High third-order optical nonlinear performance in CMOS devices integrated with 2D graphene oxide lms


 We report enhanced nonlinear optics in complementary metal-oxide-semiconductor (CMOS) compatible photonic platforms through the use of layered two-dimensional (2D) graphene oxide (GO) films. We integrate GO films with silicon-on-insulator nanowires (SOI), high index doped silica glass (Hydex) and silicon nitride (SiN) waveguides and ring resonators, to demonstrate an enhanced optical nonlinearity including Kerr nonlinearity and four-wave mixing (FWM). The GO films are integrated using a large-area, transfer-free, layer-by-layer method while the film placement and size are controlled by photolithography. In SOI nanowires we observe a dramatic enhancement in both the Kerr nonlinearity and nonlinear figure of merit (FOM) due to the highly nonlinear GO films. Self-phase modulation (SPM) measurements show significant spectral broadening enhancement for SOI nanowires coated with patterned films of GO. The dependence of GO’s Kerr nonlinearity on layer number and pulse energy shows trends of the layered GO films from 2D to quasi bulk-like behavior. The nonlinear parameter of GO coated SOI nanowires is increased 16 folds, with the nonlinear FOM increasing over 20 times to FOM > 5. We also observe an improved FWM efficiency in SiN waveguides integrated with 2D layered GO films. FWM measurements for samples with different numbers of GO layers and at different pump powers are performed, achieving up to ≈ 7.3 dB conversion efficiency (CE) enhancement for a uniformly coated device with 1 layer of GO and ≈ 9.1 dB for a patterned device with 5 layers of GO. These results reveal the strong potential of GO films to improve the nonlinear optics of silicon, Hydex and SiN photonic devices.


Introduction
All-optical integrated photonic devices are of signi cant interest for high-speed signal generation and processing in optical communication systems, due to the fact that they don't need the complex and ine cient optical-electrical-optical conversion [1,2]. Triggered by a signi cant number of applications in telecommunications [3], metrology [4], astronomy [5], ultrafast optics [6], quantum photonics [7], and many other areas [8-10], high-performance platforms for integrated nonlinear optics has attracted much attention, and no doubt silicon-on-insulator (SOI) has led this eld for several years [11][12][13][14][15].
While SOI has shown itself to be a leading platform for integrated photonic devices, it suffers from strong two-photon absorption (TPA) at near-infrared wavelengths, which greatly limits the nonlinear performance [2,16], and this has motivated the use of highly nonlinear materials on chips. Other complementary metal-oxide-semiconductor (CMOS) compatible platforms including high index doped silica glass (Hydex) [17,18] and silicon nitride (SiN) [19,20] have a much lower TPA, but they hamper the nonlinear performance due to a comparatively low Kerr nonlinearity. To overcome these limitations, twodimensional (2D) layered graphene oxide (GO) has received much attention among the various 2D materials due to its ease of preparation as well as the tunability of its material properties [21][22][23][24][25][26][27][28][29]. Previously, we reported GO lms with a giant Kerr nonlinear response about 4-5 orders of magnitude higher than that of silicon and SiN [25] and demonstrated enhanced four-wave mixing (FWM) in doped silica waveguides and microring resonators (MRRs) integrated with GO lms [30,31]. Here, we demonstrate enhanced nonlinear optics in SOI nanowires [32] and SiN waveguides [33] integrated with 2D layered GO lms. Owing to the strong light-matter interaction between the integrated waveguides and the highly nonlinear GO lms, self-phase modulation (SPM) measurements are performed to show signi cant spectral broadening enhancement for SOI nanowires coated with patterned lms of GO. The dependence of GO's Kerr nonlinearity on layer number and pulse energy shows interesting physical insights and trends of the layered GO lms in evolving from 2D monolayers to quasi bulk-like behavior. We obtain signi cant enhanced nonlinear performance for the GO hybrid devices as compared with the bare waveguides, achieving the nonlinear parameter of GO-coated SOI nanowires up by 16 times, with the nonlinear gure of merit (FOM) increasing over 20 times to FOM > 5. We obtain a signi cant improvement in the FWM conversion e ciency (CE) of ≈ 7.3 dB for a uniformly coated SiN waveguide with 1 layer of GO and ≈ 9.1 dB for a patterned device with 5 layers of GO. These results con rm the strong potential of introducing 2D layered GO lms into CMOS compatible photonic platforms to realize high-performance nonlinear photonic devices. Figure 1a shows a schematic of an SOI nanowire waveguide integrated with a GO lm. The SOI nanowire was fabricated on an SOI wafer via CMOS compatible fabrication processes, with opened windows on the silica upper cladding so as to enable GO lm coating onto the SOI nanowire. The coating of 2D layered GO lms was achieved by a solution-based method that yielded layer-by-layer GO lm deposition. Our GO coating method can achieve precise control of the lm thickness with an ultrahigh resolution of 2 nm, which is challenging for spin coating methods. Further, our GO coating approach, unlike the sophisticated transfer processes (e.g., using scotch tape) employed for coating other 2D materials such as graphene and TMDCs [34,35], enables transfer-free GO lm coating on integrated photonic devices, with highly scalable fabrication processes as well as high fabrication stability and repeatability. Figure 1b shows a microscope image of a fabricated SOI chip with a 0.4-mm-long opened window. Apart from allowing precise control of the placement and coating length of the GO lms that are in contact with the SOI nanowires, the opened windows also enabled us to test the performance of devices having a shorter length of GO lm but with higher lm thicknesses (up to 20 layers). This provided more exibility to optimize the device performance with respect to SPM spectral broadening. Figure 1c shows the scanning electron microscopy (SEM) image of an SOI nanowire conformally coated with 1 layer of GO.

Enhanced Kerr Nonlinearity In Go-coated Soi Nanowires
Note that the conformal coating (with the GO lm coated on both the top surface and sidewalls of the nanowire) is slightly different to earlier work where we reported doped silica devices with GO lms only coated on the top surface of the waveguides [30,31]. As compared with doped silica waveguides, the SOI nanowires allow much stronger light-material interaction between the evanescent eld leaking from the waveguide and the GO lm, which is critical to enhance nonlinear optical processes such as SPM and FWM. Figure 1d shows the successful integration of GO lms which is con rmed by the representative D (1345 cm -1 ) and G (1590 cm -1 ) peaks of GO observed in the Raman spectrum of an SOI chip coated with 5 layers of GO. Microscope images of the same SOI chip before and after GO coating are shown in the insets, which illustrate good morphology of the lms.  Figure 2a-i shows the normalized spectra of the optical pulses before and after transmission through the SOI nanowires with 2.2-mm-long, 1−3 layers of GO, together with the output optical spectrum for the bare SOI nanowire, all taken with the same pulse energy of ~51.5 pJ (i.e., ~13.2 W peak power, excluding coupling loss) coupled into the SOI nanowires. As compared with the input pulse spectrum, the output spectrum after transmission through the bare SOI nanowire exhibited measurable spectral broadening. This is expected and can be attributed to the high Kerr nonlinearity of silicon. The GO-coated SOI nanowires, on the other hand, show much more signi cantly broadened spectra as compared with the bare SOI nanowire, clearly re ecting the improved Kerr nonlinearity of the hybrid waveguides. Figure 2a-ii shows the corresponding results for the SOI nanowires with 0.4-mm-long, 5−20 layers of GO, taken with the same coupled pulse energy as in Figure  2a-i. The SOI nanowires with a shorter GO coating length but higher lm thicknesses also clearly show more signi cant spectral broadening as compared with the bare SOI nanowire. We also note that in either Figure 2a-i or 2a-ii, the maximum spectral broadening is achieved for a device with an intermediate number of GO layers (i.e., 2 and 10 layers of GO in a-i and a-ii, respectively). This could re ect the tradeoff between the Kerr nonlinearity enhancement (which dominates for the device with a relatively short GO coating length) and loss increase (which dominates for the device with a relatively long GO coating length) for the SOI nanowires with different numbers of GO layers.
Figures 2b-i and b-ii show the power-dependent output spectra after going through the SOI nanowires with (i) 2 layers and (ii) 10 layers of GO lms. We measured the output spectra at 10 different coupled pulse energies ranging from ~8.2 pJ to ~51.5 pJ (i.e., coupled peak power from ~2.1 W to ~13.2 W). As the coupled pulse energy was increased, the output spectra showed increasing spectral broadening as expected. We also note that the broadened spectra exhibited a slight asymmetry. This was a combined result of both the asymmetry of the input pulse spectrum and the free-carrier effect of silicon including both the free carrier absorption (FCA) and free carrier dispersion (FCD) [36]. Since the time response for the generation of free carries is slower compared to the pulse width, there is a delayed impact of FCA on the pulse shape, which leads the spectral asymmetry of the optical pulses. The FCD further broadens the asymmetry induced by FCA, resulting in more obvious spectral asymmetry at the output.
To quantitively analyze the spectral broadening of the output spectra, we introduce the concept of a broadening factor (BF, de ned as the square of the pulse'rms spectral width at the waveguide output facet divided by the corresponding value at the input [37]). Figure 2c shows the BFs of the measured output spectra after transmission through the bare SOI nanowire and the GO-coated SOI nanowires at pulse energies of 8.2 pJ and 51.5 pJ. The GO-coated SOI nanowires show higher BFs than the bare SOI nanowires (i.e., GO layer number = 0), and the BFs at a coupled pulse energy of 51.5 pJ are higher than those at 8.2 pJ, agreeing with the results in Figures 2a and 2b,  The BFs of the output spectra versus coupled pulse energy are shown in Figures 2d-i and 2d-ii for the SOI nanowires with 1−3 layers and 5−20 layers of GO, respectively. The BFs increase with coupled pulse energy, re ecting a more signi cant spectral broadening that agrees with the results in Figure 2b. Figure 3a shows Kerr coe cient (n 2 ) of the GO lms versus layer number for xed coupled pulse energies of 8.2 pJ and 51.5 pJ, which is extract from the effective nonlinear parameter (γ eff ) of the hybrid waveguides using the following equation [30]: where λ c is the pulse central wavelength, D is the integral of the optical elds over the material regions, S z is the time-averaged Poynting vector calculated using Lumerical FDTD commercial mode solving software, n 0 (x, y) and n 2 (x, y) are the linear refractive index and n 2 pro les over the waveguide cross section, respectively.
The picosecond optical pulses used in our experiment had a relatively small spectral width (< 10 nm), we therefore neglected any variation in n 2 arising from its dispersion and used n 2 instead of the more general third-order nonlinearity χ (3) in our subsequent analysis and discussion. The values of n 2 are over three orders of magnitude higher than that of silicon and agree reasonably well with our previous waveguide FWM [30] and Z-scan measurements [28]. Note that the layer-by-layer characterization of n 2 for GO is challenging for Z-scan measurements due to the weak response of extremely thin 2D lms [25,28]. The high n 2 of GO lms highlights their strong Kerr nonlinearity for not only SPM but also other third-order (c (3) ) nonlinear processes such as FWM, and possibly even enhancing (c (3) ) for third harmonic generation (THG) and stimulated Raman scattering [38][39][40]. In Figure 3a, n 2 (both at 51.5 pJ and 8.2 pJ) decreases with GO layer number, showing a similar trend to WS 2 measured by a spatial-light system [41]. This is probably due to increased inhomogeneous defects within the GO layers as well as imperfect contact between the different GO layers. Although the n 2 of GO decreases with layer number, the increase in mode overlap with GO more than compensates for this, resulting in a net increase in γ eff with layer number. At 51.5 pJ, n 2 is slightly higher than at 8.2 pJ, indicating a more signi cant change in the GO optical properties with inceasing power. We also note that the decrease in n 2 with GO layer number becomes more gradual for thicker GO lms, possibly re ecting the transition of the GO lm properties towards bulk material properties − with a thickness independent n 2 .
To quantitively analyze the improvement in the nonlinear performance of the GO-coated SOI nanowires, we further calculated the effective nonlinear FOM (FOM eff ) for the GO-coated SOI nanowires. The resulting FOM eff (normalized to the FOM of silicon) is shown in Figure 3b where we see that a very high FOM eff of 20 times that of silicon is achieved for the hybrid SOI nanowires with 20 layers of GO. This is remarkable since it indicates that by coating SOI nanowires with GO lms, not only can the nonlinearity be signi cantly enhanced but the relative effect of nonlinear absorption can be greatly reduced as well. This is interesting given that the GO lms themselves cannot be described by a nonlinear FOM since the nonlinear absorption displays saturable absorption (SA) rather than TPA, and yet nonetheless the GO lms still are able to reduce the β TPA, eff ) of the hybrid waveguides, thus improving the overall nonlinear performance. Figure 4a shows the SiN waveguide integrated with a GO lm, along with a schematic showing atomic structure of GO with different oxygen functional groups (OFGs) such as hydroxyl, epoxide and carboxylic groups. SiN waveguides with a cross section of 1.6 µm × 0.66 µm were fabricated via annealing-free and crack-free processes that are compatible with CMOS fabrication [37,42]. Layered GO lms were coated on the top surface of the chip by a solution-based method that yielded layer-by-layer lm deposition, as mentioned in Sect. 2 [29,30,43]. Figure 4b shows a microscope image of a SiN waveguide patterned with 10 layers of GO, which illustrates the high transmittance and good morphology of the GO lms.  The coupled continuous-wave (CW) pump and signal power (18 dBm for each) was the same as that in Fig. 5a-i. The SiN waveguides with patterned GO lms also had an additional insertion loss as compared with the bare waveguide, while the results for both 5 and 10 GO layers show enhanced idler output powers. In particular, there is a maximum CE enhancement of ≈ 9.1 dB for the SiN waveguide patterned with 5 layers of GO, which is even higher than that for the uniformly coated waveguide with 1 layer of GO.

Enhanced Fwm In Go-coated Sin Waveguides
This re ects the trade-off between FWM enhancement (which dominates for the patterned devices with a short GO coating length) and loss (which dominates for the uniformly coated waveguides with a much longer GO coating length) in the GO-coated SiN waveguides. Figure 5b shows the measured CE versus pump power for the uniformly coated and patterned devices, respectively. The plots show the average of three measurements on the same samples and the error bars re ect the variations, showing that the measured CE is repeatable. As the pump power was increased, the measured CE increased linearly with no obvious saturation for the bare SiN waveguide and all the hybrid waveguides, indicating the low TPA of both the SiN waveguides and the GO lms. For the bare waveguide, the dependence of CE versus pump power shows a nearly linear relationship, with a slope rate of about 2 for the curve as expected from classical FWM theory [44][45][46][47][48][49][50][51][52][53][54]. For the GO-coated waveguides, the measured CE curves have shown slight deviations from the linear relationship with a slope rate of 2, particularly at high light powers. Figure 5c compares the CE of the hybrid waveguides with four different numbers of GO layers (i.e., 1, 2, 5, 10), where we see that the hybrid waveguide with an intermediate number of GO layers has the maximum CE. This re ects the trade-off between γ and loss in the hybrid waveguides, which both increase with GO layer number. Table 1 compares the performance of SOI nanowires, SiN, and Hydex waveguides incorporated with GO lms. As we can see in the Table, the dimensions of the three CMOS compatible photonic platforms were quite different. The SOI nanowire had the smallest waveguide dimensions and the tightest mode con nement, resulting in signi cantly increased mode overlap with the GO lm. This resulted in a signi cantly increased nonlinear parameter γ, but also the largest excess propagation loss induced by the GO lm. Mode overlap is a key factor for optimizing the trade-off between the Kerr nonlinearity and loss when introducing 2D layered GO lms onto different integrated platforms to enhance the nonlinear optical performance.

Conclusion
We demonstrate enhanced nonlinear optics including Kerr nonlinearity and FWM in SOI nanowires, Hydex and SiN waveguides and ring resonators incorporated with layered GO lms. We achieve precise control of the placement, thickness, and length of the GO lms using layer-by-layer coating of GO lms followed by photolithography and lift-off. Owing to the strong mode overlap between the platforms and the highly nonlinear GO lms, we achieve a high nonlinear parameter of GO coated SOI nanowires up to 16 times and an improved nonlinear FOM of up to a factor of 20. We obtain a signi cant improvement in the FWM CE of ≈ 7.3 dB for a uniformly coated SiN waveguide with 1 layer of GO and ≈ 9.1 dB for a patterned device with 5 layers of GO. These results verify the enhanced nonlinear optical performance of silicon, Hydex and SiN photonic devices achievable by incorporating 2D layered GO lms.