Ultra-broadband near- to mid-infrared electro-optic modulator on thin-film lithium niobate | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Ultra-broadband near- to mid-infrared electro-optic modulator on thin-film lithium niobate Li Shen, Qiyuan Li, Qiyuan Yi, ChengLin Shang, Sizhe Xing, Jinlai Cui, and 9 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6748087/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 08 Jan, 2026 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Abstract The escalating capacity limitations of conventional near-infrared telecommunication bands have spurred urgent investigations into full-spectrum optical communication systems spanning from the near-infrared to mid-infrared regimes. This has motivated the development of optical components combining broadband bandwidth with high-speed operation. Thin-film lithium niobate (TFLN) modulators, while exhibiting state-of-the-art performance in voltage-length product, optical loss, and electro-optic (EO) bandwidth at telecommunication wavelengths, face challenges in achieving broad operational bandwidth due to waveguide dispersion and velocity mismatch at longer wavelengths. Here, we present a Mach-Zehnder EO modulator comprising two adiabatic power splitters and a high-bandwidth phase shifter, achieving an unprecedented 800-nm operational bandwidth that covers the full optical fibre communication spectrum and extends into the 2-µm mid-infrared band. The fabricated modulator demonstrates > 67 GHz EO bandwidth in both O/C-bands and 48 GHz (detector-limited) in the 2-µm band. This device enables single-lane transmission exceeding 100 Gbaud across O-, C-, and 2-µm bands. Notably, we achieve 100 Gbaud OOK and 65 Gbaud PAM-4 transmission in the 2-µm band - the highest rates reported for this spectral region. This breakthrough establishes TFLN as a compelling platform for multispectral photonics, bridging conventional telecom infrastructure with emerging 2-µm technologies for next-generation full-spectrum optical communications. Physical sciences/Optics and photonics/Applied optics/Integrated optics Physical sciences/Optics and photonics/Applied optics/Optoelectronic devices and components Physical sciences/Optics and photonics/Optical materials and structures/Silicon photonics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Main The relentless growth of big data, cloud computing, and artificial intelligence has escalated global data traffic to unprecedented levels, challenging the capacity limits of conventional single-mode fiber (SMF)-based optical communication systems [1]. To address this, two complementary strategies have emerged: maximizing the utilization of existing spectral resources and exploring new optical communication windows beyond traditional O- and C-bands [2, 3]. Recent advances in hollow-core photonic bandgap fibers (HC-PBGFs) have unveiled ultra-broad low-loss transmission windows spanning 1240–1940 nm [4, 5], while innovations in amplifier technologies—ranging from rare-earth-doped fibers to the newly demonstrated integrated optical parametric amplifiers—deliver broadband gain across these extended spectral regions [6–8]. Together, these developments enable a transformative ultra-wide communication spectrum that seamlessly integrates conventional telecom bands with the emerging 2-μm waveband. Beyond revolutionizing optical networks, this expanded spectral range also holds promise for applications such as quantum photonics, precision metrology, and biomedical imaging [9–11]. However, the realization of end-to-end optical systems spanning these wavelengths remains hindered by a critical gap: the absence of high-speed broadband optical transmitters and photodetectors (PDs). While integrated PDs leveraging germanium or germanium tin (GeSn) on silicon photonic platforms have achieved 40–50 GHz bandwidths up to the 2-μm band [12–14], high-speed modulators—the core of optical transmitters—remain constrained by narrow optical bandwidths. Existing integrated platforms, such as silicon or germanium modulators based on free-carrier effects [15–19] and thin-film lithium niobate (TFLN) devices utilizing the Pockels effect [20], suffer from waveguide dispersion and velocity mismatch at longer wavelengths, resulting in limited optical bandwidths. These limitations restrict current integrated modulators to single-waveband operation, impeding their utility for multi-band systems. TFLN has emerged as a uniquely promising platform to overcome these barriers. Its combination of ultra-wide optical transparency (visible to mid-infrared), high EO coefficients, low optical loss, and compatibility with CMOS voltages enables modulators with unmatched bandwidth and efficiency [21–25]. Recent milestones include TFLN-based coherent transmission at 1.96 Tb/s [26] and compact packaged modules [27], underscoring its potential for next-generation networks. Although previous TFLN modulators have operated across discrete wavelengths from visible to 2-μm bands [20, 28–31], achieving seamless broadband operation—critical for emerging multi-band systems—remains an unresolved challenge. This requires overcoming the intrinsic optical bandwidth limitations of conventional device designs. Here, we demonstrate a high-speed TFLN electro-optic modulator with a record-breaking continuous operational range of 1260–2050 nm, seamlessly bridging the O-, C-, and 2-μm bands. By integrating adiabatic mode evolution engineering with the TFLN platform, our device achieves a 3-dB EO bandwidth exceeding 67 GHz in O/C-bands and 48 GHz in the 2-μm band—the highest reported for this spectral region. The modulator exhibits exceptional efficiency, with V π ·L values of 1.92, 2.61, and 3.94 V·cm at 1310, 1550, and 2000 nm, respectively. Experimental validation using a packaged device demonstrates error-free transmission of 108/115/100 Gbaud On-Off Keying (OOK) and 85/95/65 Gbaud 4-level Pulse Amplitude Modulation (PAM-4) signals across all three bands. This breakthrough paves the way for universal optical transmitters capable of unifying fragmented spectral resources into a single ultra-broadband communication infrastructure, with transformative potential for next-generation optical networks, quantum systems, and sensing technologies. Results D esign of the broadband TNLN modulator As shown in Fig. 1(a), the proposed extended full-spectrum optical communication system leverages TFLN technology to realize broadband optical transmitters spanning the O-band to the 2-μm wavelength range. This approach integrates the TFLN platform (NanoLN) with existing multi-wavelength laser sources and multi-band optical amplifiers. At the transmitter, a single full-spectrum TFLN EO modulator converts electrical signals—including OOK and advanced modulation formats such as PAM-4—into optical signals across the entire spectral range from the O-band to the 2-μm band. The EOMs are fabricated on a 300-nm-thick x-cut TFLN layer bonded to a 4.7-μm thermally grown silicon dioxide substrate with a silicon base. As illustrated in the inset of Fig. 1(a), the waveguide architecture features a 120-nm slab height and a sidewall angle of θ = 60°. Figure 1(b) displays the microscopy images of the fabricated EO modulator, which comprises two 3-dB power splitters, a 9-mm-long EO Mach-Zehnder interferometer modulation section, and a thermo-optic phase shifter, and two spot-size converters (SSCs) as the edge coupler for light coupling. The SEM images of an SSC and a power splitter are shown in the inset of Fig. 1(b). To achieve ultra-broadband operating bandwidth, the edge coupler was designed based on broadband SSC, which is illustrated in Fig. 2(a) and (b). The structure comprises LN bi-layer tapers and a silicon oxynitride (SiON) ridge waveguide. At the chip edge, the SiON waveguide interfaces with an ultra-high numerical-aperture fiber (UHNAF), which exhibits mode field diameters (MFDs) of 3.3 µm (1310 nm), 4.0 µm (1550 nm), and 5.3 µm (2000 nm). The fundamental transverse electric (TE) mode transitions adiabatically from the SiON waveguide to the LN waveguide via tapered thickness modulation. The refractive index of SiON (n = 1.54) lies between that of LN (n ≈ 2.2) and the silicon dioxide buried oxide (SiO₂ BOX, n ≈ 1.45), enabling optical confinement within the waveguide while suppressing leakage into the BOX. Achieving low coupling loss (<1 dB) across the 1310–2000 nm spectral range requires precise matching of the SiON waveguide and fiber MFDs. For a SiON rib waveguide with a rib height of 3.2 µm and slab height of 1 µm. Fig. 2(c) shows the simulated coupling loss as a function of rib width at λ = 1310, 1550, and 2000 nm. The simulations reveal an optimal rib width of 4.4 µm, where the mode-field overlap peaks at 1550 nm (C-band) and gradually decreases at 1310 nm (O-band) and 2000 nm. This design also demonstrates robust tolerance to dimensional variations in the SiON waveguide cross-section. Table 1. Parameters of the SSC Parameter w 1 w 2 w 3 w 4 w 5 w 6 s 1 s 2 s 3 h 1 h 2 Value(μm) 0.2 0.2 2.1 0.8 5 4.4 250 150 100 3.2 1 The SSC parameters were optimized through a dual focus: maximizing the taper length to ensure adiabatic mode evolution and minimizing the tip width to enhance coupling efficiency, while maintaining fabrication feasibility. The finalized SSC parameters are summarized in Table 1. Figure 2(d) presents the calculated transmission spectrum across 1200–2100 nm, with transmission efficiencies of 91.4%, 92.2%, and 88.4% at 1310 nm, 1550 nm, and 2000 nm, respectively. Minor spectral ripples below 1500 nm originate from reflection-induced interference between the LN slab and rib waveguide, as revealed by mode field simulations. Notably, the simulation framework omitted intrinsic optical absorption losses in plasma-deposited silicon oxynitride (SiOxNy) near 1510 nm and 2000 nm—a factor that may influence experimental broadband performance. The full spectrum 3-dB power splitter is based on TFLN adiabatic waveguide, with light is adiabatically coupled from the input waveguide to two symmetric output waveguides through three tapers. Further design details of the 3-dB power splitter are provided in Ref. [32]. For shorter wavelengths (e.g., O-band), waveguides thicker than 300 nm impose stricter adiabatic mode evolution requirements, which would degrade the 3-dB power splitter performance. While the reduced electro-optic interaction at this thickness increases the half-wave voltage, this trade-off was essential to achieve a functional bandwidth spanning 1260–2050 nm. As shown in Fig. 2(e), the calculated transmission for each splitter is 48.2%, 49.6%, and 44.6% at 1310, 1550, and 2000 nm, respectively, confirming robust performance across the targeted spectrum. Outside the modulation region, the waveguides are engineered to support single-mode operation at 1310 nm, ensuring broadband compatibility across the O-/C- and 2-μm bands. This design suppresses higher-order mode excitation over the 1260–2050 nm range, thereby preventing EO performance degradation in the modulation region. Finite-difference eigenmode (FDE) simulations indicate that single-mode operation requires a waveguide top width of ~0.8 μm. However, this narrow width reduces optical mode confinement, increasing propagation losses due to stronger mode overlap with waveguide sidewalls and cladding, particularly at longer wavelengths. Additionally, the constrained geometry elevates the half-wave voltage requirement. To balance these trade-offs, we implement an adiabatic taper proximal to the 3-dB power splitter, expanding the waveguide width to W = 3.5 μm within the modulation region. A FDE solver was employed to numerically analyze the coupled electrical-optical modal properties of the device architecture. The electrode gap (G) is defined as the vertical separation between the ground plane and the signal electrode. As shown in Fig. 2(f), we calculated the half-wave voltage-length product (V π ·L) and quantified the metal-induced absorption loss, both parameterized as functions of G at wavelengths of 1310, 1550, and 2000 nm. The results reveal an inverse relationship between device modulation efficiency (V π ·L) and optical loss performance, necessitating tailored selection of G to prioritize either metric for specific applications. For this work, an electrode gap of G = 6 μm was chosen to achieve an optimal trade-off point, simultaneously ensuring strong modulation performance while limiting absorption loss to moderate levels. The traveling-wave electrodes feature a push-pull configuration based on a coplanar waveguide (CPW) structure. The signal and ground electrode widths are designed as W S = 25 μm and W G = 150 μm, respectively, with gold (Au) electrodes of height H = 1 μm and a 2-μm-thick SiO₂ cladding layer. As illustrated in the inset of Fig. 2(g), the characteristic impedance of the CPW is set to 42 Ω, slightly lower than the system’s characteristic impedance, to mitigate the low-frequency S 21 roll-off and enhance the operational bandwidth. The optical group indices are 2.18, 2.13, and 2.04 at wavelengths (λ) of 1310 nm, 1550 nm, and 2000 nm, respectively (dashed lines in Fig. 2(g)). The optical group index at the C-band is aligned with the radio frequency (RF) effective index, resulting in a trade-off in electro-optic (EO) bandwidth performance for the 2-μm band. The RF effective index is designed as 2.13 at 50 GHz. For λ = 2000 nm, the residual index mismatch leads to a theoretical 3-dB bandwidth of approximately 80 GHz for a 9-mm-long, impedance-matched, lossless modulator. Measurement and analysis The fabrication details of TFLN modulator is given in the ‘Methods’. The coupling losses of the SSC were characterized using a reference waveguide coupled to a UHNA4 fiber. Refractive index-matching oil (index = 1.46) was applied at the fiber-chip interface to minimize reflections. Coupling losses were measured using tunable lasers for the 1260–1360 nm and 1500–1630 nm bands, and a custom-built 2-μm amplified spontaneous emission (ASE) source paired with an optical spectrum analyzer (Yokogawa AQ6375B) for the 1920–2060 nm range. Using a polarization controller (PC), the measured TE mode coupling losses are 0.78 dB/facet at 1310 nm, 0.69 dB/facet at 1550 nm, and 0.82 dB/facet at 1970 nm, as shown in Fig. 3(a). The slightly higher losses at 1310 nm and 1970 nm are attributed to minor mode mismatch between the UHNA4 fiber and the SiON waveguide. Notably, two pronounced loss peaks near 1510 nm and 2000 nm correlate with intrinsic N–H infrared absorption bands in the plasma-deposited SiON layer, consistent with prior reports of N–H bond absorption [33–36]. Following characterization of the modulator’s total optical losses, the on-chip insertion losses (ILs) were deduced as 1.2 dB, 2.8 dB, and 5.8 dB at wavelengths of 1310 nm, 1550 nm, and 1970 nm, respectively. A test structure comprising 16 cascaded 3-dB power splitters was used to measure the ILs of individual splitters, yielding values of 0.23 dB, 0.11 dB, and 0.42 dB at 1310 nm, 1550 nm, and 1970 nm, closely matching simulations. After accounting for contributions from the power splitters and waveguide propagation losses, the primary source of on-chip loss was identified as absorption by the metal electrodes, which becomes more pronounced at longer wavelengths (e.g., 1970 nm). Experimental on-chip losses for a gap spacing ( G = 6 μm) show strong agreement with simulated values, as illustrated by the dashed lines in Fig. 2(f). Fig. 3(c) illustrates the half-wave voltage V π measurements with a 100 kHz triangular voltage sweep. With a modulator length of 9 mm, the V π ·L values are calculated to be 1.92/2.61/3.94 V·cm for λ = 1310/1550/2000 nm, respectively. Notably, the experimental V π ·L and loss are consistent with the simulation results in Fig. 2(f) for G = 6 μm, validating the accuracy of our numerical model. This agreement confirms the reliability of our simulations for predicting performance trends across varying values of electrode gap. For instance, the simulations suggest that prioritizing reduced metal absorption loss over minimizing V π can be achieved by increasing G slightly to 7 μm would maintain comparable high-frequency EO performance. Additionally, static extinction ratios (ERs) were measured by sweeping the heater voltage, yielding values of ~17, ~15, and ~18 dB at 1310, 1550, and 1970 nm, respectively. The observed wavelength-dependent ER variations primarily originate from power imbalance between the output ports of the 3-dB splitter, attributed to fabrication tolerances. The EO responses of the fabricated TFLN modulator were characterized using a 67-GHz vector network analyzer (VNA, Keysight N5247B). A high-speed RF probe was employed to deliver RF signals from the VNA, while a second probe provided a 50 Ω termination. The EO response was captured via electrical signals received from the PD into the VNA. For O- and C-band measurements, commercial PDs were used to record the EO S 21 curves (Fig. 3c). The results indicate that the modulator's 3-dB bandwidth exceeds 67-GHz, which is the limitation of the VNA. For the 2-μm band, where commercial PDs lack sufficient bandwidth, we employed a recently demonstrated high-speed GeSn PD (bandwidth >40 GHz, Ref. [12]) to measure the bandwidth. After system calibration and PD-response de-embedding, characterization of the frequency response and high-speed data transmission capabilities was achieved. The measured EO S 21 response at λ = 2 μm (Fig. 3c) reveals a 3-dB bandwidth of ~48 GHz, the highest reported to date for optical modulators operating in the 2-μm band. To evaluate high-speed data transmission performance, we characterized the TFLN modulator in a butterfly package across three wavelength regimes: O-band (1310 nm), C-band (1550 nm), and 2-μm band. Fig. 4(a) illustrates the experimental setup for measuring eye diagrams and bit error rate (BER). Optical signals were generated using three laser sources at 1310 nm, 1550 nm, and 2000 nm. High-frequency RF signals, synthesized by a 120 GS/s arbitrary waveform generator (AWG, Keysight M8194A), were amplified to ~2.4 Vpp using an RF amplifier and applied to the modulator. The bias point of the modulator was stabilized via an integrated heater. For O- and C-band measurements, the modulated optical output was amplified using an O-band semiconductor optical amplifier (SOA) or erbium-doped fiber amplifier (EDFA), respectively, before detection by commercial PDs. For the 2-μm band, the modulated light was amplified using a thulium-doped fiber amplifier (TDFA) and coupled into the GeSn PD, described in prior characterization. The inset of Fig. 4(a) displays measured eye diagrams at the maximum baud rates achieved while maintaining BER below the hard decision forward error correction (HD-FEC) threshold of 3.8×10⁻³. Clear eye openings were obsered for 108/115/100 Gbaud OOK signals at 1310/1550/2000 nm, respectively. To our knowledge, this represents the highest OOK transmission rate in the 2-μm band and the first packaged TFLN modulator capable of exceeding 100 Gbaud OOK operation across all three wavelength bands. We further evaluated PAM-4 performance by measuring BER curves as a function of received optical power (ROP), defined as the optical power incident on the PD. A variable optical attenuator (VOA) was inserted after the optical amplifiers to adjust ROP. Fig. 4(b) shows back-to-back (B2B) BER curves for the TFLN modulator. For 1310 nm and 1550 nm, transmission results through a 500-m standard SMF were also included and exhibit negligible power penalties compared to the B2B case. Clear eye openings for 80/95/65 Gbaud PAM-4 signals were obtained at 1310/1550/2000 nm, corresponding to net data rates of 160/190/130 Gbps. Despite the bandwidth limitations from the RF port packaging (1.85 mm aperture) might potentially constrain the modulator for higher speed data transmission, it is worth noting that the 130 Gbps achieved for the 2-μm band is the highest net bit rate reported to date for this wavelength regime [18]. Discussion and conclusion Fig. 5 compares the 3-dB EO bandwidths of integrated modulators in the short-wave infrared spectrum regime from 1.2–2.2 μm. Our TFLN modulator operates across an unprecedented 800-nm optical bandwidth (1260–2050 nm), spanning the O-band (1260–1360 nm), C-band (1530–1565 nm), and extending into the 2-μm regime (1900–2050 nm). This broad operational range enables multi-band compatibility within a single device, a critical advantage for versatile photonic systems. The TFLN modulator maintains a flat frequency response across its operational range. The 48-GHz bandwidth at 2000 nm enables, to our knowledge, demonstration of the highest single-lane 100 Gb/s OOK data transmission in this band. This represents a 2.3-fold improvement in bandwidth over existing 2-μm modulators, directly enhancing data capacity and underscoring the potential for next-generation high-speed communication systems. The detailed comparasions of the state of art integrated EO modulators operating in the 2-μm spectral band are given in the ‘Methods’. In conclusion, we demonstrate a TFLN modulator featuring high EO bandwidth and a record-breaking operational range spanning over 800 nm, from the O-band to the 2-μm spectral region. The device achieves 3-dB EO bandwidths of over 67 GHz in both the O- and C-bands and 48 GHz in the 2-μm band. High speed data transmisions of 108 Gbaud, 115 Gbaud, and 100 Gbaud OOK signals were experimentally validated at 1310 nm, 1550 nm, and 2000 nm wavelengths, with BERs below the HD-FEC threshold of 3.8 × 10⁻³. These results underscore the transformative potential of TFLN modulators in enabling ultrabroadband optical communication systems that seamlessly bridge conventional telecom bands with the emerging 2-μm window. This advancement directly addresses the escalating bandwidth demands of next-generation data centers and high-performance computing infrastructures. Methods Fabrication of the TNLF modulator The fabrication process of the TNLF modulator is outlined as follows: First, electron beam lithography (EBL) was used to define ridge waveguide structures on a 700-nm-thick layer of AR-P 6200 resist. The ridge waveguide was then formed by etching 180 nm of lithium niobate (LN) using Ar⁺-based inductively coupled plasma (ICP) dry etching. A 2-µm-thick silica cladding layer was deposited over the waveguide via plasma-enhanced chemical vapor deposition (PECVD) to encapsulate the modulation section. This silica layer was selectively etched using ICP dry etching in the electrode and spot-size converter (SSC) regions. To create the LN bi-layer tapers, the EBL and ICP etching steps were repeated to remove the remaining 120 nm of LN. A 200-nm-thick thermal titanium (Ti) layer was deposited by electron-beam evaporation (EBE), after which the heater pattern was defined via EBL. A 1-µm-thick Au traveling-wave electrode was then formed by EBE deposition and lift-off. Next, a 4.2-µm-thick SiON layer was grown across the entire chip using PECVD with silane, ammonia, and nitrous oxide precursors. The SiON layer was patterned and etched twice to form the SiON ridge waveguide in the SSC region, while the SiON above the modulation section was removed. A 2-µm-thick protective SiO₂ layer was deposited by PECVD to shield the SSC and minimize contamination-induced losses in the modulator. Finally, the cladding thickness in the modulation section was reduced to 2 µm via ICP etching, and the electrode pads were exposed. Comparisons of the 2-µm integrated EO modulators Table 2 summarizes the performance metrics of integrated electro-optic modulators operating in the 2-µm spectral band. Our TFLN modulator achieves a 48 GHz EO 3-dB bandwidth—the highest reported value in this spectral region—enabling a baud rate of 100 Gbaud (100 Gb/s for on-off keying, OOK) and surpassing prior state-of-the-art devices. This advancement establishes a new benchmark for high-speed data transmission in the 2-µm band. Table 2 Performance comparison of on-chip integrated EO modulators operating at 2-µm band Ref. Platform Structure Extinction Ratio (dB) 3-dB Bandwidth (GHz) Max. Baud Rate and Signal Format Other wavebands capable [ 15 ] SOI MZI - 10 20 Gbaud OOK C-band [ 16 ] SOI Michelson 15 - 20 Gbaud OOK 15 Gbaud PAM-4 No [ 20 ] LNOI MZI 20 22 32 Gbaud OOK No [ 17 ] SOI MZI 22 18 30 Gbaud OOK 40 Gbaud PAM-4 No [ 18 ] SOI MRM 19 18 50 Gbaud OOK No [ 19 ] SOI Racetrack Ring 21 26 34 Gbaud OOK No This work LNOI MZI 18 48 100 Gbaud OOK 65 Gbaud PAM-4 O-/C-bands Declarations Data availability All data are available in the main text or Supplementary Information. The data are available from the corresponding authors upon reasonable request. Acknowledgements This work was supported by the National Natural Science Foundation of China (62175080, 62075074) and National Major Research and Development Program (2022YFB2803600), the open research fund of Songshan Lake Materials Laboratory (2023SLABFK11). The authors thank the Center of Optoelectronic Micro and Nano Fabrication and Characterizing Facility, Wuhan National Laboratory for Optoelectronics of Huazhong University of Science and Technology for the support in device fabrication. Author contributions Q.L., Q.Y., A.P., and L.S. conceived the project. Q.L. and Q.Y. carried out simulations and designed the TFLN modulator. A.P. and C.S. fabricated the modulator. Q.L., S.X., J.C., Y.Z., and G.C. characterized the modulator and analyzed experimental data. J.Z., J.Z., C.Z., S.Z., and L.S. discussed the results. Q.L. and L.S. wrote the manuscript with contribution from all authors. L.S., N.C., J.X., and M.Z. supervised the whole project. Competing interests The authors declare no competing interests. References T. Mizuno, and Y. Miyamoto, “High-capacity dense space division multiplexing transmission,” Opt. Fiber Technol. 35, 108-117 (2017). R. Soref, “Enabling 2 -μm communications,” Nat. Photonics 9, 358-359 (2015). Z. Liu, Y. Chen, Z. Li, B. Kelly, R. Phelan, J. Carroll, T. Bradley, J. P. Wooler, N. V. Wheeler, A. M. Heidt, T. Richter, C. Schubert, M. Becker, F. Poletti, M. N. Petrovich, S. Alam, D. J. Richardson, and R. Slavik, “High-capacity directly modulated optical transmitter for 2-μm spectral region,” J. Lightwave Technol. 33(7), 1373-1379 (2015). H. Sakr, Y. Hong, T. D. Bradley, G. T. Jasion, J.R.Hayes, H.Kim, I. A. Davidson, E. N. Fokoua, Y. Chen, K. R. H. Bottrill, N. Taengnoi, N. V. Wheeler, P. Petropoulos, D. J. Richardson and F. Poletti, “Interband Short Reach Data Transmission in Ultrawide Bandwidth Hollow Core Fiber,” J. Lightwave Technol. 38(1), 159-165 (2020). P. J. Roberts, F. Couny, H. Sabert, B. J. Mangan, D. P. Williams, L. Farr, M. W. Mason, and A. Tomlinson, “Ultimate low loss of hollowcore photonic crystal fibres,” Opt. Express 13(1), 236-244 (2005). Z. Li, A. Heidt, N. Simakov, et al, “Diode-pumped wideband thuliumdoped fiber amplifiers for optical communications in the 1800-2050 nm window,” Opt. Express 21(22), 26450-26455 (2013). P. Zhao, V. Shekhawat, M. Girardi, Z. He, V. T. -Company, and P. A. Andrekson, “Ultra-broadband optical amplification using nonlinear integrated waveguides,” Nature 640, 918-923 (2025). N. Kuznetsov, A. Nardi, J. Riemensberger, A. Davydova, M. Churaev, P. Seidler, and T. J. Kippenberg, “An ultra-broadband photonic-chip-based parametric amplifier,” Nature 639, 928-934 (2025). T. Bruns, T. S. Bischof, D. K. Harris, D. Franke, Y. Shi, L. Riedemann, A. Bartelt, F. B. Jaworski, J. A. Carr, C. J. Rowlands, M. W. B. Wilson, O. Chen, H. Wei, G. W. Hwang, D. M. Montana, I. Coropceanu, O. B. Achorn, J. Kloepper, J. Heeren, P. T. C. So, D. Fukumura, K. F. Jensen, R. K. Jain and M. G. Bawendi, “Next-generation in vivo optical imaging with short-wave infrared quantum dots,” Nat. Biomed. Eng. 1, 0056 (2017). D. R. Carlson, D. D. Hickstein, W. Zhang, A. J. Metcalf, F. Quinlan, S. Diddams, and S. A. Papp, “Ultrafast electro-optic light with subcycle control,” Science 361, 1358 (2018). M. Karpiński, M. Jachura, L. J. Wright and B. J. Smith, “Bandwidth manipulation of quantum light by an electro-optic time lens,” Nat. Photonics 11, 53 (2017). J. Cui, J. Zheng, Y. Zhu, X. Liu, Y. Wu, Q. Huang, Y. Yang, Z. Liu, Z. Liu, Y. Zuo, and B. Cheng, “High-speed GeSn resonance cavity enhanced photodetectors for a 50 Gbps Si-based 2 μm band communication system,” Photon. Res. 12(4), 767-773 (2024). Y. Zhu, Z. Liu, Z. Liu, Y. Hu, Q. Huang, Y. Yang, X. Liu, T. Men, G. Zhang, J. Zheng, Y. Zuo, and B. Cheng, “112 Gbps CMOS-Compatible Waveguide Germanium Photodetector for the 2 μm Wavelength Band with a 3.64 A/W Responsivity,” Photon. Res. 12(11), 2373-2378 (2024). J. Wang, X. Zhang, Y. Li, X. Wang, S. Xiao, Q. Song, Z. He, J. Du, H. K. Tsang, and K. Xu, “112 Gb s −1 Germanium Photodetector at 2 µm Enabled by a 3D Integrated Waveguide Loop,” Adv. Mater. Technol. 2401718 (2025). W. Cao, D. Hagan, D. J. Thomson, M. Nedeljkovic, C. G. Littlejohns, A. Knights, S. Alam, J. Wang, F. Gardes, W. Zhang, S. Liu, K. Li, M. S. Rouifed, G. Xin, W. Wang, H. Wang, G. T. Reed, and G. Z. Mashanovich, “High-speed silicon modulators for the 2 μm wavelength band,” Optica 5(9), 1055-1062 (2018). W. Cao, S. Liu, C. G. Littlejohns, D. J. Thomson, M. Nedeljković, W. Zhang, K. Li, M. Banakar, Y. Tran, X. Yan, H. Du, Z. Ren, F. Gardes, G. T. Reed, and G. Z. Mashanovich, “High-speed silicon Michelson interferometer modulator and streamlined IMDD PAM-4 transmission of Mach-Zehnder modulators for the 2 μm wavelength band,” Opt. Express 29(10), 14438-14451 (2021). X. Wang, W. Shen, W. Li, Y. Liu, Y. Yao, J. Du, Q. Song, and K. Xu, “High-speed silicon photonic Mach-Zehnder modulator at 2 μm,” Photon. Res. 9, 535-540 (2021). W. Shen, G. Zhou, J. Du, L. Zhou, K. Xu, and Z. He, “High-speed silicon microring modulator at the 2 µm waveband with analysis and observation of optical bistability,” Photon. Res. 10(3), A35 (2022). X. Wang, J. Wang, Y. Yao, S. Xiao, Q. Song, and K. Xu, “Efficient and high-speed coupling modulation of silicon racetrack ring resonators at 2 µm waveband,” Opt. Lett. 49(8), 2157-2160 (2024). B. Pan, J. Hu, Y. Huang, L. Song, J. Wang, P. Chen, Z. Yu, L. Liu, and D. Dai, “Demonstration of high-speed thin-film lithium-niobate-on-insulator optical modulators at the 2-μm wavelength,” Opt. Express 29(12), 17710-17717 (2021). C. Wang, M. Zhang, X. Chen, M. Bertrand, A. Shams-Ansari, S. Chandrasekhar, P. Winzer, and M. Loncar, “Integrated lithium niobate electro-optic modulators operating at CMOS-compatible voltages,” Nature 562, 101 (2018). H. Feng, T. Ge, X. Guo, B. Wang, Y. Zhang, Z. Chen, S. Zhu, K. Zhang, W. Sun, C. Huang, Y. Yuan and C. Wang, “Integrated lithium niobate microwave photonic processing engine,” Nature 627, 80-87 (2024). M. He, M. Xu, Y. Ren, J. Jian, Z. Ruan, Y. Xu, S. Gao, S. Sun, X. Wen, L. Zhou, L. Liu, C. Guo, H. Chen, S. Yu, L. Liu, and X. Cai, “High-performance hybrid silicon and lithium niobate Mach-Zehnder modulators for 100 Gbit s−1 and beyond,” Nat. Photonics 13, 359-364 (2019). Y. Zhang, L. Shao, J. Yang, Z. Chen, K. Zhang, K. -M. Shum, D. Zhu, C. H. Chan, M. Loncar, and C. Wang, “Systematic investigation of millimeter-wave optic modulation performance in thin-film lithium niobate,” Photon. Res. 10, 2380-2387 (2022). G. Chen, Y. Gao, H. -L. Lin, and A. J. Danner, “Compact and Efficient Thin-Film Lithium Niobate Modulators,” Adv. Photonics Res. 4, 2300229 (2023). M. Xu, Y. Zhu, F. Pittalà, J. Tang, M. He, W. C. Ng, J. Wang, Z. Ruan, X. Tang, M. Kuschnerov, L. Liu, S. Yu, B. Zheng, and X. Cai, “Dual-polarization thin-film lithium niobate in-phase quadrature modulators for terabit-per-second transmission,” Optica 9, 61 (2022). S. Makino, S. Takeuchi, S. Maruyama, M. Doi, Y. Ohmori, and Y. Kubota, in Optical Fiber Communication Conference (OFC) (IEEE, 2022), paper M1D.2. D. Renaud, D. R. Assumpcao, G. Joe, A. S. -Ansari, D. Zhu, Y. Hu, N. Sinclair, and M. Loncar, “Sub-1 Volt and high-bandwidth visible to near-infrared electro-optic modulators,” Nat. Commun. 14, 1496 (2023). S. Xue, Z. Shi, J. Ling, Z. Gao, Q. Hu, K. Zhang, G. Valentine, X. Wu, J. Staffa, U. A. Javid, and Q. Lin, “Full-spectrum visible electro-optic modulator,” Optica 10, 125 (2023). F. Valdez, V. Mere, X. Wang, and S. Mookherjea, “Integrated O- and C-band silicon-lithium niobate Mach-Zehnder modulators with 100 GHz bandwidth, low voltage, and low loss,” Opt. Express 31(4), 5273-5289 (2023). X. Fang, F. Yang, X. Chen, Y. Li, and F. Zhang, “Ultrahigh-Speed Optical Interconnects With Thin Film Lithium Niobate Modulator,” J. Lightwave Technol. 41(4), 1207-1215 (2023). Q. Yi, A. Pan, J. Xia, C. Zeng, and L. Shen, “Ultra-broadband 1 × 2 3 dB power splitter using a thin-film lithium niobate from 1.2 to 2 µm wave band,” Opt. Lett. 48(20), 5375-5378 (2023). S. E. Krikorian, and M. Mahpour, “The identification and origin of N-H overtone and combination bands in the near-infrared spectra of simple primary and secondary amides,” Spectrochim. Acta A 29(7), 1233-1246 (1973). L. R. Brown, and J. S. Margolis, “Empirical line parameters of NH 3 from 4791 to 5294 cm −1 ,” J. Quant. Spectrosc. Radiat. Transfer 56(2), 283-294 (1996). C. A. Beale, A. Wong, and P. Bernath, “Infrared transmission spectra of hot ammonia in the 4800-9000 cm −1 region,” J. Quant. Spectrosc. Radiat. Transfer 246, 106911 (2020). D. R. Dadadzhanov, T. A. Vartanyan, and A. Karabchevsky, “Lattice Rayleigh Anomaly Associated Enhancement of NH and CH Stretching Modes on Gold Metasurfaces for Overtone Detection,” Nanomaterials 10(7), 1265 (2020). M. Samiul Alam, X. Li, M. Jacques, Z. Xing, A. Samani, E. E. -Fiky, P. Koh, and D. V. Plant, “Net 220 Gbps/λ IM/DD Transmssion in O-Band and C-Band With Silicon Photonic Traveling-Wave MZM,” J. Lightwave Technol. 39, 4270 (2021). C. Han, Z. Zheng, H. Shu, M. Jin, J. Qin, R. Chen, Y. Tao, B. Shen, B. Bai, F. Yang, Y. Wang, H. Wang, Z. Zhang, S. Yu, C. Peng and X. Wang, “Slow-light silicon modulator with 110-GHz bandwidth,” Sci. Adv. 9, eadi5339 (2023). J. Shen, Y. Zhang, L. Zhang, J. Li, C. Feng, Y. Jiang, H. Wang, X. Li, Y. He, X. Ji, G. Yin, Y. Tian, X. Xiao, and Y. Su, “Highly Efficient Slow‐Light Mach-Zehnder Modulator Achieving 0.21 V·cm Efficiency with Bandwidth Surpassing 110 GHz,” Laser Photonics Rev. 18(1), 2200192 (2024). M. Li, L. Wang, X. Li, X. Xiao, and S. Yu, “Silicon intensity Mach–Zehnder modulator for single lane 100 Gb/s applications,” Photon. Res. 6, 109 (2018). P. Kharel, C. Reimer, K. Luke, L. He, and M. Zhang, “Breaking voltage-bandwidth limits in integrated lithium niobate modulators using micro-structured electrodes,” Optica 8, 357 (2021). Additional Declarations There is NO Competing Interest. Cite Share Download PDF Status: Published Journal Publication published 08 Jan, 2026 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6748087","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":463094695,"identity":"1f48bfdb-f240-4d34-b4d8-cb0fe73e6490","order_by":0,"name":"Li Shen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzElEQVRIiWNgGAWjYBACPmYGNiBlkwDhshGhhQ2iJQ2ohZlYLRBlh0nRws7+7MHHHefzDM6fP8DwoewwA//sBkIO4zE3nHnmdrHBjWQGxhnnDjNI3DlAUAubNG/b7cQNN5gZmHnbDjMYSCQQ0sL+TPpv27nEDecPMzD/JU4Lg5k0Y9uBxA0HkhmYGYnTwmMm2duWnDjzRrLBwZ5z6TwSNwho4ec//kziZ5tdYt/5gw8f/CizluOfQUALCjgAxDwkqB8Fo2AUjIJRgAsAAP3nPdJloqBoAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0001-7502-3643","institution":"Huazhong University of Science and Technology","correspondingAuthor":true,"prefix":"","firstName":"Li","middleName":"","lastName":"Shen","suffix":""},{"id":463094696,"identity":"988af748-c2b1-4c0d-8338-ef0a3c3c4347","order_by":1,"name":"Qiyuan Li","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Qiyuan","middleName":"","lastName":"Li","suffix":""},{"id":463094697,"identity":"0fd5d03b-f165-4632-9527-110c7846eefb","order_by":2,"name":"Qiyuan Yi","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Qiyuan","middleName":"","lastName":"Yi","suffix":""},{"id":463094698,"identity":"fc04bb84-b171-40d0-a8c3-468b74f142f1","order_by":3,"name":"ChengLin Shang","email":"","orcid":"","institution":"Wuhan National Laboratory for Optoelectronics and School of Optical and Electronic Information, Huazhong University of Science and Technology, Wuhan 430074 Hubei, China","correspondingAuthor":false,"prefix":"","firstName":"ChengLin","middleName":"","lastName":"Shang","suffix":""},{"id":463094699,"identity":"8ff46ac5-f895-4e0c-a3d9-b1f4b9df7dda","order_by":4,"name":"Sizhe Xing","email":"","orcid":"","institution":"Fudan University","correspondingAuthor":false,"prefix":"","firstName":"Sizhe","middleName":"","lastName":"Xing","suffix":""},{"id":463094700,"identity":"6c680401-0db3-4e62-aace-88e20e7d1043","order_by":5,"name":"Jinlai Cui","email":"","orcid":"","institution":"Institute of Semiconductors, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Jinlai","middleName":"","lastName":"Cui","suffix":""},{"id":463094701,"identity":"df894c71-4fc7-46ac-9723-0cdb765a7883","order_by":6,"name":"Yupeng Zhu","email":"","orcid":"","institution":"Institute of Semiconductors, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Yupeng","middleName":"","lastName":"Zhu","suffix":""},{"id":463094702,"identity":"bb8644fd-9cd6-477a-8bd2-fdb855d0bb09","order_by":7,"name":"Jun Zheng","email":"","orcid":"","institution":"Institute of Semiconductors, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Jun","middleName":"","lastName":"Zheng","suffix":""},{"id":463094703,"identity":"0c758651-8781-4e09-8802-6bce70bda136","order_by":8,"name":"Junwen Zhang","email":"","orcid":"https://orcid.org/0000-0001-8041-1608","institution":"Fudan Univeristy","correspondingAuthor":false,"prefix":"","firstName":"Junwen","middleName":"","lastName":"Zhang","suffix":""},{"id":463094704,"identity":"899bb2d7-0e2d-47e2-9901-99ec8bead0e3","order_by":9,"name":"Nan Chi","email":"","orcid":"https://orcid.org/0000-0003-4966-3844","institution":"Fudan University","correspondingAuthor":false,"prefix":"","firstName":"Nan","middleName":"","lastName":"Chi","suffix":""},{"id":463094705,"identity":"bab24035-997b-43a0-99ea-cb46750a7d96","order_by":10,"name":"An Pan","email":"","orcid":"","institution":"Wuhan National Laboratory for Optoelectronics and School of Optical and Electronic Information, Huazhong University of Science and Technology, Wuhan 430074 Hubei, China","correspondingAuthor":false,"prefix":"","firstName":"An","middleName":"","lastName":"Pan","suffix":""},{"id":463094706,"identity":"540ec8dd-e719-4210-adcd-8f0586947d6b","order_by":11,"name":"Cheng Zeng","email":"","orcid":"","institution":"Wuhan National Laboratory for Optoelectronics, Huazhong University of Science and Technology, Wuhan 430074","correspondingAuthor":false,"prefix":"","firstName":"Cheng","middleName":"","lastName":"Zeng","suffix":""},{"id":463094707,"identity":"cd1c1d18-bb80-4848-beea-fe0a956b5423","order_by":12,"name":"Jinsong Xia","email":"","orcid":"https://orcid.org/0000-0002-9650-7839","institution":"Wuhan National Laboratory for Optoelectronics, Huazhong University of Science and Technology, Wuhan 430074","correspondingAuthor":false,"prefix":"","firstName":"Jinsong","middleName":"","lastName":"Xia","suffix":""},{"id":463094708,"identity":"c66bf23d-3079-4df6-b0c0-571a4daaa342","order_by":13,"name":"shuang zheng","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"shuang","middleName":"","lastName":"zheng","suffix":""},{"id":463094709,"identity":"39a00934-2432-4e00-aef7-0acc760aa1b8","order_by":14,"name":"Minming Zhang","email":"","orcid":"","institution":"Huazhong University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Minming","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2025-05-26 07:25:23","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6748087/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6748087/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41467-025-67902-2","type":"published","date":"2026-01-08T05:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":83600838,"identity":"6a9538d6-1f88-4049-8e14-f5035bf2381b","added_by":"auto","created_at":"2025-05-29 09:09:46","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1068619,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Concept of extended full spectrum optical communication through broadband TFLN optical modulator, multi-wavelength laser sources, multi-band optical amplifiers, wide-band optical fiber and PDs. Inset: cross-sectional view of the modulator region. (b) Microscopy images of the fabricated LN Mach–Zehnder EOM. Inset: SEM images of the SSC and power splitter.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6748087/v1/e25760af5c56a935dbb3f954.png"},{"id":83600836,"identity":"101c0c84-0800-4002-8b8b-b8b5aa7ffe8d","added_by":"auto","created_at":"2025-05-29 09:09:46","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":642787,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Schematic structure of the SSC. The inset shows the dimension parameters of the SiON waveguide. (b) Dimension parameters of the LN bi-layer tapers. (c) Dependence of the coupling loss on SiON rib width. (d) Simulated transmission spectrum of the SSC. (e) Simulated transmission spectrum of the 3-dB power splitter. (f) Simulated V\u003csub\u003eπ\u003c/sub\u003e·L and propagation loss as a function of G. (g) Simulated frequency-dependent results of RF effective index n\u003csub\u003em\u003c/sub\u003e. The three dashed lines correspond to the optical group index ng. The inset shows the simulated characteristic impedance Z\u003csub\u003e0\u003c/sub\u003e.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6748087/v1/2778673135faa04c5fe7fffa.png"},{"id":83600837,"identity":"311f6b7f-6494-489e-8945-6bb604ef3cff","added_by":"auto","created_at":"2025-05-29 09:09:46","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":662439,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Measured coupling loss of the SSC for the O-/C-/2-μm bands. (b) Measured EO S\u003csub\u003e21\u003c/sub\u003e responses. The insets show normalized optical transmissions with measured half-wave voltage V\u003csub\u003eπ\u003c/sub\u003e.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6748087/v1/cf00cd56d859614eda3053cf.png"},{"id":83601256,"identity":"3b381564-b7ec-4419-9669-14d5ea7aaa19","added_by":"auto","created_at":"2025-05-29 09:17:46","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1085280,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Experimental setup for measuring the eye diagrams and the calculated eye diagrams for OOK signals. (b) Measured BER curves versus the received optical power of PAM-4 signals for λ=1310/1550/2000 nm. The insets show the calculated eye diagrams.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6748087/v1/fa2bddd97a97373094125d40.png"},{"id":83600841,"identity":"c9a35398-12fa-4778-85ea-fe7bef54fd38","added_by":"auto","created_at":"2025-05-29 09:09:46","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":248920,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of 3-dB bandwidth of short-wave infrared high-speed EO modulators.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-6748087/v1/3200b0a8aae4aaca54fb68f6.png"},{"id":101480730,"identity":"8e75e12a-3a3f-406a-b2c0-ed00623103a5","added_by":"auto","created_at":"2026-01-30 08:06:30","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4942676,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6748087/v1/a9ca7b0e-636d-4ebd-8918-509cb86da643.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Ultra-broadband near- to mid-infrared electro-optic modulator on thin-film lithium niobate","fulltext":[{"header":"Main","content":"\u003cp\u003eThe relentless growth of big data, cloud computing, and artificial intelligence has escalated global data traffic to unprecedented levels, challenging the capacity limits of conventional single-mode fiber (SMF)-based optical communication systems [1]. To address this, two complementary strategies have emerged: maximizing the utilization of existing spectral resources and exploring new optical communication windows beyond traditional O- and C-bands [2, 3]. Recent advances in hollow-core photonic bandgap fibers (HC-PBGFs) have unveiled ultra-broad low-loss transmission windows spanning 1240\u0026ndash;1940 nm [4, 5], while innovations in amplifier technologies\u0026mdash;ranging from rare-earth-doped fibers to the newly demonstrated integrated optical parametric amplifiers\u0026mdash;deliver broadband gain across these extended spectral regions [6\u0026ndash;8]. Together, these developments enable a transformative ultra-wide communication spectrum that seamlessly integrates conventional telecom bands with the emerging 2-\u0026mu;m waveband. Beyond revolutionizing optical networks, this expanded spectral range also holds promise for applications such as quantum photonics, precision metrology, and biomedical imaging [9\u0026ndash;11]. However, the realization of end-to-end optical systems spanning these wavelengths remains hindered by a critical gap: the absence of high-speed broadband optical transmitters and photodetectors (PDs).\u003c/p\u003e\n\u003cp\u003eWhile integrated PDs leveraging germanium or germanium tin (GeSn) on silicon photonic platforms have achieved 40\u0026ndash;50 GHz bandwidths up to the 2-\u0026mu;m band [12\u0026ndash;14], high-speed modulators\u0026mdash;the core of optical transmitters\u0026mdash;remain constrained by narrow optical bandwidths. Existing integrated platforms, such as silicon or germanium modulators based on free-carrier effects [15\u0026ndash;19] and thin-film lithium niobate (TFLN) devices utilizing the Pockels effect [20], suffer from waveguide dispersion and velocity mismatch at longer wavelengths, resulting in limited optical bandwidths. These limitations restrict current integrated modulators to single-waveband operation, impeding their utility for multi-band systems. TFLN has emerged as a uniquely promising platform to overcome these barriers. Its combination of ultra-wide optical transparency (visible to mid-infrared), high EO coefficients, low optical loss, and compatibility with CMOS voltages enables modulators with unmatched bandwidth and efficiency [21\u0026ndash;25]. Recent milestones include TFLN-based coherent transmission at 1.96 Tb/s [26] and compact packaged modules [27], underscoring its potential for next-generation networks. Although previous TFLN modulators have operated across discrete wavelengths from visible to 2-\u0026mu;m bands [20, 28\u0026ndash;31], achieving seamless broadband operation\u0026mdash;critical for emerging multi-band systems\u0026mdash;remains an unresolved challenge. This requires overcoming the intrinsic optical bandwidth limitations of conventional device designs.\u003c/p\u003e\n\u003cp\u003eHere, we demonstrate a high-speed TFLN electro-optic modulator with a record-breaking continuous operational range of 1260\u0026ndash;2050 nm, seamlessly bridging the O-, C-, and 2-\u0026mu;m bands. By integrating adiabatic mode evolution engineering with the TFLN platform, our device achieves a 3-dB EO bandwidth exceeding 67 GHz in O/C-bands and 48 GHz in the 2-\u0026mu;m band\u0026mdash;the highest reported for this spectral region. The modulator exhibits exceptional efficiency, with V\u003csub\u003e\u0026pi;\u003c/sub\u003e\u0026middot;L values of 1.92, 2.61, and 3.94 V\u0026middot;cm at 1310, 1550, and 2000 nm, respectively. Experimental validation using a packaged device demonstrates error-free transmission of 108/115/100 Gbaud On-Off Keying (OOK) and 85/95/65 Gbaud 4-level Pulse Amplitude Modulation (PAM-4) signals across all three bands. This breakthrough paves the way for universal optical transmitters capable of unifying fragmented spectral resources into a single ultra-broadband communication infrastructure, with transformative potential for next-generation optical networks, quantum systems, and sensing technologies.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eD\u003c/strong\u003e\u003cstrong\u003eesign of the broadband TNLN modulator\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs shown in Fig. 1(a), the proposed extended full-spectrum optical communication system leverages TFLN technology to realize broadband optical transmitters spanning the O-band to the 2-\u0026mu;m wavelength range. This approach integrates the TFLN platform (NanoLN) with existing multi-wavelength laser sources and multi-band optical amplifiers. At the transmitter, a single full-spectrum TFLN EO modulator converts electrical signals\u0026mdash;including OOK and advanced modulation formats such as PAM-4\u0026mdash;into optical signals across the entire spectral range from the O-band to the 2-\u0026mu;m band. The EOMs are fabricated on a 300-nm-thick x-cut TFLN layer bonded to a 4.7-\u0026mu;m thermally grown silicon dioxide substrate with a silicon base. As illustrated in the inset of Fig. 1(a), the waveguide architecture features a 120-nm slab height and a sidewall angle of \u0026theta; = 60\u0026deg;. Figure 1(b) displays the microscopy images of the fabricated EO modulator, which comprises two 3-dB power splitters, a 9-mm-long EO Mach-Zehnder interferometer modulation section, and a thermo-optic phase shifter, and two spot-size converters (SSCs) as the edge coupler for light coupling. The SEM images of an SSC and a power splitter are shown in the inset of Fig. 1(b).\u003c/p\u003e\n\u003cp\u003eTo achieve ultra-broadband operating bandwidth, the edge coupler was designed based on broadband SSC, which is illustrated in Fig. 2(a) and (b). The structure comprises LN bi-layer tapers and a silicon oxynitride (SiON) ridge waveguide. At the chip edge, the SiON waveguide interfaces with an ultra-high numerical-aperture fiber (UHNAF), which exhibits mode field diameters (MFDs) of 3.3 \u0026micro;m (1310 nm), 4.0 \u0026micro;m (1550 nm), and 5.3 \u0026micro;m (2000 nm). The fundamental transverse electric (TE) mode transitions adiabatically from the SiON waveguide to the LN waveguide via tapered thickness modulation. The refractive index of SiON (n = 1.54) lies between that of LN (n \u0026asymp; 2.2) and the silicon dioxide buried oxide (SiO₂ BOX, n \u0026asymp; 1.45), enabling optical confinement within the waveguide while suppressing leakage into the BOX. Achieving low coupling loss (\u0026lt;1 dB) across the 1310\u0026ndash;2000 nm spectral range requires precise matching of the SiON waveguide and fiber MFDs. For a SiON rib waveguide with a rib height of 3.2 \u0026micro;m and slab height of 1 \u0026micro;m. Fig. 2(c) shows the simulated\u0026nbsp;coupling loss as a function of rib width at \u0026lambda; = 1310, 1550, and 2000 nm. The simulations reveal an optimal rib width of 4.4 \u0026micro;m, where the mode-field overlap peaks at 1550 nm (C-band) and gradually decreases at 1310 nm (O-band) and 2000 nm. This design also demonstrates robust tolerance to dimensional variations in the SiON waveguide cross-section.\u003c/p\u003e\n\u003cp\u003eTable 1. Parameters of the SSC\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"590\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.1111%;\"\u003e\n \u003cp\u003eParameter\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003ew\u003csub\u003e1\u003c/sub\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003ew\u003csub\u003e2\u003c/sub\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003ew\u003csub\u003e3\u003c/sub\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003ew\u003csub\u003e4\u003c/sub\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003ew\u003csub\u003e5\u003c/sub\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003ew\u003csub\u003e6\u003c/sub\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003es\u003csub\u003e1\u003c/sub\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003es\u003csub\u003e2\u003c/sub\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003es\u003csub\u003e3\u003c/sub\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003eh\u003csub\u003e1\u003c/sub\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003eh\u003csub\u003e2\u003c/sub\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 11.1111%;\"\u003e\n \u003cp\u003eValue(\u0026mu;m)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003e0.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003e0.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003e2.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003e0.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003e4.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003e250\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003e150\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003e3.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.08081%;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eThe SSC parameters were optimized through a dual focus: maximizing the taper length to ensure adiabatic mode evolution and minimizing the tip width to enhance coupling efficiency, while maintaining fabrication feasibility. The finalized SSC parameters are summarized in Table 1. Figure 2(d) presents the calculated transmission spectrum across 1200\u0026ndash;2100 nm, with transmission efficiencies of 91.4%, 92.2%, and 88.4% at 1310 nm, 1550 nm, and 2000 nm, respectively. Minor spectral ripples below 1500 nm originate from reflection-induced interference between the LN slab and rib waveguide, as revealed by mode field simulations. Notably, the simulation framework omitted intrinsic optical absorption losses in plasma-deposited silicon oxynitride (SiOxNy) near 1510 nm and 2000 nm\u0026mdash;a factor that may influence experimental broadband performance.\u003c/p\u003e\n\u003cp\u003eThe full spectrum 3-dB power splitter is based on TFLN adiabatic waveguide, with light is adiabatically coupled from the input waveguide to two symmetric output waveguides through three tapers. Further design details of the 3-dB power splitter are provided in Ref. [32]. For shorter wavelengths (e.g., O-band), waveguides thicker than 300 nm impose stricter adiabatic mode evolution requirements, which would degrade the 3-dB power splitter performance. While the reduced electro-optic interaction at this thickness increases the half-wave voltage, this trade-off was essential to achieve a functional bandwidth spanning 1260\u0026ndash;2050 nm. As shown in Fig. 2(e), the calculated transmission for each splitter is 48.2%, 49.6%, and 44.6% at 1310, 1550, and 2000 nm, respectively, confirming robust performance across the targeted spectrum.\u003c/p\u003e\n\u003cp\u003eOutside the modulation region, the waveguides are engineered to support single-mode operation at 1310 nm, ensuring broadband compatibility across the O-/C- and 2-\u0026mu;m bands. This\u0026nbsp;design suppresses higher-order mode excitation over the 1260\u0026ndash;2050 nm\u0026nbsp;range, thereby preventing EO performance degradation in the modulation region. Finite-difference eigenmode (FDE) simulations indicate that single-mode operation requires a waveguide top width of ~0.8 \u0026mu;m. However, this narrow width reduces optical mode confinement, increasing propagation losses due to stronger mode overlap with waveguide sidewalls and cladding, particularly at longer wavelengths. Additionally, the constrained geometry elevates the half-wave voltage requirement. To balance these trade-offs, we implement an adiabatic taper proximal to the 3-dB power splitter, expanding the waveguide width to W = 3.5 \u0026mu;m within the modulation region.\u0026nbsp;A FDE solver was employed to numerically analyze the coupled electrical-optical modal properties of the device architecture. The electrode gap (G) is defined as the vertical separation between the ground plane and the signal electrode. As shown in Fig. 2(f), we calculated the half-wave voltage-length product (V\u003csub\u003e\u0026pi;\u003c/sub\u003e\u0026middot;L)\u0026nbsp;and quantified the metal-induced absorption loss, both parameterized as functions of G at wavelengths of 1310, 1550, and 2000 nm. The results reveal an inverse relationship between device modulation efficiency (V\u003csub\u003e\u0026pi;\u003c/sub\u003e\u0026middot;L) and optical loss performance, necessitating tailored selection of G to prioritize either metric for specific applications. For this work, an electrode gap of G = 6 \u0026mu;m was chosen to achieve an optimal trade-off point, simultaneously ensuring strong modulation performance while limiting absorption loss to moderate levels.\u003c/p\u003e\n\u003cp\u003eThe traveling-wave electrodes feature a push-pull configuration based on a coplanar waveguide (CPW) structure. The signal and ground electrode widths are designed as W\u003csub\u003eS\u003c/sub\u003e = 25 \u0026mu;m and W\u003csub\u003eG\u003c/sub\u003e = 150 \u0026mu;m, respectively, with gold (Au) electrodes of height H = 1 \u0026mu;m and a 2-\u0026mu;m-thick SiO₂ cladding layer. As illustrated in the inset of Fig. 2(g), the characteristic impedance of the CPW is set to 42 \u0026Omega;, slightly lower than the system\u0026rsquo;s characteristic impedance, to mitigate the low-frequency S\u003csub\u003e21\u003c/sub\u003e roll-off and enhance the operational bandwidth. The optical group indices are 2.18, 2.13, and 2.04 at wavelengths (\u0026lambda;) of 1310 nm, 1550 nm, and 2000 nm, respectively (dashed lines in Fig. 2(g)). The optical group index at the C-band is aligned with the radio frequency (RF) effective index, resulting in a trade-off in electro-optic (EO) bandwidth performance for the 2-\u0026mu;m band. The RF effective index is designed as 2.13 at 50 GHz. For \u0026lambda; = 2000 nm, the residual index mismatch leads to a theoretical 3-dB bandwidth of approximately 80 GHz for a 9-mm-long, impedance-matched, lossless modulator.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMeasurement and analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe fabrication details of TFLN modulator is given in the \u0026lsquo;Methods\u0026rsquo;. The coupling losses of the SSC were characterized using a reference waveguide coupled to a UHNA4 fiber. Refractive index-matching oil (index = 1.46) was applied at the fiber-chip interface to minimize reflections. Coupling losses were measured using tunable lasers for the 1260\u0026ndash;1360 nm and 1500\u0026ndash;1630 nm bands, and a custom-built 2-\u0026mu;m amplified spontaneous emission (ASE) source paired with an optical spectrum analyzer (Yokogawa AQ6375B) for the 1920\u0026ndash;2060 nm range. Using a polarization controller (PC), the measured TE mode coupling losses are 0.78 dB/facet at 1310 nm, 0.69 dB/facet at 1550 nm, and 0.82 dB/facet at 1970 nm, as shown in Fig.\u0026nbsp;3(a). The slightly higher losses at 1310 nm and 1970 nm are attributed to minor mode mismatch between the UHNA4 fiber and the SiON waveguide. Notably, two pronounced loss peaks near 1510 nm and 2000 nm correlate with intrinsic N\u0026ndash;H infrared absorption bands in the plasma-deposited SiON layer, consistent with prior reports of N\u0026ndash;H bond absorption [33\u0026ndash;36]. Following characterization of the modulator\u0026rsquo;s total optical losses, the on-chip insertion losses (ILs) were deduced as 1.2 dB, 2.8 dB, and 5.8 dB at wavelengths of 1310 nm, 1550 nm, and 1970 nm, respectively. A test structure comprising 16 cascaded 3-dB power splitters was used to measure the ILs of individual splitters, yielding values of 0.23 dB, 0.11 dB, and 0.42 dB at 1310 nm, 1550 nm, and 1970 nm,\u0026nbsp;closely matching simulations.\u0026nbsp;After accounting for contributions from the power splitters and waveguide propagation losses, the primary source of on-chip loss was identified as absorption by the metal electrodes, which becomes more pronounced at longer wavelengths (e.g., 1970 nm). Experimental on-chip losses for a gap spacing (\u003cem\u003eG\u003c/em\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e= 6 \u0026mu;m) show strong agreement with simulated values, as illustrated by the dashed lines in Fig. 2(f).\u003c/p\u003e\n\u003cp\u003eFig. 3(c) illustrates the half-wave voltage\u0026nbsp;V\u003csub\u003e\u0026pi;\u003c/sub\u003e measurements with a 100 kHz triangular voltage sweep. With a modulator length of 9 mm, the V\u003csub\u003e\u0026pi;\u003c/sub\u003e\u0026middot;L values are calculated to be 1.92/2.61/3.94 V\u0026middot;cm for \u0026lambda; = 1310/1550/2000 nm, respectively. Notably, the experimental V\u003csub\u003e\u0026pi;\u003c/sub\u003e\u0026middot;L and loss are consistent with the simulation results in Fig. 2(f) for G = 6 \u0026mu;m, validating the accuracy of our numerical model. This agreement confirms the reliability of our simulations for predicting performance trends across varying values of electrode gap. For instance, the simulations suggest that prioritizing reduced metal absorption loss over minimizing V\u003csub\u003e\u0026pi;\u003c/sub\u003e can be achieved by increasing G slightly to 7 \u0026mu;m would maintain comparable high-frequency EO performance. Additionally, static extinction ratios (ERs) were measured by sweeping the heater voltage, yielding values of ~17, ~15, and ~18 dB at 1310, 1550, and 1970 nm, respectively. The observed wavelength-dependent ER variations primarily originate from power imbalance between the output ports of the 3-dB splitter, attributed to fabrication tolerances.\u003c/p\u003e\n\u003cp\u003eThe EO responses of the fabricated TFLN modulator were characterized using a 67-GHz vector network analyzer (VNA, Keysight N5247B). A high-speed RF probe was employed to deliver RF signals from the VNA, while a second probe provided a 50 \u0026Omega; termination. The EO response was captured via electrical signals received from the PD into the VNA. For O- and C-band measurements, commercial PDs were used to record the EO S\u003csub\u003e21\u003c/sub\u003e curves (Fig. 3c). The results indicate that the modulator\u0026apos;s 3-dB bandwidth exceeds 67-GHz, which is the limitation of the VNA. For the 2-\u0026mu;m band, where commercial PDs lack sufficient bandwidth, we employed a recently demonstrated high-speed GeSn PD (bandwidth \u0026gt;40 GHz, Ref. [12]) to measure the bandwidth. After system calibration and PD-response de-embedding, characterization of the frequency response and high-speed data transmission capabilities was achieved. The measured EO S\u003csub\u003e21\u003c/sub\u003e response at \u0026lambda; = 2 \u0026mu;m (Fig. 3c) reveals a 3-dB bandwidth of ~48 GHz, the highest reported to date for optical modulators operating in the 2-\u0026mu;m band.\u003c/p\u003e\n\u003cp\u003eTo evaluate high-speed data transmission performance, we characterized the TFLN modulator in a butterfly package across three wavelength regimes: O-band (1310 nm), C-band (1550 nm), and 2-\u0026mu;m band. Fig. 4(a) illustrates the experimental setup for measuring eye diagrams and bit error rate (BER). Optical signals were generated using three laser sources at 1310 nm, 1550 nm, and 2000 nm. High-frequency RF signals, synthesized by a 120 GS/s arbitrary waveform generator (AWG, Keysight M8194A), were amplified to ~2.4 Vpp using an RF amplifier and applied to the modulator. The bias point of the modulator was stabilized via an integrated heater.\u003c/p\u003e\n\u003cp\u003eFor O- and C-band measurements, the modulated optical output was amplified using an O-band semiconductor optical amplifier (SOA) or erbium-doped fiber amplifier (EDFA), respectively, before detection by commercial PDs. For the 2-\u0026mu;m band, the modulated light was amplified using a thulium-doped fiber amplifier (TDFA) and coupled into the GeSn PD, described in prior characterization. The inset of Fig. 4(a) displays measured eye diagrams at the maximum baud rates achieved while maintaining BER below the hard decision forward error correction (HD-FEC) threshold of 3.8\u0026times;10⁻\u0026sup3;. Clear eye openings were obsered for 108/115/100 Gbaud OOK signals at 1310/1550/2000 nm, respectively. To our knowledge, this represents the highest OOK transmission rate in the 2-\u0026mu;m band and the first packaged TFLN modulator capable of exceeding 100 Gbaud OOK operation across all three wavelength bands. We further evaluated PAM-4 performance by measuring BER curves as a function of received optical power (ROP), defined as the optical power incident on the PD. A variable optical attenuator (VOA) was inserted after the optical amplifiers to adjust ROP. Fig. 4(b) shows back-to-back (B2B) BER curves for the TFLN modulator. For 1310 nm and 1550 nm, transmission results through a 500-m standard SMF were also included and exhibit negligible power penalties compared to the B2B case. Clear eye openings for 80/95/65 Gbaud PAM-4 signals were obtained at 1310/1550/2000 nm, corresponding to net data rates of 160/190/130 Gbps. Despite the bandwidth limitations from the RF port packaging (1.85 mm aperture) might potentially constrain the modulator for higher speed data transmission, it is worth noting that the 130 Gbps achieved for the 2-\u0026mu;m band is the highest net bit rate reported to date for this wavelength regime [18].\u003c/p\u003e"},{"header":"Discussion and conclusion","content":"\u003cp\u003eFig. 5 compares the 3-dB EO bandwidths of integrated modulators in the short-wave infrared spectrum regime from 1.2\u0026ndash;2.2 \u0026mu;m. Our TFLN modulator operates across an unprecedented 800-nm optical bandwidth (1260\u0026ndash;2050 nm), spanning the O-band (1260\u0026ndash;1360 nm), C-band (1530\u0026ndash;1565 nm), and extending into the 2-\u0026mu;m regime (1900\u0026ndash;2050 nm). This broad operational range enables multi-band compatibility within a single device, a critical advantage for versatile photonic systems. The TFLN modulator maintains a flat frequency response across its operational range. The 48-GHz bandwidth at 2000 nm enables, to our knowledge, demonstration of the highest single-lane 100 Gb/s OOK data transmission in this band. This represents a 2.3-fold improvement in bandwidth over existing 2-\u0026mu;m modulators, directly enhancing data capacity and underscoring the potential for next-generation high-speed communication systems. The detailed comparasions of the state of art integrated EO modulators operating in the 2-\u0026mu;m spectral band are given in the \u0026lsquo;Methods\u0026rsquo;.\u003c/p\u003e\n\u003cp\u003eIn conclusion, we demonstrate a TFLN modulator featuring high EO bandwidth and a record-breaking operational range spanning over 800 nm, from the O-band to the 2-\u0026mu;m spectral region. The device achieves 3-dB EO bandwidths of over 67 GHz in both the O- and C-bands and 48 GHz in the 2-\u0026mu;m band. High speed data transmisions of 108 Gbaud, 115 Gbaud, and 100 Gbaud OOK signals were experimentally validated at 1310 nm, 1550 nm, and 2000 nm wavelengths, with BERs below the HD-FEC threshold of 3.8 \u0026times; 10⁻\u0026sup3;. These results underscore the transformative potential of TFLN modulators in enabling ultrabroadband optical communication systems that seamlessly bridge conventional telecom bands with the emerging 2-\u0026mu;m window. This advancement directly addresses the escalating bandwidth demands of next-generation data centers and high-performance computing infrastructures.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003eFabrication of the TNLF modulator\u003c/h2\u003e \u003cp\u003eThe fabrication process of the TNLF modulator is outlined as follows: First, electron beam lithography (EBL) was used to define ridge waveguide structures on a 700-nm-thick layer of AR-P 6200 resist. The ridge waveguide was then formed by etching 180 nm of lithium niobate (LN) using Ar⁺-based inductively coupled plasma (ICP) dry etching. A 2-\u0026micro;m-thick silica cladding layer was deposited over the waveguide via plasma-enhanced chemical vapor deposition (PECVD) to encapsulate the modulation section. This silica layer was selectively etched using ICP dry etching in the electrode and spot-size converter (SSC) regions. To create the LN bi-layer tapers, the EBL and ICP etching steps were repeated to remove the remaining 120 nm of LN. A 200-nm-thick thermal titanium (Ti) layer was deposited by electron-beam evaporation (EBE), after which the heater pattern was defined via EBL. A 1-\u0026micro;m-thick Au traveling-wave electrode was then formed by EBE deposition and lift-off. Next, a 4.2-\u0026micro;m-thick SiON layer was grown across the entire chip using PECVD with silane, ammonia, and nitrous oxide precursors. The SiON layer was patterned and etched twice to form the SiON ridge waveguide in the SSC region, while the SiON above the modulation section was removed. A 2-\u0026micro;m-thick protective SiO₂ layer was deposited by PECVD to shield the SSC and minimize contamination-induced losses in the modulator. Finally, the cladding thickness in the modulation section was reduced to 2 \u0026micro;m via ICP etching, and the electrode pads were exposed.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eComparisons of the 2-\u0026micro;m integrated EO modulators\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e summarizes the performance metrics of integrated electro-optic modulators operating in the 2-\u0026micro;m spectral band. Our TFLN modulator achieves a 48 GHz EO 3-dB bandwidth\u0026mdash;the highest reported value in this spectral region\u0026mdash;enabling a baud rate of 100 Gbaud (100 Gb/s for on-off keying, OOK) and surpassing prior state-of-the-art devices. This advancement establishes a new benchmark for high-speed data transmission in the 2-\u0026micro;m band.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePerformance comparison of on-chip integrated EO modulators operating at 2-\u0026micro;m band\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRef.\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePlatform\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eStructure\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eExtinction\u003c/p\u003e \u003cp\u003eRatio (dB)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3-dB Bandwidth (GHz)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMax. Baud Rate and Signal Format\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eOther wavebands capable\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSOI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMZI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e20 Gbaud OOK\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eC-band\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSOI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMichelson\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e20 Gbaud OOK\u003c/p\u003e \u003cp\u003e15 Gbaud PAM-4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLNOI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMZI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e32 Gbaud OOK\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSOI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMZI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e30 Gbaud OOK\u003c/p\u003e \u003cp\u003e40 Gbaud PAM-4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSOI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMRM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e50 Gbaud OOK\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSOI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRacetrack Ring\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e34 Gbaud OOK\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eThis work\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLNOI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMZI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e100 Gbaud OOK\u003c/p\u003e \u003cp\u003e65 Gbaud PAM-4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eO-/C-bands\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003eData availability\u003c/p\u003e\n\u003cp\u003eAll data are available in the main text or Supplementary Information. The data are available from the corresponding authors upon reasonable request.\u003c/p\u003e\n\u003cp\u003eAcknowledgements\u003c/p\u003e\n\u003cp\u003eThis work was supported by the National Natural Science Foundation of China (62175080, 62075074) and National Major Research and Development Program (2022YFB2803600), the open research fund of Songshan Lake Materials Laboratory (2023SLABFK11). The authors thank the Center of Optoelectronic Micro and Nano Fabrication and Characterizing Facility, Wuhan National Laboratory for Optoelectronics of Huazhong University of Science and Technology for the support in device fabrication.\u003c/p\u003e\n\u003cp\u003eAuthor contributions\u003c/p\u003e\n\u003cp\u003eQ.L., Q.Y., A.P., and L.S. conceived the project. Q.L.\u0026nbsp;and Q.Y. carried out simulations and designed the TFLN modulator. A.P. and C.S. fabricated the modulator. Q.L., S.X., J.C., Y.Z., and G.C. characterized the modulator and analyzed experimental data. J.Z., J.Z.,\u0026nbsp;C.Z., S.Z., and L.S. discussed the results. Q.L. and L.S. wrote the manuscript with contribution from all authors. L.S., N.C., J.X., and M.Z.\u0026nbsp;supervised the whole project.\u003c/p\u003e\n\u003cp\u003eCompeting interests\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eT. Mizuno, and Y. Miyamoto, \u0026ldquo;High-capacity dense space division multiplexing transmission,\u0026rdquo; Opt. Fiber Technol. 35, 108-117 (2017).\u003c/li\u003e\n\u003cli\u003eR. Soref, \u0026ldquo;Enabling 2\u0026thinsp;-\u0026mu;m communications,\u0026rdquo; Nat. Photonics 9, 358-359 (2015).\u003c/li\u003e\n\u003cli\u003eZ. Liu, Y. Chen, Z. Li, B. Kelly, R. Phelan, J. Carroll, T. Bradley, J. P. Wooler, N. V. Wheeler, A. M. Heidt, T. Richter, C. Schubert, M. Becker, F. Poletti, M. N. Petrovich, S. Alam, D. J. Richardson, and R. Slavik, \u0026ldquo;High-capacity directly modulated optical transmitter for 2-\u0026mu;m spectral region,\u0026rdquo; J. Lightwave Technol. 33(7), 1373-1379 (2015).\u003c/li\u003e\n\u003cli\u003eH. Sakr, Y. Hong, T. D. Bradley, G. T. Jasion, J.R.Hayes, H.Kim, I. A. Davidson, E. N. Fokoua, Y. Chen, K. R. H. Bottrill, N. Taengnoi, N. V. Wheeler, P. Petropoulos, D. J. Richardson and F. Poletti, \u0026ldquo;Interband Short Reach Data Transmission in Ultrawide Bandwidth Hollow Core Fiber,\u0026rdquo; J. Lightwave Technol. 38(1), 159-165 (2020).\u003c/li\u003e\n\u003cli\u003eP. J. Roberts, F. Couny, H. Sabert, B. J. Mangan, D. P. Williams, L. Farr, M. W. Mason, and A. Tomlinson, \u0026ldquo;Ultimate low loss of hollowcore photonic crystal fibres,\u0026rdquo; Opt. Express 13(1), 236-244 (2005).\u003c/li\u003e\n\u003cli\u003eZ. Li, A. Heidt, N. Simakov, et al, \u0026ldquo;Diode-pumped wideband thuliumdoped fiber amplifiers for optical communications in the 1800-2050 nm window,\u0026rdquo; Opt. Express 21(22), 26450-26455 (2013).\u003c/li\u003e\n\u003cli\u003eP. Zhao, V. Shekhawat, M. Girardi, Z. He, V. T. -Company, and P. A. Andrekson, \u0026ldquo;Ultra-broadband optical amplification using nonlinear integrated waveguides,\u0026rdquo; Nature 640, 918-923 (2025).\u003c/li\u003e\n\u003cli\u003eN. Kuznetsov, A. Nardi, J. Riemensberger, A. Davydova, M. Churaev, P. Seidler, and T. J. Kippenberg, \u0026ldquo;An ultra-broadband photonic-chip-based parametric amplifier,\u0026rdquo; Nature 639, 928-934 (2025).\u003c/li\u003e\n\u003cli\u003eT. Bruns, T. S. Bischof, D. K. Harris, D. Franke, Y. Shi, L. Riedemann, A. Bartelt, F. B. Jaworski, J. A. Carr, C. J. Rowlands, M. W. B. Wilson, O. Chen, H. Wei, G. W. Hwang, D. M. Montana, I. Coropceanu, O. B. Achorn, J. Kloepper, J. Heeren, P. T. C. So, D. Fukumura, K. F. Jensen, R. K. Jain and M. G. Bawendi, \u0026ldquo;Next-generation in vivo optical imaging with short-wave infrared quantum dots,\u0026rdquo; Nat. Biomed. Eng. 1, 0056 (2017).\u003c/li\u003e\n\u003cli\u003eD. R. Carlson, D. D. Hickstein, W. Zhang, A. J. Metcalf, F. Quinlan, S. Diddams, and S. A. Papp, \u0026ldquo;Ultrafast electro-optic light with subcycle control,\u0026rdquo; Science 361, 1358 (2018).\u003c/li\u003e\n\u003cli\u003eM. Karpiński, M. Jachura, L. J. Wright and B. J. Smith, \u0026ldquo;Bandwidth manipulation of quantum light by an electro-optic time lens,\u0026rdquo; Nat. Photonics 11, 53 (2017).\u003c/li\u003e\n\u003cli\u003eJ. Cui, J. Zheng, Y. Zhu, X. Liu, Y. Wu, Q. Huang, Y. Yang, Z. Liu, Z. Liu, Y. Zuo, and B. Cheng, \u0026ldquo;High-speed GeSn resonance cavity enhanced photodetectors for a 50 Gbps Si-based 2 \u0026mu;m band communication system,\u0026rdquo; Photon. Res. 12(4), 767-773 (2024).\u003c/li\u003e\n\u003cli\u003eY. Zhu, Z. Liu, Z. Liu, Y. Hu, Q. Huang, Y. Yang, X. Liu, T. Men, G. Zhang, J. Zheng, Y. Zuo, and B. Cheng, \u0026ldquo;112 Gbps CMOS-Compatible Waveguide Germanium Photodetector for the 2 \u0026mu;m Wavelength Band with a 3.64 A/W Responsivity,\u0026rdquo; Photon. Res. 12(11), 2373-2378 (2024).\u003c/li\u003e\n\u003cli\u003eJ. Wang, X. Zhang, Y. Li, X. Wang, S. Xiao, Q. Song, Z. He, J. Du, H. K. Tsang, and K. Xu, \u0026ldquo;112 Gb s\u003csup\u003e\u0026minus;1\u003c/sup\u003e Germanium Photodetector at 2 \u0026micro;m Enabled by a 3D Integrated Waveguide Loop,\u0026rdquo; Adv. Mater. Technol. 2401718 (2025).\u003c/li\u003e\n\u003cli\u003eW. Cao, D. Hagan, D. J. Thomson, M. Nedeljkovic, C. G. Littlejohns, A. Knights, S. Alam, J. Wang, F. Gardes, W. Zhang, S. Liu, K. Li, M. S. Rouifed, G. Xin, W. Wang, H. Wang, G. T. Reed, and G. Z. Mashanovich, \u0026ldquo;High-speed silicon modulators for the 2 \u0026mu;m wavelength band,\u0026rdquo; Optica 5(9), 1055-1062 (2018). \u003c/li\u003e\n\u003cli\u003eW. Cao, S. Liu, C. G. Littlejohns, D. J. Thomson, M. Nedeljković, W. Zhang, K. Li, M. Banakar, Y. Tran, X. Yan, H. Du, Z. Ren, F. Gardes, G. T. Reed, and G. Z. Mashanovich, \u0026ldquo;High-speed silicon Michelson interferometer modulator and streamlined IMDD PAM-4 transmission of Mach-Zehnder modulators for the 2 \u0026mu;m wavelength band,\u0026rdquo; Opt. Express 29(10), 14438-14451 (2021).\u003c/li\u003e\n\u003cli\u003eX. Wang, W. Shen, W. Li, Y. Liu, Y. Yao, J. Du, Q. Song, and K. Xu, \u0026ldquo;High-speed silicon photonic Mach-Zehnder modulator at 2 \u0026mu;m,\u0026rdquo; Photon. Res. 9, 535-540 (2021).\u003c/li\u003e\n\u003cli\u003eW. Shen, G. Zhou, J. Du, L. Zhou, K. Xu, and Z. He, \u0026ldquo;High-speed silicon microring modulator at the 2\u0026thinsp;\u0026micro;m waveband with analysis and observation of optical bistability,\u0026rdquo; Photon. Res. 10(3), A35 (2022).\u003c/li\u003e\n\u003cli\u003eX. Wang, J. Wang, Y. Yao, S. Xiao, Q. Song, and K. Xu, \u0026ldquo;Efficient and high-speed coupling modulation of silicon racetrack ring resonators at 2 \u0026micro;m waveband,\u0026rdquo; Opt. Lett. 49(8), 2157-2160 (2024).\u003c/li\u003e\n\u003cli\u003eB. Pan, J. Hu, Y. Huang, L. Song, J. Wang, P. Chen, Z. Yu, L. Liu, and D. Dai, \u0026ldquo;Demonstration of high-speed thin-film lithium-niobate-on-insulator optical modulators at the 2-\u0026mu;m wavelength,\u0026rdquo; Opt. Express 29(12), 17710-17717 (2021). \u003c/li\u003e\n\u003cli\u003eC. Wang, M. Zhang, X. Chen, M. Bertrand, A. Shams-Ansari, S. Chandrasekhar, P. Winzer, and M. Loncar, \u0026ldquo;Integrated lithium niobate electro-optic modulators operating at CMOS-compatible voltages,\u0026rdquo; Nature 562, 101 (2018).\u003c/li\u003e\n\u003cli\u003eH. Feng, T. Ge, X. Guo, B. Wang, Y. Zhang, Z. Chen, S. Zhu, K. Zhang, W. Sun, C. Huang, Y. Yuan and C. Wang, \u0026ldquo;Integrated lithium niobate microwave photonic processing engine,\u0026rdquo; Nature 627, 80-87 (2024).\u003c/li\u003e\n\u003cli\u003eM. He, M. Xu, Y. Ren, J. Jian, Z. Ruan, Y. Xu, S. Gao, S. Sun, X. Wen, L. Zhou, L. Liu, C. Guo, H. Chen, S. Yu, L. Liu, and X. Cai, \u0026ldquo;High-performance hybrid silicon and lithium niobate Mach-Zehnder modulators for 100 Gbit s\u0026minus;1 and beyond,\u0026rdquo; Nat. Photonics 13, 359-364 (2019).\u003c/li\u003e\n\u003cli\u003eY. Zhang, L. Shao, J. Yang, Z. Chen, K. Zhang, K. -M. Shum, D. Zhu, C. H. Chan, M. Loncar, and C. Wang, \u0026ldquo;Systematic investigation of millimeter-wave optic modulation performance in thin-film lithium niobate,\u0026rdquo; Photon. Res. 10, 2380-2387 (2022).\u003c/li\u003e\n\u003cli\u003eG. Chen, Y. Gao, H. -L. Lin, and A. J. Danner, \u0026ldquo;Compact and Efficient Thin-Film Lithium Niobate Modulators,\u0026rdquo; Adv. Photonics Res. 4, 2300229 (2023).\u003c/li\u003e\n\u003cli\u003eM. Xu, Y. Zhu, F. Pittal\u0026agrave;, J. Tang, M. He, W. C. Ng, J. Wang, Z. Ruan, X. Tang, M. Kuschnerov, L. Liu, S. Yu, B. Zheng, and X. Cai, \u0026ldquo;Dual-polarization thin-film lithium niobate in-phase quadrature modulators for terabit-per-second transmission,\u0026rdquo; Optica 9, 61 (2022).\u003c/li\u003e\n\u003cli\u003eS. Makino, S. Takeuchi, S. Maruyama, M. Doi, Y. Ohmori, and Y. Kubota, in Optical Fiber Communication Conference (OFC) (IEEE, 2022), paper M1D.2.\u003c/li\u003e\n\u003cli\u003eD. Renaud, D. R. Assumpcao, G. Joe, A. S. -Ansari, D. Zhu, Y. Hu, N. Sinclair, and M. Loncar, \u0026ldquo;Sub-1 Volt and high-bandwidth visible to near-infrared electro-optic modulators,\u0026rdquo; Nat. Commun. 14, 1496 (2023).\u003c/li\u003e\n\u003cli\u003eS. Xue, Z. Shi, J. Ling, Z. Gao, Q. Hu, K. Zhang, G. Valentine, X. Wu, J. Staffa, U. A. Javid, and Q. Lin, \u0026ldquo;Full-spectrum visible electro-optic modulator,\u0026rdquo; Optica 10, 125 (2023).\u003c/li\u003e\n\u003cli\u003eF. Valdez, V. Mere, X. Wang, and S. Mookherjea, \u0026ldquo;Integrated O- and C-band silicon-lithium niobate Mach-Zehnder modulators with 100 GHz bandwidth, low voltage, and low loss,\u0026rdquo; Opt. Express 31(4), 5273-5289 (2023).\u003c/li\u003e\n\u003cli\u003eX. Fang, F. Yang, X. Chen, Y. Li, and F. Zhang, \u0026ldquo;Ultrahigh-Speed Optical Interconnects With Thin Film Lithium Niobate Modulator,\u0026rdquo; J. Lightwave Technol. 41(4), 1207-1215 (2023).\u003c/li\u003e\n\u003cli\u003eQ. Yi, A. Pan, J. Xia, C. Zeng, and L. Shen, \u0026ldquo;Ultra-broadband 1\u0026thinsp;\u0026times;\u0026thinsp;2 3 dB power splitter using a thin-film lithium niobate from 1.2 to 2 \u0026micro;m wave band,\u0026rdquo; Opt. Lett. 48(20), 5375-5378 (2023).\u003c/li\u003e\n\u003cli\u003eS. E. Krikorian, and M. Mahpour, \u0026ldquo;The identification and origin of N-H overtone and combination bands in the near-infrared spectra of simple primary and secondary amides,\u0026rdquo; Spectrochim. Acta A 29(7), 1233-1246 (1973).\u003c/li\u003e\n\u003cli\u003eL. R. Brown, and J. S. Margolis, \u0026ldquo;Empirical line parameters of NH\u003csub\u003e3\u003c/sub\u003e from 4791 to 5294 cm\u003csup\u003e\u0026minus;1\u003c/sup\u003e,\u0026rdquo; J. Quant. Spectrosc. Radiat. Transfer 56(2), 283-294 (1996).\u003c/li\u003e\n\u003cli\u003eC. A. Beale, A. Wong, and P. Bernath, \u0026ldquo;Infrared transmission spectra of hot ammonia in the 4800-9000 cm\u003csup\u003e\u0026minus;1\u003c/sup\u003e region,\u0026rdquo; J. Quant. Spectrosc. Radiat. Transfer 246, 106911 (2020).\u003c/li\u003e\n\u003cli\u003eD. R. Dadadzhanov, T. A. Vartanyan, and A. Karabchevsky, \u0026ldquo;Lattice Rayleigh Anomaly Associated Enhancement of NH and CH Stretching Modes on Gold Metasurfaces for Overtone Detection,\u0026rdquo; Nanomaterials 10(7), 1265 (2020).\u003c/li\u003e\n\u003cli\u003eM. Samiul Alam, X. Li, M. Jacques, Z. Xing, A. Samani, E. E. -Fiky, P. Koh, and D. V. Plant, \u0026ldquo;Net 220 Gbps/\u0026lambda; IM/DD Transmssion in O-Band and C-Band With Silicon Photonic Traveling-Wave MZM,\u0026rdquo; J. Lightwave Technol. 39, 4270 (2021).\u003c/li\u003e\n\u003cli\u003eC. Han, Z. Zheng, H. Shu, M. Jin, J. Qin, R. Chen, Y. Tao, B. Shen, B. Bai, F. Yang, Y. Wang, H. Wang, Z. Zhang, S. Yu, C. Peng and X. Wang, \u0026ldquo;Slow-light silicon modulator with 110-GHz bandwidth,\u0026rdquo; Sci. Adv. 9, eadi5339 (2023).\u003c/li\u003e\n\u003cli\u003eJ. Shen, Y. Zhang, L. Zhang, J. Li, C. Feng, Y. Jiang, H. Wang, X. Li, Y. He, X. Ji, G. Yin, Y. Tian, X. Xiao, and Y. Su, \u0026ldquo;Highly Efficient Slow‐Light Mach-Zehnder Modulator Achieving 0.21 V\u0026middot;cm Efficiency with Bandwidth Surpassing 110 GHz,\u0026rdquo; Laser Photonics Rev. 18(1), 2200192 (2024).\u003c/li\u003e\n\u003cli\u003eM. Li, L. Wang, X. Li, X. Xiao, and S. Yu, \u0026ldquo;Silicon intensity Mach\u0026ndash;Zehnder modulator for single lane 100 Gb/s applications,\u0026rdquo; Photon. Res. 6, 109 (2018).\u003c/li\u003e\n\u003cli\u003eP. Kharel, C. Reimer, K. Luke, L. He, and M. Zhang, \u0026ldquo;Breaking voltage-bandwidth limits in integrated lithium niobate modulators using micro-structured electrodes,\u0026rdquo; Optica 8, 357 (2021).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"
[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6748087/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6748087/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe escalating capacity limitations of conventional near-infrared telecommunication bands have spurred urgent investigations into full-spectrum optical communication systems spanning from the near-infrared to mid-infrared regimes. This has motivated the development of optical components combining broadband bandwidth with high-speed operation. Thin-film lithium niobate (TFLN) modulators, while exhibiting state-of-the-art performance in voltage-length product, optical loss, and electro-optic (EO) bandwidth at telecommunication wavelengths, face challenges in achieving broad operational bandwidth due to waveguide dispersion and velocity mismatch at longer wavelengths. Here, we present a Mach-Zehnder EO modulator comprising two adiabatic power splitters and a high-bandwidth phase shifter, achieving an unprecedented 800-nm operational bandwidth that covers the full optical fibre communication spectrum and extends into the 2-\u0026micro;m mid-infrared band. The fabricated modulator demonstrates\u0026thinsp;\u0026gt;\u0026thinsp;67 GHz EO bandwidth in both O/C-bands and 48 GHz (detector-limited) in the 2-\u0026micro;m band. This device enables single-lane transmission exceeding 100 Gbaud across O-, C-, and 2-\u0026micro;m bands. Notably, we achieve 100 Gbaud OOK and 65 Gbaud PAM-4 transmission in the 2-\u0026micro;m band - the highest rates reported for this spectral region. This breakthrough establishes TFLN as a compelling platform for multispectral photonics, bridging conventional telecom infrastructure with emerging 2-\u0026micro;m technologies for next-generation full-spectrum optical communications.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e","manuscriptTitle":"Ultra-broadband near- to mid-infrared electro-optic modulator on thin-film lithium niobate","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-29 09:09:41","doi":"10.21203/rs.3.rs-6748087/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"
[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"24bf23a2-14a5-4613-a1c1-f21f44b03c7f","owner":[],"postedDate":"May 29th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":49173127,"name":"Physical sciences/Optics and photonics/Applied optics/Integrated optics"},{"id":49173128,"name":"Physical sciences/Optics and photonics/Applied optics/Optoelectronic devices and components"},{"id":49173129,"name":"Physical sciences/Optics and photonics/Optical materials and structures/Silicon photonics"}],"tags":[],"updatedAt":"2026-01-30T08:06:23+00:00","versionOfRecord":{"articleIdentity":"rs-6748087","link":"https://doi.org/10.1038/s41467-025-67902-2","journal":{"identity":"nature-communications","isVorOnly":false,"title":"Nature Communications"},"publishedOn":"2026-01-08 05:00:00","publishedOnDateReadable":"January 8th, 2026"},"versionCreatedAt":"2025-05-29 09:09:41","video":"","vorDoi":"10.1038/s41467-025-67902-2","vorDoiUrl":"https://doi.org/10.1038/s41467-025-67902-2","workflowStages":[]},"version":"v1","identity":"rs-6748087","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6748087","identity":"rs-6748087","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}
Text is read by the "Ask this paper" AI Q&A widget below.
Extraction quality varies by source — PMC NXML preserves structure
cleanly, OA-HTML may include some navigation residue, and OA-PDF can
have broken hyphenation. The publisher copy
(via DOI)
is the canonical version.