser written AWGs require a larger chip footprint as the bend radius of the waveguide array needs to be increased in order to keep bend losses at a minimum. For a 4 µm diameter, step index waveguide with 1.5 × 10−3 index contrast, theoretical bend losses of less than 0.1 dB/mm are expected for bend radii larger than 16.3 mm [24]. Furthermore, larger FPZs are needed due to the smaller diffraction angle.
Slab waveguides were fabricated by multi-scanning single mode waveguides as previously demonstrated by Ghosh et al. [17]. Creating the slab region from single-mode waveguides enables the inscription of waveguides, tapers, and FPZs with identical parameters, thus allowing for all these components to be fabricated in a single continuous translation. This is beneficial for low loss AWGs as explained later in the text. Waveguides were written in the cumulative heating regime which enables high writing speeds of mm/s [25]. The disadvantage of the cumulative heating regime is the formation of complex refractive index profiles containing regions of positive as well as negative index change [26]. The depressed regions are caused by the migration of elements [27]. This causes overlapping waveguides to individually guide as the depressed region overwrites the positive index change of the previous waveguide (see Fig. 2(a)). To enable the fabrication of a smooth slab using cumulative heated waveguides it was found that just above the cumulative heating threshold the depressed regions are minimized, hence enabling waveguides to be overlapped to fabricate continuous slab waveguides as seen in Fig. 2(b).
Fig. 2 (a) End-on image of strong cumulative heated waveguides overlapped with a 3 µm spacing. The depressed regions between the individual modifications are clearly visible and thus inhibit the fabrication of smooth slabs. (b) End-on view of a 200 µm wide slab waveguide with 0.4 µm multi-scan spacing. (c) Near-field output from a 6 mm long slab waveguide with 0.4 µm multi-scan spacing. Scale bar represents 25 µm.
To find the optimum multi-scan pitch 0.5 mm wide × 6 mm long slab waveguides were fabricated with multi-scan spacings of 0.3 – 0.8 µm providing significant overlap between the 4 µm wide waveguides. These slabs where characterized by injecting 633 nm light into the center of the slabs via a single-mode optical fiber (SM600). The near-field output is expected to have a broad Gaussian cross section if the light can freely diffract within the slab waveguide. The recorded output mode-field for multi-scan spacings of 0.3 and 0.4 µm show a Gaussian output with R2 values of 0.9644 and 0.8806, respectively. At 0.5 µm spacing the output profile no longer exhibits a Gaussian distribution, while at a modification pitch of 0.7 µm some individual waveguide guidance is observed. A multi-scan spacing of 0.4 µm was chosen for future slab fabrication since it provides good compromise between the smoothness and fabrication time. Fast fabrication is important because the writing regime is close to the cumulative heating threshold. If the laser power drifts bellow the cumulative heating threshold the slab’s index change will no longer be uniform. An example of a near-field output from a 6 mm long, 0.4 µm multi-scan spacing slab can be seen in Fig. 2(c). The output exhibits single-mode guidance in the vertical and a broad Gaussian in the horizontal.
If the laser fabricated slabs have a uniform refractive index, the slab diffraction should be linear for propagation lengths greater than the Rayleigh range, which is approximately 20 µm. To evaluate the regularity of the induced refractive index change, 200 µm wide slabs of lengths ranging from 110–710 µm were fabricated. These slabs were fabricated with a central single-mode waveguide input by extending the central multi-scan modification. Figure 3 shows the measured near-field profiles of slabs of varying lengths. These near-field outputs have been fitted with the expected near-field profile using the retrieved single-mode waveguide index contrast, size and the slab refractive index calculated in the following paragraph.
Fig. 3 Experimental and simulated near-field profiles of 633 nm light diffraction within a 200 µm wide slab waveguide at different propagation lengths. Light is injected into the slab via a 4 µm wide 1.5 × 10−3 index contrast single-mode input waveguide. The simulation assumes a slab index contrast of 1.5 × 10−3.
Gaussian profiles were fitted to the near-field profiles in order to determine the FWHM of the mode. Beam propagation software was then used to model the diffraction of up to 710 µm long slabs for refractive index contrasts ranging from 0.5 to 3 × 10−3 at 1 × 10−4 increments. Experimental and theoretical beam profile widths (FWHM) are plotted as a function of slab length in Fig. 4. The best fit for the refractive index contrast of the slab was found to be ∆n = (1.5 ± 0.1) × 10−3. This index contrast is in good agreement with the peak index contrast of the single-mode waveguides found using the inverse Helmholtz technique. Therefore the small negative index region around the single-mode waveguides does not decrease the overall index change of the slab waveguide nor does the close spacing increase the index change. The slab waveguides of different lengths were individually fabricated (e.g. not a single slab that was polished back) across different days to investigate the repeatability. As the beam diffraction of these individual slabs closely follows the model, it implies that the input waveguide and slab refractive index are both highly reproducible and uniform. The refractive index contrast of these slabs are similar to the 2.0 × 10−3 reported by Watanabe et al. [14].
Fig. 4 The FWHM of experimental slab modes of varying lengths were plotted against theoretical slab mode FWHMs for varying refractive index contrasts. The best fit for the refractive index contrast was found to be 1.5 × 10−3. The dash lines represent 95% prediction interval bounds of the fit.
Top-down differential interference contrast (DIC) images of the slab waveguides (see Fig. 5(a)) show a ripple structure that runs parallel to the slab writing direction. Quantitative phase microscopy (QPm) [28] was used to obtain a refractive index profile of the slab. QPm combines 3 bright field images, one in focus, one slightly positive and one negatively defocused image to obtain a phase image that can be converted to a refractive index map. The QPm phase images contain large low frequency image artefacts that cause the refractive index map to exhibit a slow curvature across the image. These features were carefully removed using Fourier filtering. The standard deviation of the slabs refractive index was found to be 1.97% of the slab’s total index contrast. This is larger than the standard deviation of the background arising from QPM artifacts which has a variation of only 0.31%. Fourier transforms of QPm refractive index profiles taken using the highest available magnification (100 × 1.3 NA objective) revealed a small feature with a periodicity of 0.4 µm equivalent to the pitch of the individual modifications that form the slab waveguide. However there are also low spatial frequency index contrast variations across the slab which are the main cause of the 1.97% slab refractive index standard deviation. These larger low spatial frequency variations could be due to other causes such as the laser power or translation speed varying slightly during the writing process.
Fig. 5 (a) DIC image of a 200 µm wide slab. The insert shows ripples in the slab parallel to the writing direction. The markers indicate the edge of the slab region. (b) Example of a laser written taper. The laser modifications that form the slab on the bottom of the image are extended at variable lengths to form the tapers. The dots are formed when cumulative waveguide writing is suddenly stopped.
4.2. Waveguide tapers
The interface between the FPZ and the waveguide array is a major source of loss due to the mode-profile mismatch. To adiabatically convert the FPZ Gaussian to the waveguide array, techniques such as vertical [29], segmentation [30], and adiabatic tapers can be utilized. For initial testing linear adiabatic tapers were chosen for laser fabrication since their implementation using the fs direct-writing technique is relatively straight forward. Using beam propagation modelling, tapers were designed to be 600 µm long for maximum throughput. To test the effectiveness of laser written tapers in reducing the coupling loss at the slab/waveguide transition, slabs with 11 output waveguides were fabricated with and without tapers.
Figure 5(b) shows an example of a tapered transition. To fabricate these tapers each individual cylindrical modification that forms the slab region is extended by a precise distance to create the triangular taper shape. In Fig. 5(b) small dots can be seen along the edge of the tapers. These dots are located where each individual modification is stopped by shuttering the laser. When the beam is suddenly blocked the cumulative heating process that causes the index change is suddenly stopped. This results in a small index defect that scatters light and thus possibly reduce the efficiency of the tapers. The possibility of ramping down the laser power to reduce these dots and thus improve taper efficiency will be considered in future work. Light was injected into the slabs and the output from the waveguide array was recorded. Slabs without transition tapers had a throughput of 28%, while slabs with transition tapers had a throughput of 54%. As shown in Table 1 these tapers match the theoretical throughput value within error. These measurements have an error of 8% due to probe laser power fluctuations. Using a more stable laser would be beneficial to accurately measure the efficiency of the fabricated tapers and the losses caused by the small index scattering centers.
Table 1. Experimental and theoretical taper transmission.
5. Integration
To fabricate a complete AWG on a single chip, the previously discussed components, FPZs, tapers, and waveguides need to be integrated. This can be accomplished in two ways; stitching where each component is fabricated separately or, continuously where the whole structure is fabricated using continuous scans without stopping at the interfaces between each component. One major design element that causes fabrication difficulty are the radially positioned waveguides/tapers with respect to the Rowland curvature. As this transition is non-tangential the stitching fabrication technique can easily manufacture these designs while the continuous contour fabrication of these non-tangential transitions is challenging due to mechanical restrictions of the translation stages. However the stitching technique was avoided due to the laser on/off defects that would cause large scattering losses at each transition. To fabricate AWGs continuously whilst avoiding fabrication errors an exponential smoothing function was used. This function smooths the non-tangential transitions making the motion possible for mechanical translation stages. The smoothing function alters the outer waveguide taper’s position by less than 3 µm. Therefore with the correct parameters non-tangential transitions can be made with negligible effect on the waveguide array phase shift.
Figure 6 shows a DIC image of a single-mode waveguide input into a slab (a) and a slab/waveguide transition (b). Figure 6(a) demonstrates how sharp features such as the Row-land curvature can be manufactured in conjugation with a 800 µm large slab region. Figure 6(b) illustrates the tapers fanning out in a perpendicular orientation to the slab curvature and linearly tapering down into a single waveguide. This demonstrates the possibility to integrate and laser fabricate all the key AWG components on a single chip. This initial prototype has a total chip footprint of 28.7×3.88 mm with two 6.9×0.8 mm FPZs and a minimum bend radius of 27 mm in the waveguide array to keep bend losses to a minimum. Initial testing of the AWG prototypes has shown an aperiodic interference pattern at the output. Similar output interference patterns could be reproduced in beam propagation simulations by inducing a random phase error between 0 – 100 degrees in each waveguide of the waveguide array. Therefore phase control in the waveguide array is a major factor restricting the fabrication of a fully functional device.
Fig. 6 (a) and (b) are DIC images of an AWG prototype. (a) Shows the input FPZ while (b) shows the slab/waveguide transition. The white dotted line highlights the inner Rowland curvature.
Using an interferometric free space method [31] the magnitude and possible causes of these phase errors could be determined. AWGs can also be designed with a larger path length difference between each waveguide thus reducing the effect of small phase errors but decreasing the devices’ free spectral range. The waveguide separation can also be increased to reduce the possibility of cross talk however this increases the size of the FPZ. A possible technique to improve the phase control is laser post tuning. Current lithographically fabricated AWGs undergo a post processing technique known as UV trimming to adjust the phase of individual waveguides or the whole waveguide array. This technique can adjust the central wavelength [32], reduce cross talk [33], and control dispersion [34]. A similar technique using the fs laser could fine tune the refractive index to correct for phase errors.
6. Summary
We have successfully fabricated slab waveguides for the use in devices such as AWGs and MMIs at speeds 4 × faster than previously demonstrated. The fabricated slabs have a refractive index contrast of ∆n = (1.5 ± 0.1) × 10−3. From quantitative phase measurements the slabs were found to have a highly uniform refractive index with a standard deviation of 1.97% relative to the total index contrast. The ability to laser fabricate low loss tapers to improve the slab/waveguide transition throughput by 90% was demonstrated. Finally, the integration of all the key AWG components ‘on-the-fly’ was demonstrated in a complete AWG prototype. So far this work has been restricted to a central wavelength of 633 nm. However techniques such as vertical stacking [10] could enable slab waveguides to be developed for longer wavelengths, thus enabling laser written AWGs to be fabricated in a range of wavelengths for specific applications.
Acknowledgments
This research was supported by the Australian Research Council Centre of Excellence for Ultrahigh bandwidth Devices for Optical Systems (project number CE110001018) and was performed in part at the OptoFab node of the Australian National Fabrication Facility utilizing Commonwealth as well as NSW and SA State Government funding. G. Douglass acknowledges the support of the MQRES scholarship. Felix Dreisow gratefully acknowledges support by the Carl Zeiss Stiftung and the European Commission (Curie FP7-PEOPLE-IRSES project e-FLAG, contract 247635). S. Gross acknowledges a Macquarie University Research Fellowship (MQRF).
References and links
1. K. Suzuki, Y. Hida, T. Shibata, Y. Inoue, H. Takahashi, and K. Okamoto, “Silica-based arrayed-waveguide gratings for the visible wavelength range,” NTT Tech. Rev. 4, 48–52 (2006).
2. B. I. Akca, V. D. Nguyen, J. Kalkman, N. Ismail, G. Sengo, F. Sun, A. Driessen, T. G. van Leeuwen, M. Pollnau, K. Wörhoff, and R. M. de Ridder, “Toward spectral-domain optical coherence tomography on a chip,” IEEE J. Sel. Top. Quantum Electron. 18, 1223–1233 (2012). [CrossRef]
3. N. Ismail, L. P. Choo-Smith, K. Wörhoff, A. Driessen, A. C. Baclig, P. J. Caspers, G. J. Puppels, R. M. de Ridder, and M. Pollnau, “Raman spectroscopy with an integrated arrayed-waveguide grating,” Opt. Lett. 36, 4629–4631 (2011). [CrossRef] [PubMed]
4. T. Saida, T. Shibata, M. Ishii, and H. Takahashi, “Integrated optical tweezers using arrayed waveguide grating,” in Conference on Lasers and Electro-Optics OSATechnical Digest Series (Optical Society of America), paper CMEE6 pp. 556–558 (2005).
5. Y. Komai, H. Nagano, K. Kodate, K. Okamoto, and T. Kamiya, “Application of arrayed-waveguide grating to compact spectroscopic sensors,” Jpn. J. Appl. Phys. 43, 5795–5799 (2004). [CrossRef]
6. N. Cvetojevic, N. Jovanovic, C. Betters, J. S. Lawrence, S. C. Ellis, G. Robertson, and J. Bland-Hawthorn, “First starlight spectrum captured using an integrated photonic micro-spectrograph,” Astron. Astrophys. 544, L1 (2012). [CrossRef]
7. C. Florea and K. A. Winick, “Fabrication and characterization of photonic devices directly written in glass using femtosecond laser pulses,” J. Light. Technol. 21, 246–253 (2003). [CrossRef]
8. S. Nolte, M. Will, J. Burghoff, and a. Tuennermann, “Femtosecond waveguide writing: A new avenue to three-dimensional integrated optics,” Appl. Phys. A. 77, 109–111 (2003). [CrossRef]
9. S. Taccheo, G. D. Valle, R. Osellame, G. Cerullo, N. Chiodo, P. Laporta, O. Svelto, A. Killi, U. Morgner, M. Lederer, and D. Kopf, “Er:Yb-doped waveguide laser fabricated by femtosecond laser pulses,” Opt. Lett. 29, 2626–2628 (2004). [CrossRef] [PubMed]
10. R. R. Thomson, T. A. Birks, S. G. Leon-Saval, A. K. Kar, and J. Bland-Hawthorn, “Ultrafast laser inscription of an integrated photonic lantern,” Opt. Express 19, 5698–5705 (2011). [CrossRef] [PubMed]
11. I. Spaleniak, N. Jovanovic, S. Gross, M. J. Ireland, J. S. Lawrence, and M. J. Withford, “Integrated photonic building blocks for next-generation astronomical instrumentation II: the multimode to single mode transition,” Opt. Express 21, 27197–27208 (2013). [CrossRef] [PubMed]
12. N. Jovanovic, P. G. Tuthill, B. Norris, S. Gross, P. Stewart, N. Charles, S. Lacour, M. Ams, J. S. Lawrence, A. Lehmann, C. Niel, J. G. Robertson, G. D. Marshall, M. Ireland, A. Fuerbach, and M. J. Withford, “Starlight demonstration of the Dragonfly instrument : an integrated photonic pupil-remapping interferometer for high-contrast imaging,” Mon. Not. R. Astron. Soc. 427, 806–815 (2012). [CrossRef]
13. I. Spaleniak, S. Gross, N. Jovanovic, R. J. Williams, J. S. Lawrence, M. J. Ireland, and M. J. Withford, “Multiband processing of multimode light: Combining 3D photonic lanterns with waveguide Bragg gratings,” Laser Photonics Rev. 8, 1–5 (2014). [CrossRef]
14. W. Watanabe, Y. Note, and K. Itoh, “Fabrication of multimode interference waveguides in glass by use of a femtosecond laser,” Opt. Lett. 30, 2888–2890 (2005). [CrossRef] [PubMed]
15. L. Da-Yong, L. Yan, D. Yan-Ping, G. Heng-Chang, Y. Hong, and g. Qi-Huang, “Transverse writing of multimode interference waveguides inside silica glass by femtosecond laser pulses,” Chinese Phys. Lett .25, 2500–2503 (2008). [CrossRef]
16. A. Szameit, F. Dreisow, M. Heinrich, R. Keil, S. Nolte, A. Tünnermann, and S. Longhi, “Geometric potential and transport in photonic topological crystals,” Phys. Rev. Lett. 104, 1–4 (2010). [CrossRef]
17. S. Ghosh, N. D. Psaila, R. R. Thomson, B. P. Pal, R. K. Varshney, and A. K. Kar, “Ultrafast laser inscribed waveguide lattice in glass for direct observation of transverse localization of light,” Appl. Phys. Lett. 100, 101102 (2012). [CrossRef]
18. F. Dreisow, Institute of Applied Physics, Friedrich-Schiller-University Jena, Max-Wien-Platz 1, 07743 Jena, Germany (personal communication, 2014).
19. S. M. Eaton, H. Zhang, M. L. Ng, J. Li, W.-J. Chen, S. Ho, and P. R. Herman, “Transition from thermal diffusion to heat accumulation in high repetition rate femtosecond laser writing of buried optical waveguides,” Opt. Express 16, 9443–9458 (2008). [CrossRef] [PubMed]
20. N. Cvetojevic, N. Jovanovic, J. Lawrence, M. Withford, and J. Bland-Hawthorn, “Developing arrayed waveguide grating spectrographs for multi-object astronomical spectroscopy,” Opt. Express 20, 2062–2072 (2012). [CrossRef] [PubMed]
21. M. K. Smit and C. V. Dam, “PHASAR-based WDM-devices : principles, design and applications,” Sel. Top. Quantum Electron. 2, 236–250 (1996). [CrossRef]
22. T. Feger, C. Bacigalupo, T. Bedding, J. Bento, D. Coutts, M. J. Ireland, Q. Parker, A. Rizzuto, and I. Spaleniak, “RHEA: the ultra-compact Replicable High-resolution Exoplanet Asteroseismology spectrograph,” Proc. SPIE 9147, 914771 (2014).
23. I. Mansour and F. Caccavale, “An improved procedure to calculate the refractive index profile from the measured near-field intensity,” J. Light. Technol. 14, 423–428 (1996). [CrossRef]
24. A. W. Snyder and J. Love, Optical Waveguide Theory, (Springer Science & Business Media, 2012), Chap. 24.
25. C. Schaffer, J. García, and E. Mazur, “Bulk heating of transparent materials using a high-repetition-rate femtosecond laser,” Appl. Phys. A. 76, 351–354 (2003). [CrossRef]
26. A. Arriola, S. Gross, N. Jovanovic, N. Charles, P. G. Tuthill, S. M. Olaizola, A. Fuerbach, and M. J. Withford, “Low bend loss waveguides enable compact, efficient 3D photonic chips,” Opt. Express 21, 2978–2986 (2013). [CrossRef] [PubMed]
27. S. Kanehira, K. Miura, and K. Hirao, “Ion exchange in glass using femtosecond laser irradiation,” Appl. Phys. Lett. 93, 023112 (2008). [CrossRef]
28. A. Barty, K. A. Nugent, D. Paganin, and A. Roberts, “Quantitative optical phase microscopy,” Opt. Lett. 23, 817–819 (1998). [CrossRef]
29. A. Sugita, A. Kaneko, K. Okamoto, M. Itoh, A. Himeno, and Y. Ohmori, “Very low insertion loss arrayed-waveguide grating with vertically tapered waveguides,” IEEE Photonics Technol. Lett. 12, 1180–1182 (2000). [CrossRef]
30. P. L. Yuan and G. Duluth, “Optical device having low insertion loss,” US Patent 5745618 (1998).
31. R. Zektzer, D. Sinefeld, N. Goldshtein, and D. Marom, “Interferometric method for phase error measurement in an arrayed waveguide grating with a free space output,” in National Fiber Optic Engineers Conference, OSA Technical Digest Series (Optical Society of America, 2012), paper JTh2A.1 [CrossRef]
32. D. Zauner, J. Hubner, K. Malone, and M. Kristensen, “UV trimming of arrayed-waveguide grating wavelength division demultiplexers,” Electron. Lett. 34, 780–781 (1998). [CrossRef]
33. J. Gehler and F. Knappe, “Crosstalk Reduction of Arrayed Waveguide Gratings by UV trimming of individual waveguides without H2-Loading,” in Optical Fiber Communication ConferenceOSA Technical Digest Series (Optical Society of America, 2000), paper WM9.
34. M. Abe, K. Takada, T. Tanaka, M. Itoh, T. Kitoh, and Y. Hibino, “Reduction in dispersion of silica-based AWG using photosensitive phase trimming technique,” Electron. Lett. 38, 1673–1675 (2002). [CrossRef]
