Mitigation of optical losses is of prime importance for the performance of integrated micro-photonic devices. In this paper, we demonstrate strip-loaded guiding optical components realized on a 27 nm ultra-thin silicon-on-insulator (SOI) platform. The absence of physically etched boundaries within the guiding core majorly suppresses the scattering loss, as shown by us previously for a silicon nitride () platform. Unexpectedly, the freshly fabricated Si devices showed large losses of 5.1 dB/cm originating from absorption by free carriers, accumulated under the positively charged loading layer. We show how ultraviolet (UV, 254 nm) light exposure can progressively and permanently neutralize ’s bulk charge, associated with diamagnetic defects. Consequently, the net decrease of electron concentration in the SOI layer reduces the propagation loss down to 0.9 dB/cm. Accurate cavity linewidth measurements demonstrate how the intrinsic cavity’s boosts from 70,000 up to 500,000 after UV illumination. Our results may open routes towards engineering of new functionalities in photonic devices employing UV modification of space-charge-associated local electric fields, unveil the origin of induced optical nonlinearities in micro-photonic systems, as well as envisage possible integration of these with both standard and ultra-thin SOI electronics.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Light confinement by resonant circulation of electromagnetic radiation in mirrorless microresonators makes them a key building block for planar integrated photonics [1,2]. The amount of optical power that is lost per round trip of circulation manifests in the spectral width of the resonances: the smaller the loss, the narrower the resonances become. In general, the possible channels of intrinsic losses in a cavity are the material absorption , the scattering due to boundary imperfections , and the out-radiation , due to the curved geometry. The overall intrinsic loss of a high-finesse cavity, combined with the extrinsic loss due to coupling to an external waveguide, thus defines the spectral width of a Lorentzian-shaped resonance peaked at a frequency of via
A reduced intrinsic loss is essential for a number of applications such as passive filtering in optical communication networks [3–5], quantum optics [6–10], space [11,12], or sensing [13–15]. Suppressing the intrinsic loss to only that of the material itself is challenging because it can push the device characteristics to an ultimate limit . Minute contributions from radiative loss can be achieved via a strong modal confinement within the guiding components (high-index contrast) and/or by using large radii of curvature [16–20]. Scattering loss can be suppressed by realization of smooth device boundaries during fabrication by either shallow etching or by completely excluding physically etched sidewalls [16–19,21–23]. In addition to the improvement of intrinsic losses , careful tailoring of the modal coupling to integrated waveguides must be performed to avoid excessive extrinsic losses . Single-mode operation, on one hand, can allow for omitting parasitic coupling to higher-order modes, while engineering of adiabatic couplers by weakly tapering the gap can reduce the modal mismatch in the coupler arms and lower the excess coupling loss .
In an optical device, the material loss is largely dependent on the choice of the operation frequency. In particular, most silicon micro-photonic devices operate at telecom frequencies, where the photon energy is below silicon’s electronic band gap [24,25]. The interband absorption, thus, is negligible, and the associated loss can be as low as 0.005 dB/cm for a typical -type Si of resistivity. On the other hand, a non-negligible intraband two-photon absorption (TPA) and excited-carrier absorption (ECA) can increase the loss by up to 3 orders of magnitude [26–28]. This last can be still mitigated by integrating p-n junction devices to operate under reverse-bias conditions, thus depleteing silicon from the electrical charge carriers in the guiding region .
Silicon nitride is widely used in integrated circuit technology , flat-panel displays , and solar cells . For the last decade, silicon nitride has attracted interest among the photonics community for CMOS-compatibile integrated photonics. It is a key dielectric material for linear micro-optical guiding circuits, transparent from visible to mid-infrared (MIR) wavelengths [4,32,33], as well as for applications in nonlinear frequency conversion schemes either due to its intrinsic nonlinearities [34–36] or induced nonlinearities when a silicon nitride film is applied to crystalline silicon [37–40]. Recent developments in the field of nonlinear optics have extended the potential of silicon nitride to be used in combination with lithium niobate within an integration-compatible process .
Here we report on the design, fabrication, and characterization of high- micro-optical components on an ultra-thin, 27 nm thick SOI platform, where light guiding is enabled by a patterned layer of strip-loading stoichiometric . The use of such a thin Si layer avoids the presence of guided slab modes where the loading nitride strip is absent. The absence of physically etched device boundaries in the SOI layer is expected to provide ultimately low losses, limiting them to that of the intrinsic absorption of the lightly doped -type silicon device layer. Such an approach was successfully implemented in our earlier study, where s of were achieved on a 80 nm thick platform . Surprisingly, the freshly fabricated silicon devices showed unexpectedly high propagation losses of up to 5.1 dB/cm. We related the origin of these losses to the absorption of free carriers within the Si layer, which are generated due to the presence of fixed positive charges in the deposited silicon nitride. Next, we successfully and permanently neutralized the charge in by exposing the devices to 254 nm wavelength UV light. This led to an improvement of losses down to 0.9 dB/cm, allowing us to boost the quality factors of the ring-resonator devices from an initial 70,000 up to 500,000.
Our results open the door to the implementation of UV-induced charge modification for the design and study of new photonic devices in which the space-charge-related static electric fields can be engineered to modulate the linear and nonlinear refractive indices of materials, for example, with patterned UV exposure. Moreover, our findings are general and may be implemented also in other geometries of guiding devices and material systems where silicon nitride is present, most notably standard SOI waveguides. We also envisage the possibility of compact integration of micro-photonic components with ultra-thin SOI electronics in the future.
2. MATERIALS AND METHODS
A. Device Fabrication
The samples were realized starting from 6” SOI wafers with a 3 μm thick buried oxide (BOX) and a 250 nm (100)-Si device layer (Soitec). The latter is lightly doped with boron and has a nominal resistivity of . First, the device layer thickness was reduced by thermally oxidizing the Si and removing the grown oxide in a buffered hydrofluoric acid solution. A fine tuning of the final thickness, , was achieved by means of standard RCA cleaning steps . Next, a 145 nm thick low-pressure chemical vapor deposition film was deposited at 780°C and patterned lithographically using an i-line Nikon stepper. The pattern was then transferred to using an inductively coupled plasma etch, terminating with a wet etching step for the last 10 nm to guarantee a smooth top surface of the underlying Si layer. Finally, the fabricated chips were diced using a polishing-grade saw to define the waveguides’ input-output facets. The cross-sectional view of a generic device and the distribution of the simulated electric field intensity of a transverse electric (TE)-polarized mode are shown in Fig. 1(a). Optical micrographs of a typical ring resonator and a spiral waveguide are shown in Fig. 1(b) and Fig. 1(c), respectively.
B. Optical Characterization
The devices were tested in waveguide transmission experiments in a 100 nm range around 1550 nm of wavelength. A tunable laser (Yenista Tunics T100S HP) was butt-coupled to the waveguides using a lensed fiber (total insertion loss of ). The signal polarization was controlled at the waveguide input, while the transmitted signal was collected with a second fiber at the waveguide output. The signal was then sent to an InGaAs photodiode and analyzed using a high-resolution oscilloscope (PicoScope 4224).
C. MOS Capacitance Measurements
Quasi-static capacitance-voltage (C-V) measurements were conducted at 10 kHz using an MDC 802-150 Mercury Probe (spot diameter 787 μm) and acquired with an Agilent 4156C Parameter Analyzer. Different samples, cut from the same wafer, were exposed to 254 nm UV light from a Hg bulb lamp (254 nm, ). The exposure times were varied from 1 min up to 32 h. As a reference, we also measured an as-deposited sample with no UV exposure. In order to acquire enough statistics per sample, the C-V measurements were repeated on at least five different points on each sample.
Based on the results from C-V experiments, several chips containing micro-photonic components such as centimeter-long spirals and ring resonators were exposed to UV light in order to directly study the variation of optical characteristics of these devices due to electrical charge neutralization.
3. RESULTS AND DISCUSSION
A. Diamagnetic and Paramagnetic Centers in Silicon Nitride
Silicon nitride, both in its stoichiometric () and -rich form (), is known to host charge trapping centers via silicon dangling bond defects. Such defects are typically present and homogeneously distributed in the film, either in a neutrally charged paramagnetic state, known as the -centers (), or in a charged diamagnetic state . These last can be positively (, ) or negatively charged () [43–47]. Interestingly, exposing films to UV radiation with energies greater than 3.5 eV () can induce a change in the spin and charge state of diamagnetic states , increasing the concentration of neutral -centers and, therefore, leading to a partial or complete compensation of space charge in the film . The initial spin/charge state may vary depending on film composition and growth technique, and the exact mechanism of charge neutralization is not fully understood yet. However, it has been proved unambiguously that the UV radiation generally causes a modification of initial charging conditions [43–48]. From the point of view of device functionality, UV-induced charge reduction can largely modify the electrical characteristics, for example, via a redistribution of space charge in the semiconductor material in contact with the film.
Capacitance measurements of a dielectric film in a two-terminal metal-oxide-semiconductor (MOS) configuration offer a wealth of information on the fabrication process, in particular, the nature, sign, and amount of electrical charge in the bulk of the dielectric film and at the interface with the semiconductor substrate [Fig. 2(a)]. In order to reveal and quantify possible charges within the silicon nitride, we have deposited a 145 nm thick film on top of Si substrates and performed C-V measurements in a MOS configuration. For this, boron-doped -type substrates with resistivity, matching with that of the device layer of our SOI wafers, were chosen. In addition, prior to deposition, a thin 5 nm layer was grown during a standard RCA cleaning step. Thus, the test wafers exactly replicate the strip-loaded SOI devices.
We found that the as-deposited layer contains a large amount of net positive electrical charge, as already reported for films deposited using other techniques [39,40]. The typical C-V curve for this reference sample [Fig. 2(b), circles] passes from an “accumulation” to a “depletion” state at negative gate voltages, with a characteristic flat-band voltage value of . This last is estimated graphically once the flat-band capacitance is calculated, following39].
Capacitance measurements on UV-exposed samples show a gradual shift of the C-V curves towards positive voltages as a function of exposure time. Repeated measurements at a distance of a couple of weeks showed that this shift is permanent. Selected examples of C-V curves are shown in Fig. 2(b), while in Fig. 2(c) we show the dependence of the extracted flat-band voltage on UV illumination time. We notice that decreases exponentially within the first couple of hours of exposure. For the next 30 h of exposure, continues to shift monotonically at a much slower rate (almost linearly) towards the asymptotic value of metal-semiconductor work function , which for our -type Si system amounts to .
The corresponding variation of charge areal density against is shown in Fig. 2(d). The UV illumination decreases the positive charge density by 3 orders of magnitude, leading to a change of the flat-band voltage from its initial value down to for the longest exposures. The estimated residual charge density amounts to , which is comparable to the density of charge traps of very-high-quality oxide/Si interfaces, which is also close to the density of dopant ions per centimeter squared () for a silicon layer of resistivity .
B. UV Exposure Effect on Optical Losses in Micro-Photonic Devices
Following the encouraging results obtained from MOS capacitance measurements, we studied the evolution of optical losses of -loaded SOI devices in response to UV light illumination. For this reason, several chips containing both spiral waveguides of different length and ring resonators were characterized in optical transmission experiments prior to and after UV exposure.
1. Loss Characterization from Waveguides
The studied waveguides were composed of 27 nm thick continuous SOI slab and a 145 nm thick loading strip of 1300 nm width. Finite-element numerical simulations, performed in the design phase, suggest that this geometry does not guide the transverse magnetic (TM) polarization and supports one single TE mode in the wavelength range, while the Si slab alone does not guide light (see Supplement 1). The typical values for the mode effective refractive index , the group index , and the effective mode area are 1.588, 2.08, and , respectively.
Figure 3 reports the results of waveguide transmission experiments. The measured waveguides had lengths of 0.61, 3.22, and 6.17 cm. After conversion to the decibel per centimeter (dB/cm) scale, the propagation losses in the as-prepared devices (red squares) amount to according to the Beer–Lambert law , and, if attributed to sidewall scattering, are unexpectedly high for the considered strip-loaded configuration and the adopted processing technology . As anticipated in Section A, we expect that the net positive charge in the loading layer recalls negative charge carriers (mirror charges) in the underlying Si core, thus increasing the free-carrier absorption within the waveguide . In fact, the same devices show much improved characteristics after 21 h of UV exposure (blue diamonds), and the resulting propagation loss is decreased down to .
According to Figs. 2(c) and 2(d), the areal charge density is estimated to decrease from in the as-prepared samples down to after 21 h of UV exposure. From total charge balance conditions, the equivalent free-electron density in the Si core would be reduced from a value of to . In a first-order approximation, it is possible to relate the change in to the absorption loss within a crystalline Si core by assuming that:
- • all losses are due to free-carrier-induced absorption and no scattering losses are present,
- • no other sources of charge are present in the vicinity of the mode (e.g., charges at the Si/BOX interface),
- • the effective loss coefficient is reduced from its bulk value proportional to the power confinement factor of the mode within the Si layer ( from finite-elements method (FEM) calculations at ).
We calculated the expected using different empirical models known from literature [51–53]. Such values range from 2 to 6 dB/cm for as-prepared samples and from 0.15 to 0.25 dB/cm after 21 h of UV exposure. It is clear that while our estimation of for the first case is well within the theoretical range, the UV-exposed case with is larger by more than 1 dB/cm with respect to expectations.
Our conclusion at this point is that either an additional loss source is present in the studied devices, or the Beer–Lambert approach is not precise enough in the conditions where few experimental data are available and the insertion losses from one to another waveguide differ due to facet imperfections. This last is in fact reflected by the large error bars of experimental points in Fig. 3, given by the variance over three different chips. This limitation can be surpassed by extracting the loss from spectral characteristics of ring resonators , which are less affected by fluctuations of the waveguide facet quality.
2. Loss Characterization from Resonators
A generic circular resonator induces spectral dips in the transmission spectrum of the waveguide to which it is coupled. These dips become visible as soon as the resonator’s intrinsic loss and the coupling loss become comparable [see Eq. (1)]. In particular, for a fixed geometry, i.e., a constant , the lossy () resonator’s spectral dips start to spot out from the bare waveguide’s transmission spectrum while lowers towards .
Figure 4(a) reports examples of spectra calculated by the following:54]. Here, we have introduced doublet resonances to account for formation of symmetric, , and anti-symmetric, , traveling waves within the resonator due to possible backscattering mechanisms . These last are described as , where is the frequency detuning from the resonance and is the backscattering rate (fixed in these examples). Equation (3) has been considered since this is the situation of our experiments with rings when the UV improvement results in narrower resonances, as will be discussed in the following.
We notice that when , no resonant features can be observed out from the oscillating FP background [dashed line, Fig. 4(a)]. A resonant dip starts to appear when is decreased, first appearing as an “unstructured bump” and transforming progressively into a brighter and more well-defined doublet when .
Experimentally, we have realized on the same chip nominally identical ring resonators (radius of 60 μm and width of 1300 nm) coupled to the waveguides through three different gaps of , 1620 nm, and 1800 nm. In particular, the largest gap provides an external coupling . This situation was considered to fulfill a critical coupling, , based on the numerical simulations for lowest possible intrinsic loss of the resonators (bend-loss limited), without accounting for other residual losses. Figure 4(b) shows examples of typical spectra for as-prepared devices at two different coupling gaps. Namely, in the case of , we observe a featureless spectrum of the typical waveguide transmission, which means that the resonator’s intrinsic loss is much higher than that of the coupling. In other words, we deal with the strongly undercoupled situation. In fact, for devices with , we observe critically coupled resonances, from which we extract , corresponding to an intrinsic loss of about (see Supplement 1).
Panels (c), (d), and (e) of Fig. 4 show the evolution of the ring spectrum upon exposure to UV at various times. All the results are for a coupling gap of 1800 nm. We note that while devices with an intermediate gap () provide deeper resonances, they hide important spectral features of the doublets due to larger coupling loss (). Therefore, we concentrate on analyzing clear doublets that manifest in devices with the highest . The spectral form of the doublets and their peak transmission are more sensible to the intrinsic loss value; therefore, they permit extraction of with higher accuracy with respect to the Beer–Lambert method, described previously. We notice that a 5 h UV treatment improves the loss to 2.72 dB/cm [Fig. 4(c)] and, further, down to after 23 h [panel (d)].
Contrary to expectations from the results of capacitance measurements, we did not observe further improvements of losses upon exposing devices to UV for times longer than 21 h. Our conclusion, at this point, is that in the case of our devices we either deal with residual scattering losses due to fabrication or the SOI structure provides other charge-related losses that cannot be improved with UV exposure. These last can originate from positive charges situated at the Si/BOX interface or in the BOX oxide itself. For this, we performed additional sintering of the chips and of a UV-improved (23 h) MOS test sample at 350°C for 3 h in forming gas to improve the interface. Control MOS capacitance measurements showed that some positive charge in the was re-activated, which rapidly annihilated upon a post-sintering exposure to UV for 2 h. We repeated the same procedure on the device chip and measured the rings’ spectra. Figure 4(e) shows an example spectrum of resonances, where the loss has been further improved down to 0.91 dB/cm. These results may suggest that a certain amount of charge was possibly present in the SOI device and was partially neutralized after the sintering procedure. Further UV exposures did not improve the situation, from which we conclude that the main part of remaining loss has other origins, such as scattering.
At this point, it is possible to explicitly relate the variation of the intrinsic -factor of the ring resonators to the flat-band voltage , extracted from MOS capacitance measurements. The mirror-charge-induced absorption can be estimated using the empirical formula for -type Si , where is the free carriers’ concentration. Finally, the intrinsic can be related to via
In Fig. 5(a) we show the calculated trend (continuous blue line) of the intrinsic against the flat-band voltage calculated using Eq. (4a) assuming a constant of 195 pF. The variation of reflects the effect of UV exposure, and, thus, the gradual neutralization of the positive charge in the layer according to Fig. 2(c). We also plot as open circles (°) the results of Eq. (4a) using the values of and , including their corresponding experimental error bars, extracted experimentally in Section 3.A. It is worth noting that the -factor for the first point of these data at has been calculated considering an effectively -type Si since the MOS experiments showed an inversion of conductivity type only for the UV-untreated samples [Fig. 2(b)]. In this case, the estimated amounts to about .
The -factors estimated from optical measurements on ring resonators are shown as red diamonds in Fig. 5(a). Each point is the result of the analysis of at least ten different resonances. The first point of this dataset, corresponding to the UV-untreated devices, shows an average in accordance with the corresponding MOS capacitance experiment. Finally, the rest of the data from the ring resonators, corresponding to long exposures to UV light, can be fit to good approximation using Eq. (4a) by including an additional residual loss with an associated (red dashed-dotted curve). The corresponding loss values, based on the results from the ring resonators, and the calculated curve for -type Si in the situation of residual loss are shown in Fig. 5(b).
These results show the advantage of the loss-estimation method from resonant features with respect to a classical Beer–Lambert approach. The estimated residual loss , which we attributed to scattering, perhaps may still contain a component associated with charging effects at the interface or within the bulk of the BOX oxide. At present, this aspect is uncertain and is a subject of further investigations.
We reported in this work the design, fabrication, and characterization of high-Q micro-optical components on an ultra-thin SOI platform. Waveguiding is supported by loading a 27 nm thick SOI slab layer with micrometer-wide stripes of stoichiometric . Such a configuration omits the need to etch physical boundaries in the SOI layer and, therefore, largely suppresses the scattering loss .
Contrary to expectations, we revealed that the as-prepared devices were subject to significant loss of about 5.1 dB/cm. We related this to free-carrier absorption effects due to the presence of a large amount of electrical charge in the layer, originating from paramagnetic defects (dangling bonds). This hypothesis was, in a first step, confirmed by detailed MOS capacitance measurements, revealing a complete inversion of the conductivity type of the -type Si substrate. As a result, the resistivity of the original substrate was reduced from down to , confirming the observed large optical losses. Next, we exposed test samples to 254 nm UV light for different times and observed gradual neutralization of the space charge in the nitride layer. Our estimations from these experiments showed that the equivalent areal electrical charge can be reduced by 3 orders of magnitude, reaching a value of , comparable to the density of charge traps of very-high-quality oxide/Si interfaces .
Then, UV exposure was performed on SOI devices, and the optical loss was directly measured from spiral waveguides by the Beer–Lambert approach and ring resonators from the spectral linewidth of resonances. We revealed a net improvement of losses down to 0.9 dB/cm at the longest exposures, improving the rings’ intrinsic -factors from 70,000 to 500,000. Finally, we explicitly related the carrier-induced optical losses to the MOS capacitance measurements of the flat-band voltage.
We foresee that these results will go far beyond the target of the current study. The permanent nature of the UV neutralization of the charges in can have important implications for the design of micro-photonic devices. For example, charged domains of material can be alternated to create static electrical poling on top of waveguiding components by using appropriate masking during UV illumination. The sign and the amount of bulk charges are specific to various deposited by different techniques; therefore, their modulation with UV light can be further studied in view of photonic applications. It is expected that at such charging levels of , the observed phenomena, such as the inversion of the conductivity type of the silicon layer and the large carrier-induced losses, may also affect the more conventional 220 nm thick SOI circuits. Last but not least, it appears promising to implement the UV exposure procedure in current hot topics such as the studies of the origin of dressed nonlinearities in nitride-strained silicon waveguides [38,39]. Finally, our results may open the door to future developments of compactly integrated micro-photonic components with electronics on the same ultra-thin SOI platform.
Ministero dell’Istruzione, dell’Università e della Ricerca (MIUR) (2015KEZNYM).
The authors gratefully thank Georg Pucker for support and fruitful discussions and Lorenzo Pavesi for providing access to the optical measurement facility of the NanoScience Laboratory at the University of Trento. The authors also acknowledge fabrication facility support by the Micro-Nano Fabrication Laboratory of FBK.
See Supplement 1 for supporting content.
1. D. G. Rabus, Integrated Ring Resonators (Springer, 2007).
2. J. Heebner, R. Grover, T. Ibrahim, and T. A. Ibrahim, Optical Microresonators: Theory, Fabrication, and Applications (Springer, 2008).
3. D. Dai, J. Bauters, and J. E. Bowers, “Passive technologies for future large-scale photonic integrated circuits on silicon: polarization handling, light non-reciprocity and loss reduction,” Light Sci. Appl. 1, e1 (2012). [CrossRef]
4. A. Subramanian, P. Neutens, A. Dhakal, R. Jansen, T. Claes, X. Rottenberg, F. Peyskens, S. Selvaraja, P. Helin, B. Du Bois, K. Leyssens, S. Severi, P. Deshpande, R. Baets, and P. Van Dorpe, “Low-loss singlemode PECVD silicon nitride photonic wire waveguides for 532–900 nm wavelength window fabricated within a CMOS pilot line,” IEEE Photon. J. 5, 2202809 (2013). [CrossRef]
5. L. Zhuang, D. Marpaung, M. Burla, W. Beeker, A. Leinse, and C. Roeloffzen, “Low-loss, high-index-contrast Si3N4/SiO2 optical waveguides for optical delay lines in microwave photonics signal processing,” Opt. Express 19, 23162–23170 (2011). [CrossRef]
6. E. Engin, D. Bonneau, C. M. Natarajan, A. S. Clark, M. Tanner, R. Hadfield, S. N. Dorenbos, V. Zwiller, K. Ohira, N. Suzuki, H. Yoshida, N. Iizuka, M. Ezaki, J. L. O’Brien, and M. G. Thompson, “Photon pair generation in a silicon micro-ring resonator with reverse bias enhancement,” Opt. Express 21, 27826–27834 (2013). [CrossRef]
7. D. Grassani, S. Azzini, M. Liscidini, M. Galli, M. J. Strain, M. Sorel, J. Sipe, and D. Bajoni, “Micrometer-scale integrated silicon source of time-energy entangled photons,” Optica 2, 88–94 (2015). [CrossRef]
8. J. W. Silverstone, R. Santagati, D. Bonneau, M. J. Strain, M. Sorel, J. L. O’Brien, and M. G. Thompson, “Qubit entanglement between ring-resonator photon-pair sources on a silicon chip,” Nat. Commun. 6, 7948 (2015). [CrossRef]
9. J. W. Silverstone, D. Bonneau, J. L. O’Brien, and M. G. Thompson, “Silicon quantum photonics,” IEEE J. Sel. Top. Quantum Electron. 22, 390–402 (2016). [CrossRef]
10. J. B. Christensen, J. G. Koefoed, K. Rottwitt, and C. McKinstrie, “Engineering spectrally unentangled photon pairs from nonlinear microring resonators by pump manipulation,” Opt. Lett. 43, 859–862 (2018).
11. T. J. Kippenberg, R. Holzwarth, and S. A. Diddams, “Microresonator-based optical frequency combs,” Science 332, 555–559 (2011). [CrossRef]
12. A. Pasquazi, L. Caspani, M. Peccianti, M. Clerici, M. Ferrera, L. Razzari, D. Duchesne, B. E. Little, S. T. Chu, D. J. Moss, and R. Morandotti, “Self-locked optical parametric oscillation in a CMOS compatible microring resonator: a route to robust optical frequency comb generation on a chip,” Opt. Express 21, 13333–13341 (2013). [CrossRef]
13. A. M. Armani and K. J. Vahala, “Heavy water detection using ultra-high-Q microcavities,” Opt. Lett. 31, 1896–1898 (2006). [CrossRef]
14. J. Zhu, S. K. Ozdemir, Y.-F. Xiao, L. Li, L. He, D.-R. Chen, and L. Yang, “On-chip single nanoparticle detection and sizing by mode splitting in an ultrahigh-Q microresonator,” Nat. Photonics 4, 46–49 (2010). [CrossRef]
15. A. Samusenko, D. Gandolfi, G. Pucker, T. Chalyan, R. Guider, M. Ghulinyan, and L. Pavesi, “A SION microring resonator-based platform for biosensing at 850 nm,” J. Lightwave. Technol. 34, 969–977 (2016). [CrossRef]
16. H. Lee, T. Chen, J. Li, K. Y. Yang, S. Jeon, O. Painter, and K. J. Vahala, “Chemically etched ultrahigh-Q wedge-resonator on a silicon chip,” Nat. Photonics 6, 369–373 (2012). [CrossRef]
17. D. T. Spencer, J. F. Bauters, M. J. Heck, and J. E. Bowers, “Integrated waveguide coupled Si3N4 resonators in the ultrahigh-Q regime,” Optica 1, 153–157 (2014). [CrossRef]
18. L. Stefan, M. Bernard, R. Guider, G. Pucker, L. Pavesi, and M. Ghulinyan, “Ultra-high-Q thin-silicon nitride strip-loaded ring resonators,” Opt. Lett. 40, 3316–3319 (2015). [CrossRef]
19. M. P. Nezhad, O. Bondarenko, M. Khajavikhan, A. Simic, and Y. Fainman, “Etch-free low loss silicon waveguides using hydrogen silsesquioxane oxidation masks,” Opt. Express 19, 18827–18832 (2011). [CrossRef]
20. X. Ji, F. A. Barbosa, S. P. Roberts, A. Dutt, J. Cardenas, Y. Okawachi, A. Bryant, A. L. Gaeta, and M. Lipson, “Ultra-low-loss on-chip resonators with sub-milliwatt parametric oscillation threshold,” Optica 4, 619–624 (2017). [CrossRef]
21. F. Ramiro-Manzano, N. Prtljaga, L. Pavesi, G. Pucker, and M. Ghulinyan, “A fully integrated high-Q whispering-gallery wedge resonator,” Opt. Express 20, 22934–22942 (2012). [CrossRef]
22. L.-W. Luo, G. S. Wiederhecker, J. Cardenas, C. Poitras, and M. Lipson, “High quality factor etchless silicon photonic ring resonators,” Opt. Express 19, 6284–6289 (2011). [CrossRef]
23. A. Biberman, M. J. Shaw, E. Timurdogan, J. B. Wright, and M. R. Watts, “Ultralow-loss silicon ring resonators,” Opt. Lett. 37, 4236–4238 (2012). [CrossRef]
24. R. Soref, “The past, present, and future of silicon photonics,” IEEE J. Sel. Top. Quantum Electron. 12, 1678–1687 (2006). [CrossRef]
25. L. Vivien and L. Pavesi, Handbook of Silicon Photonics (Taylor & Francis, 2016).
26. T. Liang and H. Tsang, “Role of free carriers from two-photon absorption in Raman amplification in silicon-on-insulator waveguides,” Appl. Phys. Lett. 84, 2745–2747 (2004). [CrossRef]
27. G. Priem, P. Dumon, W. Bogaerts, D. Van Thourhout, G. Morthier, and R. Baets, “Optical bistability and pulsating behaviour in silicon-on-insulator ring resonator structures,” Opt. Express 13, 9623–9628 (2005). [CrossRef]
28. J. Leuthold, C. Koos, and W. Freude, “Nonlinear silicon photonics,” Nat. Photonics 4, 535–544 (2010). [CrossRef]
29. J. Milek, Silicon Nitride for Microelectronic Applications: Part 1 Preparation and Properties (Springer, 2013).
30. J. Kanicki, Amorphous and Microcrystalline Semiconductor Devices: Materials and Device Physics (Artech House, 1992), Vol. 2.
31. R. E. Schropp and M. Zeman, Amorphous and Microcrystalline Silicon Solar Cells: Modeling, Materials and Device Technology (Springer, 1998).
32. A. Gorin, A. Jaouad, E. Grondin, V. Aimez, and P. Charette, “Fabrication of silicon nitride waveguides for visible-light using PECVD: a study of the effect of plasma frequency on optical properties,” Opt. Express 16, 13509–13516 (2008). [CrossRef]
33. S. Romero-Garca, F. Merget, F. Zhong, H. Finkelstein, and J. Witzens, “Silicon nitride CMOS-compatible platform for integrated photonics applications at visible wavelengths,” Opt. Express 21, 14036–14046 (2013). [CrossRef]
34. J. S. Levy, M. A. Foster, A. L. Gaeta, and M. Lipson, “Harmonic generation in silicon nitride ring resonators,” Opt. Express 19, 11415–11421 (2011). [CrossRef]
35. Y. Okawachi, K. Saha, J. S. Levy, Y. H. Wen, M. Lipson, and A. L. Gaeta, “Octave-spanning frequency comb generation in a silicon nitride chip,” Opt. Lett. 36, 3398–3400 (2011). [CrossRef]
36. D. J. Moss, R. Morandotti, A. L. Gaeta, and M. Lipson, “New CMOS-compatible platforms based on silicon nitride and hydex for nonlinear optics,” Nat. Photonics 7, 597–607 (2013). [CrossRef]
37. R. S. Jacobsen, K. N. Andersen, P. I. Borel, J. Fage-Pedersen, L. H. Frandsen, O. Hansen, M. Kristensen, A. V. Lavrinenko, G. Moulin, H. Ou, C. Peucheret, B. Zsigri, and A. Bjarklev, “Strained silicon as a new electro-optic material,” Nature 441, 199–202 (2006). [CrossRef]
38. M. Cazzanelli, F. Bianco, E. Borga, G. Pucker, M. Ghulinyan, E. Degoli, E. Luppi, V. Véniard, S. Ossicini, D. Modotto, S. Wabnitz, R. Pierobon, and L. Pavesi, “Second-harmonic generation in silicon waveguides strained by silicon nitride,” Nat. Mater. 11, 148–154 (2012). [CrossRef]
39. C. Schriever, F. Bianco, M. Cazzanelli, M. Ghulinyan, C. Eisenschmidt, J. de Boor, A. Schmid, J. Heitmann, L. Pavesi, and J. Schilling, “Second-order optical nonlinearity in silicon waveguides: inhomogeneous stress and interfaces,” Adv. Opt. Mater. 3, 129–136 (2015). [CrossRef]
40. S. S. Azadeh, F. Merget, M. Nezhad, and J. Witzens, “On the measurement of the Pockels effect in strained silicon,” Opt. Lett. 40, 1877–1880 (2015). [CrossRef]
41. L. Chang, Y. Li, N. Volet, L. Wang, J. Peters, and J. E. Bowers, “Thin film wavelength converters for photonic integrated circuits,” Optica 3, 531–535 (2016). [CrossRef]
42. W. Kern and D. A. Puotinen, “Cleaning solutions based on hydrogen peroxide for use in silicon semiconductor technology,” RCA Rev. 31, 187–206 (1970).
43. D. Krick, P. Lenahan, and J. Kanicki, “Electrically active point defects in amorphous silicon nitride: an illumination and charge injection study,” J. Appl. Phys. 64, 3558–3563 (1988). [CrossRef]
44. M. Kumeda, H. Yokomichi, and T. Shimizu, “Photo-induced ESR in amorphous Si1-xNx: H films,” Jpn. J. Appl. Phys. 23, L502–L504 (1984). [CrossRef]
45. W. L. Warren, P. Lenahan, and S. E. Curry, “First observation of paramagnetic nitrogen dangling-bond centers in silicon nitride,” Phys. Rev. Lett. 65, 207–210 (1990). [CrossRef]
46. W. Warren, P. Lenahan, and J. Kanicki, “Electrically neutral nitrogen dangling-bond defects in amorphous hydrogenated silicon nitride thin films,” J. Appl. Phys. 70, 2220–2225 (1991). [CrossRef]
47. K. Kobayashi and K. Ishikawa, “Ultraviolet light-induced conduction current in silicon nitride films,” Jpn. J. Appl. Phys. 50, 031501 (2011). [CrossRef]
48. W. Warren, J. Kanicki, J. Robertson, E. Poindexter, and P. McWhorter, “Electron paramagnetic resonance investigation of charge trapping centers in amorphous silicon nitride films,” J. Appl. Phys. 74, 4034–4046 (1993). [CrossRef]
49. P. Zhang, E. Tevaarwerk, B.-N. Park, D. E. Savage, G. K. Celler, I. Knezevic, P. G. Evans, M. A. Eriksson, and M. G. Lagally, “Electronic transport in nanometre-scale silicon-on-insulator membranes,” Nature 439, 703–706 (2006). [CrossRef]
50. A. Beer, “Determination of the absorption of red light in colored liquids,” Ann. Phys. Chem. 86, 78–88 (1852). [CrossRef]
51. R. Soref and B. Bennett, “Electrooptical effects in silicon,” IEEE J. Quantum Electron. 23, 123–129 (1987). [CrossRef]
52. J. Degallaix, R. Flaminio, D. Forest, M. Granata, C. Michel, L. Pinard, T. Bertrand, and G. Cagnoli, “Bulk optical absorption of high resistivity silicon at 1550 nm,” Opt. Lett. 38, 2047–2049 (2013). [CrossRef]
53. M. Nedeljkovic, R. Soref, and G. Z. Mashanovich, “Free-carrier electrorefraction and electroabsorption modulation predictions for silicon over the 1–14 μm infrared wavelength range,” IEEE Photon. J. 3, 1171–1180 (2011). [CrossRef]
54. S. Taebi, M. Khorasaninejad, and S. S. Saini, “Modified Fabry–Perot interferometric method for waveguide loss measurement,” Appl. Opt. 47, 6625–6630 (2008). [CrossRef]
55. M. Borselli, T. J. Johnson, and O. Painter, “Beyond the Rayleigh scattering limit in high-Q silicon microdisks: theory and experiment,” Opt. Express 13, 1515–1530 (2005). [CrossRef]