A method to compensate for fabrication tolerances and to fine-tune individual photonic circuit components is inevitable for wafer-scale photonic systems even with most-advanced CMOS-fabrication tools. We report a cost-effective and highly accurate method for the permanent trimming of hydrogenated amorphous silicon photonic devices by UV-irradiation. Microring resonators and Mach-Zehnder-interferometers were utilized as photonic test devices. The MZIs were tuned forth and back over their complete free spectral range of 5.5 nm by locally trimming the two MZI-arms. The trimming range exceeds 8 nm for compact ring resonators with trimming accuracies of 20 pm. Trimming speeds of ≥ 10 GHz/s were achieved. The components did not show any substantial device degradation.
© 2014 Optical Society of America
Silicon material based photonic devices have experienced tremendous progress in recent years and a large scale fabrication of photonic and combined photonic-electronic integrated circuits show a strong uptrend in developing markets like datacom, telecommunication, and biosensing. In particular, low optical loss and CMOS-compatible photonic materials with high refractive index contrast (HIC:Δn ≈ 2) like crystalline (SOI), hydrogenated amorphous (a-Si:H), and poly-silicon are attractive due to their strong light confinement which enables densely packed photonic systems with low-footprint. Among these materials, a-Si:H is interesting since it is backend-compatible and provides unique fabrication and material properties. The material exhibits low absorption loss at telecommunication wavelengths which supports low-loss waveguiding [1–4] and the fabrication of all common passive devices like e.g. resonators and interferometers [4, 5]. The flexible deposition process enables the fabrication of three-dimensional shaped fiber-chip-couplers [6, 7], and facilitates multilayer-stacking of photonic circuits which allows for efficient waveguide crossings or the separation of photonic devices on different vertically arranged layers that are connected by optical vias [8–11]. Dependent on the deposition conditions, the material possesses high nonlinearities favorable for numerous all-optical and ultra-fast signal processing applications [12–17]. Furthermore, a-Si:H can be doped and crystallized for e.g. electro-optical modulation [18–20], and it can be deposited on various substrate materials like glasses and flexible plastics which makes it interesting as sensing platform for the detection of physical and biochemical quantities [21, 22].
Although low-loss a-Si:H can be deposited with high uniformity , the employment of any HIC photonic material implies strong fabrication challenges since almost all integrated-optical applications and devices strongly suffer from local variations of waveguide dimensions and refractive index (RI). This results in deviations of the targeted device performance and the spectral characteristics will slighly differ from their ideal design on a wafer-scale level. Thus, it is common practice to actively tune the photonic components by e.g. microheaters or free carrier injectors employing the thermo-optic (TOE) and free carrier dispersion (FCD) effects, and/or to permanently trim the photonic components by a direct modification of the guided mode. Compared to an active tuning, the main advantage of a permanent device trimming is the reduction of photonic circuits static power consumption which is low for a single device, however, becomes a crucial issue for hundreds or even thousends of devices per wafer. Furthermore, as soon as highly desirable athermal solutions are available the TOE naturally loses performance , so that e.g. the weaker FCD effect needs to be employed for the active tuning and reconfiguration of complex photonic networks and the individual circuit components need to function as precise as possible to the design specifications after fabrication.
Recently, various promising compensation techniques for HIC photonic devices have been explored, which are either based on a permanent RI modification of the waveguide core or the surrounding cladding material. Both approaches tailor the guided mode and hence permit to counterbalance the fabrication imperfections. Waveguide core material trimming methods have been proposed with local oxidation e.g. by using an electrically biased atomic force microscope tip  or by local heating with high power lasers , with ion implantation inducing point defects , and by amorphization and milling techniques using femtosecond laser pulses . Cladding material modification methods were reported by straining the bottom oxide with electron beam irradiation , and by using photo-sensitive materials like chalcogenides  or polymers [31,32] that can be altered in their RI either by UV-light exposure or by electron beam bleaching. For the global trimming on wafer-scale atomic layer deposition techniques , and silicon nitride overlayers have been reported to reduce device non-uniformities .
In this work, we present a precise and spatially resolved post-fabrication UV-trimming method for a-Si:H photonic devices. Mach-Zehnder-interferometers (MZIs) and microring resonators (MRRs) were fabricated as photonic test devices. The capability to reconfigure multistage filters is proven by successively trimming the MZI-arms which resulted in a full 2π shift to both directions without any notable device degradation. Compact 10 μm radius MRRs were permanently trimmed by a free spectral range (FSR) of 8.5 nm, with a mean accuracy of ≤ 20 pm. The trimming results, underlying physical phenomena, and fabricational aspects are discussed.
2. Modeling and fabrication
2.1. Theoretical background and simulations
The functionality of photonic wire based integrated optical interferometers and resonators depend on the guided mode index (neff) properties. Thus, if the propagating mode index is changed the spectral characteristics of the photonic component can be controlled. In case of a MRR the resonance wavelength shift Δλres (RWS) as a function of the mode index change Δneff which is locally modified over a trimming length Ltrim can be expressed as follows:Fig. 1. Due to their compact size HIC photonic wires are highly sensitive to small RI variations such that a large trimming range of e.g. 10 nm can be realized with sub-percentile index modifications of Δna–Si:H ≈ Δneff ≈ 0.027 for the TE-mode, whereas for the TM-mode the trimming range is even higher.
The photonic devices were fabricated on low-loss hydrogenated amorphous silicon on oxide (a-SOI) platform. Standard 10 cm silicon wafers (SSP) were used as substrates, on which 3 μm thermal oxide was grown by furnace process at 1050°C. Then optimized hydrogenated amorphous silicon was deposited to a targeted thickness of 200 nm as described more precisely in . After photonic core layer deposition the lithography was carried out with electron beam lithography (EBL) at HHI-Berlin. Prior to the a-Si:H etch, an O2 descum process with low platen and coil powers was applied in order to smooth the resist which helps to reduce waveguide side-wall roughness and minimizes detrimental scattering loss. Subsequently, the photonic device patterning was performed in an inductively coupled plasma (ICP) etcher assisted by SF6 and C4F8 reactive gases. Remaining EBL-resist was directly removed by an O2 cleaning plasma. The wafer was fragmented and the individual chips were cleaned with wet chemicals and left uncladded for the experiments. Compact MRRs with 10 μm radius and MZIs with directional coupler based 3 dB splitters and a path length difference of 150 μm were utilized as photonic demonstrators. Completely etched apodized grating couplers were used for the light coupling which are linearly tapered down to the photonic wires. This simplifies the fabrication because it only requires a single lithography and etch step. The photonic wire propagation loss was measured with virtual cut-back method and was determined to be 3.25 dB/cm at 1550 nm wavelength with a protective upper cladding of I-line photo resist.
3. Trimming procedure and optical device characterization
3.1. Trimming setup
The MRRs and the MZIs were characterized with spectral transmission measurements. The grating couplers were designed for high coupling efficiency for TE-mode only, thus polarization paddles were sufficient to maximize the power being coupled to the photonic chip. Three-axis flexure alignment stages with fiber mounts were used to maximize the fiber-chip-coupling. The outcoupled light was 3dB-splitted such that it could be guided among two different optical paths. An optical spectrum analyzer (OSA) was used for the inline monitoring during the trimming procedure employing the broadband emission of an erbium doped fiber amplifier (EDFA) as light source. The other paths included a higher resolution photodetector for more precise spectral analysis. After each trimming step the NIR-light from the tunable laser source (TLS, Agilent 8064A) was swept over the desired wavelength range and data-logged with an builtin high-sensitivity photodetector (PD, Agilent 81634B). In order to compensate for the inherently strong TOE of a-Si:H (≈ 2 · 10−4 [1/°C]) a thermo-electric controller (TEC) was used to stabilize the temperature of the chip-mount and stable conditions with ΔT ≤ 0.01°C (Δλres ≤ 1 pm) were ensured throughout the measurements. The measurement setup and microscope pictures of the photonic devices under test (DUT) are depicted in Fig. 2.
A low-cost continuous wave UV-laser with 405 nm wavelength (≈3.01 eV) was used for the RI trimming. The light was supplied into a single-mode UV-fiber by a fiber-collimator such that the cleaved fiber-end could be positioned with μm-precision on top of the photonic devices. The fiber mode field diameter (MFD) is about 2.59 μm and allows for high irradiation densities. The maximum output power was adjusted up to P=6 mW. Since a free space Gaussian beam is a good approximation for the mode profile of single-mode-fibers, the intensity profile is calculated according to Eq. (2), with as maximal beam intensity at the middle of the focal point r=z=0 (e.g. ≈ 3.8 · 10−4W/μm2 for 1 mW), ω2(z) as spot size with respect to propagation direction z, and as lateral fiber to photonic waveguide offset:
3.2. Trimming experiments
The UV-trimming experiments were carried out with MZIs and MRRs as demonstrators since a RI change is directly translated into measurable spectral quantities which are the MZI beat fringes and the MRR resonances. Moreover these devices are key building blocks in almost any photonic circuit. MZIs were used to prove the RI trimming direction because the beat fringes shift either to longer or shorter wavelength dependent on which MZI-arm is irradiated. In case of an RI increase in the long arm the output signal experiences a redshift, whereas a blueshift occurs by irradiation of the short arm. The shifts are vice versa for a RI decrease.
The device trimming procedure was performed as follows. The fiber was positioned on top of the photonic waveguides using alignment cameras and the optimum fiber position was found by maximizing the shift of the characteristic device peaks due to the TOE-responce with low input powers which was monitored with the OSA. After that the power was increased and the actual measurements were carried out. First the trimming was performed with the MZIs aiming to tune a full FSR forth and back in π/2 steps. Therefore, the 150 μm longer arm was irradiated until the signal was moved by the FSR of 5.5 nm. After that the spectrum was shifted back to the initial position by trimming the short arm. As presented in Fig. 3(a), the long arm irradiation results in a successive blueshift to lower wavelength, whereas the shift is oppositely directed for the short arm. The spectra are offset by 5 dB to guide the eye. From this it is concluded that the mode index and hence the RI decreases. The MZI performance in C-band is analyzed by the beat fringe extinction ratios (ER) and no significant degradation is observed as shown in Fig. 3(b).
The MRR trimming experiments were conducted in order to analyze following key aspects: The time dependence of the trimming with various power levels including the trimming velocity, the device tuning accuracy, the maximal trimming range, the impact on the MRR Q-factors which reflect a change in propagation loss, and finally the temporal stability. The MRRs free spectral range is ≈ 8.5 nm so a distinct allocation of the RWS is possible. First the temporal behaviour of the RWS was analyzed by trimming the MRRs with three different input power levels supplied to the UV-fiber. It should be noted that due to the small fiber MFD only a small fraction of the MRRs waveguide area of ≈ 5% was locally exposed at the opposite side of the directional coupler region. The experimentally determined RWS within an time interval of 10 minutes for 1, 3, and 5 mW are summarized in Fig. 4(a). The results for a fixed intensity follow an exponential behaviour which allows to accurately estimate the RWS for a given light intensity. The trimming velocity for P=5 mW in the unsaturated close to linear regime is determined to be 0.83 GHz/s from which it is estimated that when irradiating the whole MRR area instead of a small fraction with same irradiation density a velocity of ≥ 10 GHz/s is possible. However, the trimming speed does not linearly scale upon device exposure dose as evidenced by trimming the same MRR area increasing the power after 10 min as shown in Fig. 4(b).
The trimming accuracy was determined by shifting the resonance peaks from their initial spectral positions in 1 nm steps to lower wavelengths. A Lorentzian fit was performed on each peak in order to improve the spectral resolution of the measured data and to evaluate the Q-factors from the resonance linewidth at full width at half maximum. The measured resonances that are separated by their modal order in the subfigures are presented in Fig. 5(a). The accuracy is determined by calculating the offset of each RWS from the targeted value. The results for each trimming step are shown in Fig. 5(b). The mean offsets ΔRWS(λres − λtarget) are about 20 pm (≈ 2.5 GHz) with the minima well below 10 pm, and the maxima ≤45 pm. The MRR Q-factor mean values of each trimming step are presented in Fig. 5(c).
The results reveal that there is no distinct trimming induced device degradation observable. The potential UV-trimming range for highly focused 405 nm light of about 5 mW is estimated to be ≥ 10 nm for the TE-mode and the used waveguide dimensions of 480 x 200 nm since the irradiation of ≈ 1/20 of the total ring area resulted in a 1 nm RWS as shown in Fig. 4. However, the maximal shift achieved during the experiments so far was about one FSR of 8.5 nm in case of a 10 μm MRR. The temporal stability of this MRR that was simply stored in ambient environment was analyzed several times over a time period of about 4 months and the relative deviation with respect to the total RWS was found to be ≤ 3%. This minor effect is most-likely attributed to a structural relaxation of strain.
The arrangement of the amorphous silicon network, in particular the structural groups (Si-H1,2,3) in which hydrogen (H) is incorporated in the rigid silicon matrix, has a significant influence on the a-Si:H material properties . High quality a-Si:H for photonic applications is preferentially fabricated with optimized plasma-enhanced chemical vapor deposition processes that minimize the dangling bond density and defects like impurities, microvoids, and H-clusters in order to form the optical bandgap of 1.6 – 1.8 eV with low defect states in the mid-bandgap which promotes a low-loss material at telecommunication wavelength [1–4]. For that reason, a-Si:H is commonly deposited at temperatures of 250–300°C such that hydrogen is bound predominantly in monohydride Si-H facilitating bulk material losses down to 0.04 dB/cm at 1550 nm . The RI of optimized a-Si:H is close to n=3.5 and can be slightly tailored around the transparency window by the plasma chemistry e.g. process temperature, pressure, gas flow rate, precursor to dilutor concentration (usually SiH4 with Ar, He, or H2), and RF-power .
Apart from modifying the material properties in-situ during the plasma process, a-Si:H thin films can be significantly altered post-deposition e.g. by thermal and laser processing. In this work a 405 nm wavelength laser was utilized to permanently tune the spectral characteristics of a-Si:H photonic devices by trimming the material RI. Thus, both photothermal and photolytic effects are involved. However, due to the complex interaction of the various influencing factors, the experimental data do not allow to clearly distinguish the origin of the measured RWS at this point and hence the results are discussed in the mainframe of the probably most relevant physiochemical phenomena involved and in comparison with other trimming methods.
4.1. Thermo- and photo-induced effects in a-Si:H
First the thermo-induced effects are discussed. The structural changes due to heating can be in a simplified way categorized into three temperature regimes: 1. The material properties can be termed stable up to the deposition temperature of 300 ± 100°C. 2. In the temperature range of 300 ± 100 – 600°C significant structural changes of the a-Si:H material are observed and the material transparency strongly degrades due to dehydrogenation. 3. At elevated temperatures starting from 600°C the amorphous nature is basically transformed into nano-/micro-, or poly-crystalline morphology and the material can exhibit tolerable propagation losses with better electrical conductivity .
This coarse classification is in agreement with studies which have evidenced an increase of photonic waveguide propagation loss as a result of hydrogen diffusion at temperatures starting from 350 – 400°C [37, 38]. In particular Ref.  provides relevant data in context of this work since the permanent trimming of MZIs has been studied with thermal annealing on chip-scale. A wavelength blueshift of 6.6 nm has been measured with a linear relationship in the 200 – 400°C temperature range upon 30 min furnace process in a nitrogen ambience. The decrease in RI is explained by H desorption upon annealing, without substantially increasing the dangling bond density due to a rearrangement of the amorphous network. Rapid thermal annealing experiments of our a-Si:H material showed qualitatively similar characteristics as has been reported in . The RI is slightly decreased up to 500°C accompanied with a slight increase in layer thickness and experiences a sharp RI rise up to n ≈ 3.7 for H-free a-Si at temperatures ≥ 550°C. Coincidentally the layer thickness decreases due to dehydrogenation and the layer is compacted. From Raman spectra an onset of crystallization starting at 600°C was deduced. The thermo-induced structural changes are probably contributing factors relevant for the RI-trimming with highly focused laser beams. The thermal conductivity κtc of a-Si:H material is reported to range from 1–5 W/Km [40–42], from which the maximum steady-state surface temperature (Tmax) at the center of the Gaussian spot can be calculated according to . This results in temperatures ranging from 140°C for κtc = 5 W/Km up to 700°C for κtc = 1 W/Km at the maximum laser power of 6 mW that was used in this work. From the above considerations we assume that the waveguide did not notably exceed the dehydrogenation temperature during the trimming procedure since the RI decreased and the devices did not inherently suffer from an increase in propagation loss.
In addition to the photothermal contribution during the UV-trimming the material is subject to photo-induced effects. The laser source of 405 nm provides high photon energies of 3.01 eV that are in principle energetic enough to directly break weak Si-Si, and Si-Hx bonds in the amorphous network with bonding energies of ESi–Si = 1.84 eV and ESi–Hx ≈ 1.8 – 3.1 eV, respectively [44,45]. Furthermore, various metastable phenomena have been identified in amorphous semiconductors . The probably most explored effect in a-Si:H is the Staebler-Wronski effect (SWE) that describes a decrease in conductivity after intense light illumination or by carrier injection . It is widely accepted that photoinduced electron-hole pairs generate dangling bonds due to their energy release during recombination via the conduction- and valence-band-tail states. As reported in , the defect density (ND) increases upon light irradiation following a nonlinear function ND ∝ Gαtβ, e.g. α=0.6 and β=0.33 for continuous wave light, with G being the photoexcited carrier generation rate and t the illumination time.
Various other models have been proposed in order to explain the SWE , with still novel insights being reported about e.g. H-trapping in microvoids , and to solve fundamental questions like the correlation of atomic arrangement on micro- and macroscopic scales . However, merely all related studies do not primarily focus on the relevant material parameters for photonic applications at near-infrared telecommunication wavelengths like absorption or RI, and thus are beyond the scope of this paper. Further material analysis and comparative thermal and light-irradiation studies are required to better understand the proposed UV-trimming method and will be in the focus of our future research work.
4.2. Experimental results and comparison with other trimming methods
The most evident finding of this work, which is to the best of our knowledge the first experimental proof that a-Si:H photonic devices can be locally trimmed by UV-irradiation, is the change of the guided mode index which is directly confirmed by the photonic device performance. The results from Section 3 demonstrate that the mode index is decreased which is deduced from a blue wavelength shift for the long and a redshift for the short MZI-arm, respectively. The RWS upon 405 nm irradiation under the described trimming conditions is determined to be nonlinear in time as well as with regard to the exposure dose. Nevertheless, the inline trimming procedure is proven to be highly accurate achieving low spectral offsets ΔRWS ≤ 20 pm with MRRs as demonstrators. The trimming range was experimentally determined to reach a FSR of 8.5 nm for a 10 μm radius MRR.
Both analyzed devices, MRRs and MZIs, did not inherently degrade in their optical functionalities and the effect is temporally stable within a few percent on month-scale in ambient environment which makes the UV-trimming suitable to counterbalance fabrication inhomogeneities. Apart from the UV-trimming, we did not notice any degradation of photonic samples due to 1550 nm wavelength or to e.g. direct white light illumination for instance over hours on the measurement setup, nor any substantial degradation of samples stored over years in ambient environment and kept under daylight conditions, although similarly as for SOI based devices there will be a small RWS due to natural oxidation for hermetically unsealed photonic devices over years. Nevertheless, these issues need to be explored in more detail.
Compared to the methods described in [25,27–29,33,34] the proposed UV-trimming method is simple and cost-effective since it is based on a low-cost laser source coupled to an optical fiber and does not require advanced fabrication tools. Moreover, this technique is locally selective within a few μm and facilitates a straight-forward way for the real-time compensation during photonic device inspection. From a photonic device perspective such a permanent fine-trimming will significantly improve the static power consumption of passive photonic circuits, or, due to the large trimming range allows a reconfiguration of complex photonic filters and routing networks. The trimming procedure applies as well for other photonic devices which can be modified by their RI like power splitters or fiber-chip-couplers. This is advantageous compared to the global trimming methods which are more useful for the compensation of uniform errors across the wafer but less appropriate to correct random errors. From a fabricational viewpoint, a local trimming through cladding layers or of multilayer-stacked photonic circuits should be possible since commonly used SiO2 is transparent at 405 nm wavelength, which makes this method more flexible than the cladding modification methods in [31, 32]. A drawback of this method can be seen in the slight structural relaxation which was evidenced. For a large scale trimming of thousands of devices the throughput of the current process with velocities of 0.8 GHz/s is too slow, however, only 1/20 of the resonator area was modified at the same time so that the trimming speed can be improved by irradiation of a larger area with higher laser powers or by using more sensitive waveguide geometries. The usage of pulsed and lower wavelength lasers is applicable as well. First experiments with a frequency-doubled argon ion laser at 244 nm wavelength (spot-size ≈ 150μm at Pcw=20 mW) exhibited a RWS of 100 pm (≥ 10 GHz/s) in less than 100 ms so that a fast step-and-repeat process can be realized.
A post fabrication UV-trimming procedure for CMOS backend-compatible a-Si:H photonic devices is presented. This low-cost trimming method allows to permanently fine-trim individual photonic devices with high spectral accuracy and to correct devices like resonators and interferometers over a large spectral range of several nanometers in a step-and-repeat process during device inspection. The time dependence of the trimming with various power levels, the device tuning accuracy, the maximal trimming range and speed, the impact on the device functionalities, and the temporal stability were analyzed. A trimming range of 8 nm, spectral accuracies ≤ 20 pm, and trimming speeds of ≥ 10 GHz/s were determined. The photonic device characteristics do not notably degrade by the trimming procedure.
The authors thank German Research Fund ( DFG grant FOR-653) for funding. Ralf Steingrüber from Fraunhofer Heinrich Hertz Institute (HHI-Berlin) is acknowledged for performing the EBL process. Prof. Ernst Brinkmeyer and Dr. Jost Müller from the Institute of Optical Communication Technology at TUHH are acknowledged for support with the argon ion laser. This publication was supported by the German Research Foundation (DFG) and the Hamburg University of Technology (TUHH) in the funding programme “Open Access Publishing”.
References and links
1. G. Cocorullo, F. G. D. Corte, I. Rendina, C. Minarini, A. Rubino, and E. Terzini, “Amorphous silicon waveguides and light modulators for integrated photonics realized by low-temperature plasma-enhanced chemical-vapor deposition,” Opt. Lett. 21(24), 2002–2004 (1996). [CrossRef] [PubMed]
2. A. Harke, M. Krause, and J. Müller, “Low-loss singlemode amorphous silicon waveguides,” Electronics Letters 41(25), 1377–1379 (2005). [CrossRef]
3. R. Takei, S. Manako, E. Omoda, Y. Sakakibara, M. Mori, and T. Kamei, “Sub-1 dB/cm submicrometer-scale amorphous silicon waveguide for backend on-chip optical interconnect,” Opt. Express 22(4), 4779–4788 (2014). [CrossRef] [PubMed]
4. S. K. Selvaraja, E. Sleeckx, M. Schaekers, W. Bogaerts, D. V. Thourhout, P. Dumon, and R. Baets, “Low-loss amorphous silicon-on-insulator technology for photonic integrated circuitry,” Opt. Commun. 282(9), 1767–1770 (2009). [CrossRef]
5. T. Lipka, O. Horn, J. Amthor, and J. Müller, “Low-loss multilayer compatible a-Si:H optical thin films for photonic applications,” J. Eur. Opt. Soc. Rap. Publicat. 7, 12033 (2012). [CrossRef]
6. A. Harke, T. Lipka, J. Amthor, O. Horn, M. Krause, and J. Müller, “Amorphous silicon 3-D tapers for Si photonic wires fabricated with shadow masks,” IEEE Photon. Technol. Lett. 20(17), 1452–1454 (2008). [CrossRef]
7. H. Yoda, K. Shiraishi, A. Ohshima, T. Ishimura, H. Furuhashi, H. Tsuchiya, and C. Tsai, “A two-port single-mode fiber-silicon wire waveguide coupler module using spot-size converters,” J. Lightwave Technol. 27(10), 1315–1319 (2009). [CrossRef]
8. R. Sun, M. Beals, A. Pomerene, J. Cheng, C. Y. Hong, L. Kimerling, and J. Michel, “Impedance matching vertical optical waveguide couplers for dense high index contrast circuits,” Opt. Express 16,(16), 11682–11690 (2008). [CrossRef] [PubMed]
9. J. Kang, Y. Atsumi, M. Oda, T. Amemiya, N. Nishiyama, and S. Arai, “Low-loss amorphous silicon multilayer waveguides vertically stacked on silicon-on-insulator substrate,” Jpn. J. Appl. Phys. 50, 120208 (2011). [CrossRef]
10. K. Furuya, R. Takei, T. Kamei, Y. Sakakibara, and M. Mori, “Basic study of coupling on three-dimensional crossing of Si photonic wire waveguide for optical interconnection on inter or inner chip,” Jpn. J. Appl. Phys. 51, 4DG12 (2012). [CrossRef]
12. Y. Shoji, T. Ogasawara, T. Kamei, Y. Sakakibara, S. Suda, K. Kintaka, H. Kawashima, M. Okano, T. Hasama, H. Ishikawa, and M. Mori, “Ultrafast nonlinear effects in hydrogenated amorphous silicon wire waveguide,” Opt. Express 18(6), 5668–5673 (2010). [CrossRef] [PubMed]
14. B. Kuyken, H. Ji, S. Clemmen, S. K. Selvaraja, H. Hu, M. Pu, M. Galili, P. Jeppesen, G. Morthier, S. Massar, L. Oxenlwe, G. Roelkens, and R. Baets, “Nonlinear properties of and nonlinear processing in hydrogenated amorphous silicon waveguides,” Opt. Express 19(26), B146–B153 (2011). [CrossRef]
16. C. Grillet, L. Carletti, C. Monat, P. Grosse, B. Ben Bakir, S. Menezo, J. M. Fedeli, and D. J. Moss, “Amorphous silicon nanowires combining high nonlinearity, FOM and optical stability,” Opt. Express 20(20), 22609–22615 (2012). [CrossRef] [PubMed]
17. J. Safioui, F. Leo, B. Kuyken, S. Gorza, S. Selvaraja, R. Baets, P. Emplit, G. Roelkens, and S. Massar, “Supercontinuum generation in hydrogenated amorphous silicon waveguides at telecommunication wavelengths,” Opt. Express 22(3), 3089–3097 (2014). [CrossRef] [PubMed]
18. F. G. Della Corte, S. Rao, G. Coppola, and C. Summonte, “Electro-optical modulation at 1550 nm in an as-deposited hydrogenated amorphous silicon p-i-n waveguiding device,” Opt. Express 19(4), 2941–2951 (2011). [CrossRef] [PubMed]
21. L. Fan, L. Varghese, Y. Xuan, J. Wang, B. Niu, and M. Qi, “Direct fabrication of silicon photonic devices on a flexible platform and its application for strain sensing,” Opt. Express 20(18), 20564–20575 (2012). [CrossRef] [PubMed]
22. T. Lipka, L. Wahn, H. K. Trieu, L. Hilterhaus, and J. Müller, “Label-free photonic biosensors fabricated with low-loss hydrogenated amorphous silicon resonators,” J. Nanophoton. 7(1), 073793 (2013). [CrossRef]
23. T. Lipka, J. Amthor, and J. Müller, “Process and device uniformity of low-loss a-Si:H,” in Proceedings of IEEE Photonics Conf. (IPC) (BurlingameCalifornia, 2012), pp. 923–924.
24. S. Grillanda, V. Raghunathan, V. Singh, F. Morichetti, J. Michel, L. Kimerling, A. Melloni, and A. Agarwal, “Post-fabrication trimming of athermal silicon waveguides,” Opt. Lett. 38(24), 5450–5453 (2013). [CrossRef] [PubMed]
25. Y. Shen, I. Divliansky, D. Basov, and S. Mookherjea, “Electric-field-driven nano-oxidation trimming of silicon microrings and interferometers,” Opt. Lett. 36(14), 2668–2670 (2011). [CrossRef] [PubMed]
26. C. J. Chen, J. Zheng, T. Gu, J. F. McMillan, M. Yu, G. Q. Lo, D. L. Kwong, and C. W. Wong, “Selective tuning of high-Q silicon photonic crystal nanocavities via laser-assisted local oxidation,” Opt. Express 19(13), 12480–12489 (2011). [CrossRef] [PubMed]
27. J. Ackert, J. Doylend, D. Logan, P. Jessop, R. Vafaei, L. Chrostowski, and A. Knights, “Defect-mediated resonance shift of silicon-on-insulator racetrack resonators,” Opt. Express 19(13), 11969–11976 (2011). [CrossRef] [PubMed]
28. O. Bachman, Z. Chen, R. Fedosejevs, Y. Y. Tsui, and V. Van, “Permanent fine tuning of silicon microring devices by femtosecond laser surface amorphization and ablation,” Opt. Express 21(9), 11048–11056 (2013). [CrossRef] [PubMed]
30. A. Canciamilla, F. Morichetti, S. Grillanda, P. Velha, M. Sorel, V. Singh, A. Agarwal, L. C. Kimerling, and A. Melloni, “Photo-induced trimming of chalcogenide-assisted silicon waveguides,” Opt. Express 20(14), 15807–15817 (2012). [CrossRef] [PubMed]
31. L. Zhou, K. Okamoto, and S. J. B. Yoo, “Athermalizing and trimming of slotted silicon microring resonators with uv-sensitive pmma upper-cladding,” IEEE Photon. Technol. Lett. 21(17), 1175–1177 (2009). [CrossRef]
32. S. Prorok, A. Y. Petrov, M. Eich, J. Luo, and A. K. Y. Jen, “Trimming of high-q-factor silicon ring resonators by electron beam bleaching,” Opt. Lett. 37(15), 3114–3116 (2012). [CrossRef] [PubMed]
33. M. Erdmanis, L. Karvonen, M. R. Saleem, M. Ruoho, V. Pale, A. Tervonen, S. Honkanen, and I. Tittonen, “ALD-assisted multiorder dispersion engineering of nanophotonic strip waveguides,” J. Lightwave Technol. 30(15), 2488–2493 (2012). [CrossRef]
35. R. A. Street, “Hydrogenated Amorphous Silicon,” (Cambridge University Press, 1991).
36. T. Lipka, A. Harke, O. Horn, J. Amthor, and J. Müller, “Amorphous silicon as high index photonic material,” Proc. SPIE 7366, paper 73661Z (2009). [CrossRef]
37. S. Y. Zhu, G. Q. Lo, and D. L. Kwong, “Low-loss amorphous silicon wire waveguide for integrated photonics: effect of fabrication process and the thermal stability,” Opt. Express 18(24), 25283–25291 (2010). [CrossRef] [PubMed]
38. S. K. Selvaraja, W. Bogaerts, D. Van Thourhout, and M. Schaekers, “Thermal trimming and tuning of hydrogenated amorphous silicon nanophotonic devices,” Appl. Phys. Lett. 97, 071120 (2010). [CrossRef]
39. T. Lipka, A. Harke, O. Horn, J. Amthor, and J. Müller, “Amorphous waveguides for high index photonic circuitry,” in Optical Fiber Communication Conference, OSA Technical Digest (CD) (OSA, 2009), paper OMJ2.
40. D. G. Cahill, M. Katiyar, and J. R. Abelson, “Thermal conductivity of a-Si:H thin films,” Phys. Rev. B 50(9), 6077–6081 (1994). [CrossRef]
41. M. Iodice, G. Mazzi, and L. Sirleto, “Thermo-optical static and dynamic analysis of a digital optical switch based on amorphous silicon waveguide,” Opt. Express 14, 5266–5278 (2006). [CrossRef] [PubMed]
42. S. Li, Y. Jiang, Z. Wu, J. Wu, Z. Ying, Z. Wang, W. Li, and G. J. Salamo, “Effect of structure variation on thermal conductivity of hydrogenated silicon film,” Appl. Surf. Sci. 257(20), 8326–8329 (2011). [CrossRef]
43. J F. Ready, “Effects of high-power laser radiation,” (Academic Press, New York, 1971).
44. L. Pauling, “The nature of the chemical bonding,” (Cornell Univ. Press, New York, 1982).
45. N. H. Nickel, K. Brendel, and R. Saleh, “Laser crystallization of hydrogenated amorphous silicon,” Phys. Stat. Sol.C 1(5), 1154–1168 (2004).
46. K. Shimakawa, A. Kolobov, and S. Elliott, “Photoinduced effects and metastability in amorphous semiconductors and insulators,” Adv. Phys. 44, 475–588 (1995). [CrossRef]
47. D. L. Staebler and C. R. Wronski, “Reversible conductivity changes in discharge-produced amorphous Si,” Appl. Phys. Lett. 31, 292 (1977). [CrossRef]
48. M. Stutzmann, W. B. Jackson, and C. C. Tsai, “Light-induced metastable defects in hydrogenated amorphous silicon: A systematic study,” Phys. Rev. B 32, 23–47 (1985). [CrossRef]
49. T. Shimizu, “Staebler-Wronski effect in hydrogenated amorphous silicon and related alloy films,” Jpn. J. Appl. Phys. 43(6A), 3257–3268 (2004). [CrossRef]
50. M. Fehr, A. Schnegg, B. Rech, O. Astakhov, F. Finger, R. Bittl, C. Teutloff, and K. Lips, “Metastable defect formation at microvoids identified as a source of light-induced degradation in a-Si:H,” Phys. Rev. Lett. 112, 066403 (2014). [CrossRef] [PubMed]
51. F. Gaspari, “Optoelectronic properties of amorphous silicon the role of hydrogen: from experiment to modeling,” in Optoelectronics - Materials and Techniques, P. Predeep, ed. (InTech, 2011), pp. 3–26. [CrossRef]