TFLN modulators and transmission lines

Abstract

An electro-optic modulator includes a thin-film lithium niobate (TFLN) waveguide that guides an optical signal, and a transmission line. A signal electrode of the transmission line receives a modulation signal that modulates the optical signal. The signal electrode is formed from a plurality of metal layers, and includes a metal-oxide-metal (MOM) capacitor formed between two metal layers of the plurality of metal layers. A ground electrode of the transmission is formed from one or more metal layers and is disposed adjacent to the signal electrode on a first side of the signal electrode. The TFLN waveguide is positioned between the signal electrode and the ground electrode.

Description

[0001] TFLN MODULATORS AND TRANSMISSION LINES

[0002] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of U. S . Provisional Application No . 63 / 778, 247, filed March 26, 2025, and U. S . Provisional Application No . 63 / 779, 270, filed March 27, 2025, the disclosures of which are incorporated herein by reference in their entirety.

[0003] TECHNICAL FIELD

[0004] The present disclosure relates generally to electrooptic modulators, and more particularly but not exclusively to thin-film lithium niobate transmission line structures for high-speed optical communication systems .

[0005] BACKGROUND

[0006] Electro-optic modulators are used in optical communication systems to convert electrical signals into modulated optical signals for transmission through optical fibers . Such modulators are widely deployed within and between data centers to support high-speed data transmission .

[0007] Coherent optical communication systems employ modulators implemented using travelling wave transmission lines on the arms of a Mach-Zehnder interferometer (MZI ) . Silicon-on-insulator (SOI ) based modulators using p-n junctions have been commonly used for this purpose . However, the demand for higher data rates continues to increase, driving the need for modulators with greater electro-optic bandwidth and improved modulation efficiency.

[0008] Thin-film lithium niobate (TFLN) has emerged as a promising material platform for electro-optic modulatorsdue to its strong Pockels effect, which enables high-speed modulation with low drive voltages .

[0009] SUMMARY

[0010] There is provided in accordance with an embodiment of the present disclosure, an electro-optic modulator comprising a thin-film lithium niobate (TFLN) waveguide configured to guide an optical signal, and a transmission line comprising a signal electrode formed from a plurality of metal layers and configured to receive a modulation signal that modulates the optical signal, the signal electrode comprising a metal-oxide-metal (MOM) capacitor formed between two metal layers of the plurality of metal layers, and a ground electrode formed from one or more metal layers and disposed adj acent to the signal electrode on a first side of the signal electrode . The TFLN waveguide is positioned between the signal electrode and the ground electrode .

[0011] Further in accordance with an embodiment of the present disclosure, the two metal layers of the MOM capacitor are adj acent metal layers .

[0012] Still further in accordance with an embodiment of the present disclosure, the two metal layers of the MOM capacitor are non-adj acent metal layers .

[0013] Additionally in accordance with an embodiment of the present disclosure, one or more intermediate metal layers between the two non-adj acent metal layers are omitted from the signal electrode .

[0014] Moreover in accordance with an embodiment of the present disclosure, the ground electrode comprises the one or more intermediate metal layers .

[0015] Further in accordance with an embodiment of the present disclosure, the electro-optic modulator furthercomprises a second ground electrode disposed adjacent to the signal electrode on a second side of the signal electrode opposite the first side . The ground electrode comprises a first number of metal layers and the second ground electrode comprises a second number of metal layers less than the first number of metal layers .

[0016] Still further in accordance with an embodiment of the present disclosure, the electro-optic modulator further comprises a second ground electrode disposed adjacent to the signal electrode on a second side of the signal electrode opposite the first side . The second ground electrode is disposed at a greater distance from the signal electrode than the ground electrode .

[0017] Additionally in accordance with an embodiment of the present disclosure, the electro-optic modulator further comprises one or more slow-wave structures on one or more metal layers of the plurality of metal layers . The one or more slow-wave structures are configured to provide velocity matching between the modulation signal and the optical signal .

[0018] Moreover in accordance with an embodiment of the present disclosure, the one or more slow-wave structures comprise T-rails .

[0019] Further in accordance with an embodiment of the present disclosure, the one or more slow-wave structures comprise a corrugated coplanar waveguide (C-CPW) structure .

[0020] Still further in accordance with an embodiment of the present disclosure, the MOM capacitor is configured to provide velocity matching between the modulation signal and the optical signal .

[0021] Additionally in accordance with an embodiment of the present disclosure, the MOM capacitor is configured toprovide a separate electrical path for a bias control signal .

[0022] Moreover in accordance with an embodiment of the present disclosure, the electro-optic modulator further comprises a second signal electrode, and the transmission line is configured in a GSGSG arrangement for differential drive operation.

[0023] There is also provided in accordance with another embodiment of the present disclosure, an electro-optic modulator comprising a thin-film lithium niobate (TFLN) waveguide configured to guide an optical signal, a transmission line comprising a signal electrode formed from a plurality of metal layers and configured to receive a modulation signal that modulates the optical signal, a first ground electrode formed from one or more metal layers and disposed adj acent to the signal electrode on a first side of the signal electrode, the first ground electrode having a first structural configuration, the TFLN waveguide being positioned between the signal electrode and the first ground electrode, and a second ground electrode formed from one or more metal layers and disposed adj acent to the signal electrode on a second side of the signal electrode opposite the first side, the second ground electrode having a second structural configuration different from the first structural configuration.

[0024] Further in accordance with an embodiment of the present disclosure, the second structural configuration comprises one or more slow-wave structures .

[0025] Still further in accordance with an embodiment of the present disclosure, the first structural configuration is defined by the one or more metal layers of the first ground electrode, the second structural configuration is definedby the one or more metal layers of the second ground electrode, the one or more metal layers of the first ground electrode consists of a first number of metal layers, and the one or more metal layers of the second ground electrode consists of a second number of metal layers less than the first number of metal layers .

[0026] Additionally in accordance with an embodiment of the present disclosure, the first structural configuration is defined by a first distance between the first ground electrode and the signal electrode, and the second structural configuration is defined by a second distance between the second ground electrode and the signal electrode . The second distance is greater than the first distance .

[0027] Moreover in accordance with an embodiment of the present disclosure, the signal electrode comprises a metal-oxide-metal (MOM) capacitor formed between two metal layers of the plurality of metal layers .

[0028] There is also provided in accordance with still another embodiment of the present disclosure, an electrooptic modulator comprising a thin-film lithium niobate (TFLN) waveguide configured to guide an optical signal, and a transmission line comprising a signal electrode formed from a plurality of electrically connected metal layers and configured to receive a modulation signal that modulates the optical signal, and a ground electrode formed from one or more metal layers and disposed adj acent to the signal electrode . The TFLN waveguide is positioned between the signal electrode and the ground electrode, and the ground electrode comprises a metal-insulator-metal (MIM) capacitor .Further in accordance with an embodiment of the present disclosure, the MIM capacitor includes a capacitor structure inserted between two metal layers of the ground electrode .

[0029] Still further in accordance with an embodiment of the present disclosure, the MIM capacitor comprises a fingertype electrode structure .

[0030] Additionally in accordance with an embodiment of the present disclosure, the electro-optic modulator further comprises a semiconductor waveguide layer disposed beneath the TFLN waveguide .

[0031] Moreover in accordance with an embodiment of the present disclosure, the electro-optic modulator further comprises a second ground electrode formed from a plurality of electrically connected metal layers and disposed adj acent to the signal electrode on a side opposite the ground electrode .

[0032] There is also provided in accordance with still another embodiment of the present disclosure, an electrooptic modulator comprising a thin-film lithium niobate (TFLN) waveguide configured to guide an optical signal, a semiconductor waveguide layer disposed beneath the TFLN waveguide, and a transmission line comprising a signal electrode formed from a plurality of electrically connected metal layers and configured to receive a modulation signal that modulates the optical signal, and a ground electrode formed from one or more metal layers and disposed adj acent to the signal electrode . The TFLN waveguide is positioned between the signal electrode and the ground electrode .

[0033] Further in accordance with an embodiment of the present disclosure, the semiconductor waveguide layer isconfigured to enhance velocity matching between the modulation signal and the optical signal .

[0034] Still further in accordance with an embodiment of the present disclosure, the semiconductor waveguide layer comprises silicon.

[0035] Additionally in accordance with an embodiment of the present disclosure, the signal electrode comprises a metal-oxide-metal (MOM) capacitor formed between two metal layers of the plurality of metal layers .

[0036] Moreover in accordance with an embodiment of the present disclosure, the ground electrode comprises a metalinsulator-metal (MIM) capacitor .

[0037] There is also provided in accordance with still another embodiment of the present disclosure, an electrooptic modulator comprising a thin-film lithium niobate (TFLN) waveguide configured to guide an optical signal, a plurality of metal layers, a signal electrode formed from a first part of the plurality of metal layers and configured to receive a modulation signal that modulates the optical signal, the signal electrode comprising a metal-oxide-metal (MOM) capacitor formed between two metal layers of the plurality of metal layers, a first ground electrode formed from a second part of the plurality of metal layers and disposed adj acent to the signal electrode on a first side of the signal electrode, the TFLN waveguide being positioned between the signal electrode and the ground electrode, and a second ground electrode formed from a third part of the plurality of metal layers and disposed adj acent to the signal electrode on a second side of the signal electrode opposite the first side .

[0038] Further in accordance with an embodiment of the present disclosure, the first ground electrode has a firststructural configuration, and the second ground electrode has a second structural configuration different from the first structural configuration.

[0039] Still further in accordance with an embodiment of the present disclosure, the first ground electrode comprises a first number of metal layers of the plurality of metal layers and the second ground electrode comprises a second number of metal layers of the plurality metal layers less than the first number of metal layers .

[0040] Additionally in accordance with an embodiment of the present disclosure, the second ground electrode is disposed at a greater distance from the signal electrode than the first ground electrode .

[0041] There is also provided in accordance with still another embodiment of the present disclosure, an apparatus comprising a first thin-film lithium niobate (TFLN) waveguide configured to guide a first optical signal, a second TFLN waveguide configured to guide a second optical signal, and a transmission line arranged in a ground-signal-ground (GSG) configuration. The transmission line comprises a signal electrode formed from one or more metal layers and configured to receive a modulation signal, a first ground electrode formed from one or more metal layers and disposed adj acent to the signal electrode on a first side of the signal electrode, the first TFLN waveguide being positioned between the signal electrode and the first ground electrode at a first distance from the signal electrode, and a second ground electrode formed from one or more metal layers and disposed adj acent to the signal electrode on a second side of the signal electrode opposite the first side, the second TFLN waveguide being positioned between the signal electrode and the second groundelectrode at a second distance from the signal electrode . The second distance is greater than the first distance .

[0042] Further in accordance with an embodiment of the present disclosure, the second distance is approximately ten times the first distance .

[0043] Still further in accordance with an embodiment of the present disclosure, the signal electrode is formed from a plurality of metal layers and comprises a metal-oxide-metal (MOM) capacitor formed between two metal layers of the plurality of metal layers .

[0044] Additionally in accordance with an embodiment of the present disclosure, the first ground electrode has a first structural configuration, and the second ground electrode has a second structural configuration different from the first structural configuration.

[0045] Moreover in accordance with an embodiment of the present disclosure, the second ground electrode is disposed at a greater distance from the signal electrode than the first ground electrode .

[0046] There is also provided in accordance with still another embodiment of the present disclosure, an integrated electro-optic transceiver comprising a die comprising a plurality of metal layers, a thin-film lithium niobate (TFLN) layer bonded to the die, a transmitter section comprising one or more modulators, each modulator comprising a TFLN waveguide associated with the TFLN layer, a signal electrode formed from a first part of the plurality of metal layers and configured to receive a modulation signal that modulates an optical signal propagating through the TFLN waveguide, and a ground electrode formed from a second part of the plurality of metal layers and disposed adj acent to the signal electrode, the TFLN waveguide beingpositioned between the signal electrode and the ground electrode, and a receiver section comprising one or more photodetectors formed in the die .

[0047] Further in accordance with an embodiment of the present disclosure, the die is a silicon photonics (SiPho) die .

[0048] Still further in accordance with an embodiment of the present disclosure, the TFLN layer is bonded to the die by a one-step die-to-wafer or waf er-to-waf er bonding process .

[0049] Additionally in accordance with an embodiment of the present disclosure, the transmitter section comprises one or more modulators .

[0050] Moreover in accordance with an embodiment of the present disclosure, the one or more photodetectors comprise germanium disposed on silicon.

[0051] Further in accordance with an embodiment of the present disclosure, the integrated electro-optic transceiver further comprises one or more heating elements formed in the die and configured for thermal tuning.

[0052] Still further in accordance with an embodiment of the present disclosure, the die further comprises at least one of : i) one or more silicon nitride (SiN) waveguides, or ii) one or more silicon (Si) waveguides .

[0053] Additionally in accordance with an embodiment of the present disclosure, the signal electrode comprises a metal-oxide-metal (MOM) capacitor formed between two metal layers of the plurality of metal layers .

[0054] Moreover in accordance with an embodiment of the present disclosure, the plurality of metal layers is configured to support co-integration with one or more electronic integrated circuits (EICs) comprising aplurality of RF drivers and a plurality of transimpedance amplifiers (TIAs) .

[0055] There is also provided in accordance with still another embodiment of the present disclosure, a method of fabricating an integrated electro-optic transceiver, the method comprising obtaining a die comprising a plurality of metal layers, bonding a thin-film lithium niobate (TFLN) layer to the die, forming one or more modulators in a transmitter section, each modulator comprising a TFLN waveguide associated with the TFLN layer, a signal electrode formed from a first part of the plurality of metal layers, and a ground electrode formed from a second part of the plurality of metal layers, the TFLN waveguide being positioned between the signal electrode and the ground electrode, and forming one or more photodetectors in a receiver section of the die .

[0056] Further in accordance with an embodiment of the present disclosure, the die is a silicon photonics (SiPho) die .

[0057] Still further in accordance with an embodiment of the present disclosure, obtaining the die comprises obtaining a wafer comprising a silicon substrate, one or more silicon photonic device layers, and the plurality of metal layers .

[0058] Additionally in accordance with an embodiment of the present disclosure, bonding the TFLN layer comprises a one-step die-to-wafer or waf er-to-waf er bonding process .

[0059] Moreover in accordance with an embodiment of the present disclosure, the method further comprises forming a metal-oxide-metal (MOM) capacitor in the signal electrode between two metal layers of the plurality of metal layers .Further in accordance with an embodiment of the present disclosure, forming the one or more photodetectors comprises forming germanium on silicon.

[0060] Still further in accordance with an embodiment of the present disclosure, the method further comprises forming one or more heating elements in the die for thermal tuning.

[0061] Additionally in accordance with an embodiment of the present disclosure, bonding the TFLN layer to the die is performed prior to forming the plurality of metal layers .

[0062] Moreover in accordance with an embodiment of the present disclosure, the plurality of metal layers is formed prior to bonding the TFLN layer to the die .

[0063] Further in accordance with an embodiment of the present disclosure, the method further comprises integrating one or more electronic integrated circuits (EICs) in the die through flip-chip bonding.

[0064] Still further in accordance with an embodiment of the present disclosure, each of the one or more EICs includes a plurality of RE drivers and a plurality of transimpedance amplifiers (TIAs) .

[0065] As used herein, the term "electrode" refers to a conductive structure formed from one or more metal layers of a transmission line, such as a signal electrode or a ground electrode . An electrode carries electrical signals and applies an electric field to the TFLN waveguide . Slow-wave structures, such as T-rails and Corrugated Coplanar Waveguide (C-CPW) structures, are extensions of the electrodes that extend laterally from a metal layer of the electrode into the gap between adj acent electrodes . In some contexts, the slow-wave structures are considered part of the electrodes from which they extend. In other contexts, the slow-wave structures are considered separate structuresassociated with the electrodes . Both interpretations are consistent with the embodiments described herein. The terms "signal line" and "signal electrode" are used interchangeably herein, as are the terms "ground line" and "ground electrode" .

[0066] The present disclosure will be more fully understood from the following detailed description of the embodiments thereof, taken together with the drawings in which:

[0067] BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a top view that schematically illustrates an electro-optic modulator with a Ground-Signal-Ground-Signal-Ground (GSGSG) transmission line configuration, in accordance with an embodiment that is described herein;

[0068] FIG. 2 is a cross-sectional view that schematically illustrates a Ground-Signal-Ground (GSG) portion of the transmission line structure of the electro-optic modulator of FIG. 1, in accordance with an embodiment that is described herein;

[0069] FIG. 3 is a cross-sectional view that schematically illustrates a transmission line structure with a metal-oxide-metal capacitor on a signal electrode, in accordance with an embodiment that is described herein;

[0070] FIG. 4 is a cross-sectional view that schematically illustrates a transmission line structure with a metal-oxide-metal capacitor formed between non-adj acent metal layers on a signal electrode, in accordance with an embodiment that is described herein;

[0071] FIG. 5 is a cross-sectional view that schematically illustrates a transmission line structure with slow-wave structures on the electrodes, in accordance with an embodiment that is described herein;FIG. 6A is a top view that schematically illustrates T-rail slow-wave structures extending from opposite sides of an electrode of the transmission line structure of FIG.

[0072] 5, in accordance with an embodiment that is described herein;

[0073] FIG. 6B is a top view that schematically illustrates Corrugated Coplanar Waveguide (C-CPW) slow-wave structures extending from opposite sides of an electrode of the transmission line structure of FIG. 5, in accordance with an embodiment that is described herein;

[0074] FIG. 7 is a circuit diagram that schematically illustrates an electrical model of the transmission line structure of FIGS . 3 and 4, in accordance with an embodiment that is described herein;

[0075] FIG. 8 is a cross-sectional view that schematically illustrates an asymmetric transmission line structure with a second ground electrode shifted relative to the transmission line structure of FIG. 5, in accordance with an embodiment that is described herein;

[0076] FIG. 9 is a cross-sectional view that schematically illustrates an asymmetric transmission line structure similar to FIG. 8 with slow-wave structures removed from the second ground electrode, in accordance with an embodiment that is described herein;

[0077] FIG. 10 is a cross-sectional view that schematically illustrates an asymmetric transmission line structure with fewer metal layers on a second ground electrode, in accordance with an embodiment that is described herein;

[0078] FIG. 11 is a cross-sectional view that schematically illustrates an asymmetric transmission line structure with a second ground electrode shifted relative to thetransmission line structure of FIG. 10, in accordance with an embodiment that is described herein;

[0079] FIG. 12 is a cross-sectional view that schematically illustrates an asymmetric transmission line structure combining the MOM capacitor configuration of FIG. 3 with the asymmetric ground electrode configuration of FIG. 10, in accordance with an embodiment that is described herein;

[0080] FIG. 13 is a cross-sectional view that schematically illustrates a transmission line structure with a semiconductor waveguide layer disposed beneath a TFLN waveguide, in accordance with an embodiment that is described herein;

[0081] FIG. 14 is a cross-sectional view that schematically illustrates a transmission line structure with a metalinsulator-metal capacitor on a ground electrode and a semiconductor waveguide layer disposed beneath a TFLN waveguide, in accordance with an embodiment that is described herein;

[0082] FIG. 15 is a top view that schematically illustrates a finger-type electrode structure for a metal-insulator-metal capacitor, in accordance with an embodiment that is described herein;

[0083] FIG. 16 is a cross-sectional view that schematically illustrates a test structure with two waveguides positioned at different distances from a signal electrode, in accordance with an embodiment that is described herein;

[0084] FIG. 17 is a cross-sectional view that schematically illustrates a test structure similar to FIG. 16 with fewer metal layers on a second ground electrode, in accordance with an embodiment that is described herein;FIG. 18 is a top view that schematically illustrates the test structure of FIG. 16 in a GSG configuration, in accordance with an embodiment that is described herein;

[0085] FIG. 19 is a cross-sectional view that schematically illustrates an integrated die comprising a TFLN modulation region, a heater region, and a photodetector region, in accordance with an embodiment that is described herein; and FIG. 20 is a block diagram that schematically illustrates a die layout of an integrated electro-optic transceiver comprising a transmitter section and a receiver section on the integrated die of FIG. 19, in accordance with an embodiment that is described herein.

[0086] DETAILED DESCRIPTION OF EMBODIMENTS

[0087] OVERVIEW

[0088] Embodiments that are described herein provide electrooptic modulators comprising thin-film lithium niobate (TFLN) waveguides and multi-layer metal transmission line structures, test structures for characterizing such modulators, integrated electro-optic transceivers, and methods of fabricating such transceivers .

[0089] In some embodiments, an electro-optic modulator comprises a TFLN waveguide and a transmission line comprising a signal electrode and a ground electrode . The signal electrode is formed from a plurality of metal layers and comprises a Metal-Oxide-Metal (MOM) capacitor formed between two metal layers of the plurality of metal layers . The TFLN waveguide is positioned between the signal electrode and the ground electrode . In certain embodiments, the two metal layers of the MOM capacitor are adj acent metal layers . In other embodiments, the two metal layers are non-adj acent metal layers, wherein one or moreintermediate metal layers between the two non-adj acent metal layers are omitted from the signal electrode . The MOM capacitor enhances velocity matching between a modulation signal and an optical signal propagating through the TFLN waveguide, thereby improving electro-optic bandwidth. The MOM capacitor also provides a separate electrical path for a bias control signal, enabling precise modulation bias control .

[0090] In certain embodiments, the transmission line comprises slow-wave structures on one or more metal layers of the plurality of metal layers . The slow-wave structures are configured to provide velocity matching between the modulation signal and the optical signal . In certain embodiments, the slow-wave structures comprise T-rails . In other embodiments, the slow-wave structures comprise a Corrugated Coplanar Waveguide (C-CPW) structure .

[0091] In some embodiments, the electro-optic modulator comprises an asymmetric transmission line structure . The transmission line comprises a signal electrode, a first ground electrode having a first structural configuration, and a second ground electrode having a second structural configuration different from the first structural configuration. The TFLN waveguide is positioned between the signal electrode and the first ground electrode . In certain embodiments, the second ground electrode comprises fewer metal layers than the first ground electrode . In other embodiments, the second ground electrode is disposed at a greater distance from the signal electrode than the first ground electrode . The asymmetric configuration reduces microwave loss and improves velocity and impedance matching, thereby improving electro-optic bandwidth. In embodiments where the signal electrode comprises a MOMcapacitor, the asymmetric configuration also mitigates a voltage divider effect introduced by the MOM capacitor, thereby improving modulation efficiency.

[0092] In certain embodiments, the electro-optic modulator comprises a semiconductor waveguide layer disposed beneath the TFLN waveguide . The semiconductor waveguide layer enhances velocity matching between the modulation signal and the optical signal . In certain embodiments, the semiconductor waveguide layer comprises silicon.

[0093] In some embodiments, the electro-optic modulator comprises a ground electrode having a Metal-Insulator-Metal (MIM) capacitor . The MIM capacitor provides a separate electrical path for a bias control signal . In certain embodiments, the MIM capacitor comprises a capacitor structure inserted between two metal layers of the ground electrode . In other embodiments, the MIM capacitor comprises a finger-type electrode structure .

[0094] In certain embodiments, an apparatus for characterizing an electro-optic modulator comprises a first TFLN waveguide and a second TFLN waveguide, and a transmission line arranged in a Ground-Signal-Ground (GSG) configuration. The first TFLN waveguide is positioned at a first distance from a signal electrode, and the second TFLN waveguide is positioned at a second distance from the signal electrode, wherein the second distance is greater than the first distance . This configuration enables efficient characterization of modulator performance while reducing the size of the test structure .

[0095] In some embodiments, an integrated electro-optic transceiver comprises a die comprising a plurality of metal layers, a TFLN layer bonded to the die, a transmitter section, and a receiver section. The transmitter sectioncomprises one or more modulators, each modulator comprising a TFLN waveguide associated with the TFLN layer, a signal electrode formed from a first part of the plurality of metal layers, and a ground electrode formed from a second part of the plurality of metal layers . The TFLN waveguide is positioned between the signal electrode and the ground electrode . The receiver section comprises one or more photodetectors formed in the die . In certain embodiments, the die is a Silicon Photonics (SiPho) die, and the TFLN layer is bonded to the die by a one-step die-to-wafer or waf er-to-waf er bonding process . In certain embodiments, the transmitter section comprises four modulators configured as a coherent In-phase / Quadrature ( IQ) modulator . The TFLN layer is employed exclusively for high-speed electro-optic modulation, while all other optical and optoelectronic functionalities are implemented in the SiPho material stack .

[0096] In certain embodiments, a method of fabricating an integrated electro-optic transceiver comprises obtaining a die comprising a plurality of metal layers, bonding a TFLN layer to the die, forming one or more modulators in a transmitter section, and forming one or more photodetectors in a receiver section of the die . In some embodiments, the bonding is performed before metallization. In other embodiments, the bonding is performed after metallization.

[0097] The disclosed techniques employ multi-layer metal structures compatible with Complementary Metal-Oxide-Semiconductor (CMOS) foundry processes, and in various embodiments leverage such foundry processes . The MOM capacitor on the signal electrode enhances velocity matching between the modulation signal and the optical signal, thereby improving Electro-Optic (EO) bandwidth. Theasymmetric transmission line configurations reduce microwave loss and mitigate the capacitive voltage divider effect introduced by the MOM capacitor, thereby improving modulation efficiency. The integration of TFLN-based modulators with SiPho components on a single die enables a fully integrated optical transceiver that combines highspeed electro-optic modulation with mature silicon photonics device functionality.

[0098] SYSTEM OVERVIEW FIG. 1 is a top view that schematically illustrates an electro-optic modulator 10 with a GSGSG transmission line configuration, in accordance with an embodiment that is described herein.

[0099] Electro-optic modulator 10 comprises an input waveguide 12 that receives an optical signal and directs it to an input splitter 14. Input splitter 14 divides the optical signal into two paths, directing portions of the optical signal into a first TFLN waveguide 16 and a second TFLN waveguide 18. First TFLN waveguide 16 and second TFLN waveguide 18 extend through a modulation region 34 of electro-optic modulator 10.

[0100] Within modulation region 34, a transmission line structure is arranged in a GSGSG configuration comprising a ground electrode 20, a signal electrode 22, a ground electrode 24, a signal electrode 26, and a ground electrode 28. First TFLN waveguide 16 is positioned between ground electrode 20 and signal electrode 22. Second TFLN waveguide 18 is positioned between ground electrode 24 and signal electrode 26.

[0101] Signal electrodes 22 and 26 are configured to receive modulation signals that modulate the optical signalspropagating through TFLN waveguides 16 and 18, respectively. Ground electrodes 20, 24, and 28 provide reference potentials for the transmission line structure . The GSGSG configuration is compatible with differential RF drivers, which provide complementary modulation signals to signal electrode 22 and signal electrode 26. In the illustrated embodiment, the GSGSG configuration supports differential push-pull driving, in which the modulation signals applied to signal electrodes 22 and 26 are of opposite polarity.

[0102] After passing through modulation region 34, first TFLN waveguide 16 and second TFLN waveguide 18 converge at an output combiner 30, which combines the modulated optical signals . The combined optical signal exits electro-optic modulator 10 through an output waveguide 32.

[0103] In the illustrated embodiment, electro-optic modulator 10 is configured as a Mach-Zehnder interferometer (MZI ) . In an MZI configuration, the phase difference between the optical signals in first TFLN waveguide 16 and second TFLN waveguide 18, introduced by the modulation signals, results in amplitude modulation of the output optical signal at output waveguide 32.

[0104] In the illustrated embodiment, electro-optic modulator 10 further comprises a phase shifter 35 located near output combiner 30. Phase shifter 35 comprises a heating element configured to set or adjust the bias point of electro-optic modulator 10 by thermally inducing a phase shift on one of the waveguide arms . It is also noted that although the two optical paths associated with first TFLN waveguide 16 and second TFLN waveguide 18 appear to have equal physical lengths in FIG. 1, in some embodiments, adeliberate path-length difference is introduced between the two arms to create an inherent phase offset .

[0105] In certain embodiments, electro-optic modulator 10 further comprises termination heating elements 37a and 37b connected to the ends of signal electrodes 22 and 26, respectively, near output combiner 30. Termination heating elements 37a and 37b are connected to a voltage node 39, which is connectable to ground or a voltage source . Termination heating elements 37a and 37b provide on-chip termination for the transmission line structure .

[0106] Although the embodiment of FIG. 1 shows a GSGSG transmission line configuration with two signal electrodes 22, 26 and two TFLN waveguides 16, 18, in other embodiments, an electro-optic modulator comprises a GSG transmission line configuration with a single signal electrode and a single TFLN waveguide positioned between the signal electrode and one of the ground electrodes . In the GSGSG configuration, each signal electrode has an active side (where the TFLN waveguide is positioned) and an opposite side that contains dummy structures . For example, for signal electrode 22, the active side is between ground electrode 20 and signal electrode 22 (where TFLN waveguide 16 is positioned) , and the opposite side is between signal electrode 22 and ground electrode 24. The cross-sectional views described below with reference to FIGS . 2-5 and FIGS .

[0107] 8-14 illustrate a GSG portion of the GSGSG configuration of FIG. 1, wherein for the signal electrode, one of the adj acent ground electrodes is on the side where the TFLN waveguide is positioned (the "active" side) , and the other adj acent ground electrode is on the side where no TFLN waveguide is positioned.BASELINE LAYER STACK

[0108] FIG. 2 is a cross-sectional view that schematically illustrates a GSG portion of the transmission line structure of electro-optic modulator 10 of FIG. 1, in accordance with an embodiment that is described herein.

[0109] The GSG portion comprises a ground electrode 40, a signal electrode 42, and a ground electrode 44. Ground electrode 40 is disposed adj acent to signal electrode 42 on a first side of signal electrode 42 (the "active" side) . Ground electrode 44 is disposed adj acent to signal electrode 42 on a second side of signal electrode 42 opposite the first side . A TFLN waveguide 36 is positioned between ground electrode 40 and signal electrode 42. Electrodes 40, 42, and 44 respectively correspond to electrodes 20, 22, and 24 of FIG. 1. TFLN waveguide 36 corresponds to TFLN waveguide 16 of FIG. 1.

[0110] Ground electrode 40, signal electrode 42, and ground electrode 44 are each formed from a plurality of metal layers provided by a common fabrication process, such as a CMOS foundry process . In the illustrated embodiment, the plurality of metal layers comprises a first metal layer Mi, a second metal layer M2, and an nth metal layer Mn, interconnected by via layers Vi through Vn-i . In certain embodiments, the plurality of metal layers comprises three metal layers, i . e . , n=3 .

[0111] In the baseline configuration of FIG. 2, all metal layers of each electrode are electrically connected through the respective via layers, and the metal layers are shared among the electrodes . Signal electrode 42 is configured to receive a modulation signal that modulates an optical signal propagating through TFLN waveguide 36.TFLN waveguide 36 is a waveguide structure that is formed by etching a portion of a TFLN layer 56. In other embodiments, TFLN waveguide 36 is formed on or as an extension of TFLN layer 56. In some embodiments, the etching forms a ridge structure while leaving a remaining portion of TFLN layer 56 beneath the ridge . In other embodiments, the TFLN layer surrounding the ridge is fully etched, such that the waveguide ridge retains substantially the full thickness of the original TFLN layer 56. In the figures, TFLN waveguide 36 is shown as a separate structure from TFLN layer 56 for clarity of illustration. TFLN layer 56 may be one or more layers, for example a plurality of films . TFLN layer 56 is disposed on an oxide layer 58, and oxide layer 58 is disposed on a substrate 60. In certain embodiments, substrate 60 comprises silicon. The metal layers and via layers of electrodes 40, 42, and 44 are embedded in an oxide material (schematically represented in FIG. 2 as backdrop 59) , such as silicon dioxide (SiCh) , that fills the spaces between and around the metal layers . This oxide material 59 serves as an insulating dielectric between adj acent metal layers where via connections are not present . In certain embodiments, oxide material 59 and oxide layer 58 are portions of a continuous oxide structure . In other embodiments, oxide material 59 and oxide layer 58 are separate oxide structures formed at different stages of the fabrication process .

[0112] Parenthetically, throughout this document metal layer Mi is referred to interchangeably as the bottom metal layer, as it is the metal layer closest to the substrate stack ( formed from substrate 60, oxide layer 58, and TFLN layer 56) . Similarly, metal layer Mnis referred to interchangeably as the top metal layer, as it is the metallayer farthest from the substrate stack. In certain embodiments, an additional metal layer, formed from a material different from Mi through Mn, is disposed on top of Mnto serve as electrical probing pads for die-level or wafer-level testing.

[0113] MOM CAPACITOR ON SIGNAL ELECTRODE FIG. 3 is a cross-sectional view that schematically illustrates a transmission line structure with a MOM capacitor on a signal electrode, in accordance with an embodiment that is described herein. Here, signal electrode 42 comprises a Metal-Oxide-Metal (MOM) capacitor 61, having a capacitance CMOM, formed between two adj acent metal layers of the plurality of metal layers . In the illustrated embodiment, MOM capacitor 61 is formed between Mi and M2 by omitting via layer Vi between Mi and M2. With via layer Vi omitted, Mi and M2 of signal electrode 42 function as two metal plates of MOM capacitor 61, and the oxide material between Ml and M2 serves as the dielectric of MOM capacitor 61. The remaining metal layers of signal electrode 42 (M2 through Mn) are electrically connected through the respective via layers .

[0114] Ground electrode 40 and ground electrode 44 retain all metal layers (Mi through Mn) electrically connected through the respective via layers, as in the baseline configuration of FIG. 2.

[0115] In other embodiments, MOM capacitor 61 is formed between other pairs of adj acent metal layers, for example between M2 and M3 by omitting via layer V2 .

[0116] MOM capacitor 61 serves two functions . First, MOM capacitor 61 enhances velocity matching between the modulation signal and the optical signal propagatingthrough TFLN waveguide 36, thereby improving electro-optic bandwidth. In addition, the capacitance introduced by MOM capacitor 61 reduces microwave loss in the transmission line, further improving the electro-optic bandwidth. Second, MOM capacitor 61 provides a separate electrical path for a bias control signal, enabling precise modulation bias control . Bias control is used to set and maintain the operating point of Mach-Zehnder interferometers, ensuring stable modulation performance .

[0117] In yet other embodiments, MOM capacitor 61 is formed between non-adj acent metal layers on the signal electrode . Such an embodiment is schematically illustrated in FIG. 4. Here, signal electrode 42 comprises MOM capacitor 61 formed between Mi and an upper metal layer designated MMOM . One or more intermediate metal layers between Mi and MMOM are omitted from signal electrode 42, along with the corresponding via layers . The metal layers above MMOM (MMOM through Mn) are electrically connected through the respective via layers .

[0118] Ground electrode 40 and ground electrode 44 retain all metal layers (Mi through Mn) electrically connected through the respective via layers, comprising the non-adj acent metal layers and the one or more intermediate metal layers that are omitted from signal electrode 42.

[0119] The non-adj acent MOM capacitor configuration of FIG.

[0120] 4 provides a reduced capacitance CMOM compared to the adj acent MOM capacitor configuration of FIG. 3, due to the increased distance between the capacitor plates . The reduced CMOM improves the electro-optic bandwidth, providing greater bandwidth margin. This bandwidth margin allows designers to extend the length of the modulator toimprove modulation efficiency without sacrificing electrooptic bandwidth.

[0121] In addition to the MOM capacitor configurations described above, in certain embodiments the transmission line structure is further optimized using slow-wave structures, as described below.

[0122] SLOW-WAVE STRUCTURES

[0123] FIG. 5 is a cross-sectional view that schematically illustrates a transmission line structure with slow-wave structures on the electrodes, in accordance with an embodiment that is described herein.

[0124] In the embodiment of FIG. 5, signal electrode 42 comprises MOM capacitor 61 formed between two metal layers, as described above with reference to FIGS . 3 and 4. In addition, the transmission line structure comprises slow-wave structures disposed on one or more metal layers of the electrodes . In the illustrated embodiment, the slow-wave structures are disposed on the Mi layer . The slow-wave structures extend laterally from the electrodes into the gaps between adj acent electrodes .

[0125] Specifically, a slow-wave structure 66 extends from the Mi layer of ground electrode 40 toward signal electrode 42. A slow-wave structure 68 extends from the Mi layer of signal electrode 42 toward ground electrode 40. A slow-wave structure 70 extends from the Mi layer of signal electrode 42 toward ground electrode 44. A slow-wave structure 72 extends from the Mi layer of ground electrode 44 toward signal electrode 42.

[0126] The slow-wave structures 66, 68, 70, and 72 are configured to provide velocity matching between the modulation signal and the optical signal propagatingthrough TFLN waveguide 36, as well as impedance tuning of the transmission line . The slow-wave structures add capacitance to the transmission line, which slows down the modulation signal to better match the propagation velocity of the optical signal in TFLN waveguide 36.

[0127] In the embodiment of FIG. 5, slow-wave structures 66, 68, 70, and 72 are disposed symmetrically on both sides of signal electrode 42. In other embodiments, the slow-wave structures are disposed asymmetrically, as described below with reference to FIGS . 8 and 9.

[0128] In certain embodiments, slow-wave structures 66, 68, 70, and 72 comprise T-rails . T-rails are T-shaped structures that extend laterally from a metal layer (e . g. , the Mi layer) of the respective electrode . FIG. 6A is a top view that schematically illustrates T-rail slow-wave structures extending from the electrodes of the transmission line structure of FIG. 5, in accordance with an embodiment that is described herein. As shown in FIG.

[0129] 6A, T-rail 68 and T-rail 70 extend from opposite sides of signal electrode 42. T-rail 66 extends from ground electrode 40 toward signal electrode 42 and T-rail 72 extends from ground electrode 44 toward signal electrode 42, such that T-rail 66 and T-rail 68 extend toward each other in the gap between ground electrode 40 and signal electrode 42 and T-rail 72 and T-rail 70 extend toward each other in the gap between signal electrode 42 and ground electrode 44. The base of each "T" extends outward into the gap between adj acent electrodes, and the cross part of the "T" acts as the main electrode body.

[0130] In other embodiments, slow-wave structures 66, 68, 70, and 72 comprise a Corrugated Coplanar Waveguide (C-CPW) structure . C-CPW structures are wider rectangularextensions from a metal layer (e . g. , the Mi layer) of the respective electrode, and are elongated in the direction of elongation of the electrodes / waveguides and extend outward into the gap between adj acent electrodes . FIG. 6B is a top view that schematically illustrates C-CPW slow-wave structures extending from the electrodes of the transmission line structure of FIG. 5, in accordance with an embodiment that is described herein. As shown in FIG.

[0131] 6B, C-CPW structure 68 and C-CPW structure 70 extend from opposite sides of signal electrode 42. C-CPW structure 66 extends from ground electrode 40 toward signal electrode 42 and C-CPW structure 72 extends from ground electrode 44 toward signal electrode 42, such that C-CPW structure 66 and C-CPW structure 68 extend toward each other in the gap between ground electrode 40 and signal electrode 42 and C-CPW structure 72 and C-CPW structure 70 extend toward each other in the gap between signal electrode 42 and ground electrode 44 .

[0132] For simplicity of illustration, both T-rails and C-CPW structures are represented schematically in the cross-sectional views as rectangular extensions from the Mi layer .

[0133] The slow-wave structures are disposed on the Mi layer in the illustrate embodiment because Mi is the metal layer closest to TFLN waveguide 36, where the electric field interaction with the optical signal is strongest . In other embodiments, slow-wave structures are disposed on other metal layers in addition to or instead of Mi . For example, in certain embodiments, slow-wave structures are disposed on M2, on both Mi and M2, or on any other combination of metal layers .PARASITIC CAPACITANCE

[0134] FIG. 7 is a circuit diagram that schematically illustrates an electrical model of the transmission line structure of FIGS . 3 and 4, in accordance with an embodiment that is described herein.

[0135] As illustrated in FIG. 7, signal electrode 42 is shown as comprising MOM capacitor 61 having capacitance CMOM . AS mentioned above with reference to FIGS . 3 and 4, one of the functions served by MOM capacitor 61 is that it provides a separate electrical path for a bias control signal . This is shown in FIG. 7, where a bias control signal is applied through a resistive path connected below MOM capacitor 61. MOM capacitor 61 blocks low-frequency bias control signal from the high-frequency modulation signal path, while allowing the modulation signal to pass through.

[0136] In addition to capacitance CMOM, a capacitance CGSL exists between ground electrode 40 and signal electrode 42, and a capacitance CGSR exists between signal electrode 42 and ground electrode 44. Capacitances CGSL and CGSR are parasitic capacitances that arise from the proximity of the electrodes .

[0137] Capacitances CGSL, CGSR, and CMOM form a capacitive voltage divider . The actual modulation voltage Vmat the Mi layer of signal electrode 42 is estimated as :

[0138]

[0139] where CMOM is the capacitance of MOM capacitor 61, and CGSL and CGSR represent the parasitic capacitances between signal electrode 42 and ground electrodes 40 and 44, respectively. The voltage divider effect reduces the actual modulation voltage Vmapplied to TFLN layer 56 (and TFLN waveguide 36) relative to the voltage applied to signalelectrode 42. For example, for a typical cross-section, Vmis estimated to be approximately 0.5, meaning that the half-wave voltage Vn(MOM) is approximately twice the halfwave voltage Vno of a conventional TFLN modulator without a MOM capacitor . In other words, the modulation efficiency is reduced by approximately a factor of two . Techniques for mitigating this voltage divider effect are described below with reference to FIGS . 8-12.

[0140] ASYMMETRIC DESIGNS

[0141] The embodiments described below with reference to FIGS . 8-12 introduce asymmetry into the transmission line structure by configuring ground electrode 44 differently from ground electrode 40. Since ground electrode 44 is on the side without TFLN waveguide 36, modifications to ground electrode 44 do not directly affect the optical modulation . Such modifications provide multiple benefits . First, modifying ground electrode 44 reduces microwave loss in the transmission line and improves velocity and impedance matching, thereby improving electro-optic bandwidth. Second, in embodiments where signal electrode 42 comprises a MOM capacitor 61, modifying ground electrode 44 reduces the parasitic capacitance CGSR, which according to the voltage divider equation described above with reference to FIG. 7, increases the modulation voltage Vmand thereby mitigates the voltage divider effect introduced by MOM capacitor 61. These benefits are independent of each other, and the asymmetric configurations described below are applicable both with and without a MOM capacitor on signal electrode 42 . It is noted, however, that the voltage divider effect and its mitigation are applicable only in embodiments where signal electrode 42 comprises MOMcapacitor 61 or any other suitable structures containing a capacitance .

[0142] FIG. 8 is a cross-sectional view that schematically illustrates an asymmetric transmission line structure with a second ground electrode shifted relative to the transmission line structure of FIG. 5, in accordance with an embodiment that is described herein.

[0143] In the embodiment of FIG. 8, signal electrode 42 comprises MOM capacitor 61, and slow-wave structures 66, 68, 70, and 72 are disposed on the Mi layers of the electrodes, as described above with reference to FIG. 5. Ground electrode 44 is shifted to a greater distance from signal electrode 42 compared to the distance between ground electrode 40 and signal electrode 42. Specifically, a distance Di separates ground electrode 40 from signal electrode 42, and a distance D2 separates ground electrode 44 from signal electrode 42, wherein D2 is greater than Di .

[0144] Shifting ground electrode 44 further from signal electrode 42 reduces microwave loss in the transmission line, thereby improving electro-optic bandwidth. Shifting ground electrode 44 also reduces the parasitic capacitance CGSR between signal electrode 42 and ground electrode 44 , thereby improving modulation efficiency in embodiments where signal electrode 42 comprises MOM capacitor 61.

[0145] FIG. 9 is a cross-sectional view that schematically illustrates an asymmetric transmission line structure similar to FIG. 8 with slow-wave structures removed from the second ground electrode, in accordance with an embodiment that is described herein.

[0146] In the embodiment of FIG. 9, the transmission line structure is identical to that of FIG. 8, except that slow-wave structure 72 is removed from ground electrode 44.Slow-wave structures 66, 68, and 70 remain on the Mi layers of ground electrode 40 and signal electrode 42. Ground electrode 44 is shifted to a greater distance D2 from signal electrode 42, as in FIG. 8. Although not shown, in certain embodiments, slow-wave structure 70 is also removed from signal electrode 42.

[0147] Removing slow-wave structure 72 from ground electrode 44 further reduces microwave loss and improves velocity and impedance matching on the side of the transmission line where no TFLN waveguide is present, thereby further improving electro-optic bandwidth. Removing slow-wave structure 72 also reduces the parasitic capacitance CGSR, thereby improving modulation efficiency in embodiments where signal electrode 42 comprises MOM capacitor 61.

[0148] FIG. 10 is a cross-sectional view that schematically illustrates an asymmetric transmission line structure with fewer metal layers on a second ground electrode, in accordance with an embodiment that is described herein.

[0149] In the embodiment of FIG. 10, signal electrode 42 comprises MOM capacitor 61 formed between two metal layers . Ground electrode 40 retains all metal layers (Mi through Mn) electrically connected through the respective via layers . Ground electrode 44 comprises fewer metal layers than ground electrode 40. In the illustrated embodiment, one or more bottom metal layers and corresponding via layers are removed from ground electrode 44. For example, Mi and Vi are removed, such that ground electrode 44 comprises M2 through Mn. In other embodiments, Mi, Vi, M2, and V2 are removed, such that ground electrode 44 comprises M3 through Mn. In an extreme case, all metal layers except Mnare removed from ground electrode 44, such that ground electrode 44 comprises a single metal layer .Removing metal layers from ground electrode 44 reduces microwave loss, thereby improving electro-optic bandwidth. Removing metal layers from ground electrode 44 also reduces the parasitic capacitance CGSR between signal electrode 42 and ground electrode 44. In embodiments where signal electrode 42 comprises MOM capacitor 61, the reduced CGSR increases the modulation voltage Vm, thereby improving modulation efficiency. For example, removing one or more metal layers from ground electrode 44 improves Vn(MOM) by approximately 20%, and achieves a 3-dB EO bandwidth exceeding 100 GHz .

[0150] FIG. 11 is a cross-sectional view that schematically illustrates an asymmetric transmission line structure with a second ground electrode shifted relative to the transmission line structure of FIG. 10, in accordance with an embodiment that is described herein.

[0151] In the embodiment of FIG. 11, the transmission line structure is identical to that of FIG. 10, except that ground electrode 44 is additionally shifted to a greater distance from signal electrode 42. As in FIG. 8, a distance Di separates ground electrode 40 from signal electrode 42, and a distance D2 separates ground electrode 44 from signal electrode 42, wherein D2 is greater than Di .

[0152] The embodiment of FIG. 11 combines the benefits of removing metal layers from ground electrode 44 (as in FIG.

[0153] 10) with the benefits of shifting ground electrode 44 further from signal electrode 42 (as in FIG. 9) , providing both reduced microwave loss and parasitic capacitance CGSR .

[0154] The asymmetric configurations of FIGS . 10 and 11 are particularly advantageous when combined with the non-adj acent MOM capacitor configuration of FIG. 4. As described above, the non-adj acent MOM capacitorconfiguration reduces CMOM, which improves the electro-optic bandwidth by approximately 50% . However, the reduced CMOM also reduces the modulation voltage Vmby approximately 20% compared to the adj acent MOM capacitor configuration of FIG. 3. By additionally removing metal layers from ground electrode 44 (as in FIGS . 10 and 11 ) , CGSR is reduced, which boosts Vmby up to 30%, more than compensating for the reduction caused by the non-adj acent MOM capacitor . Combined with a longer modulator length (e . g. , approximately 30% longer) , this configuration achieves a Vnas low as 3. 6 V while maintaining a 3-dB EO bandwidth exceeding 100 GHz .

[0155] The various asymmetric configurations described with reference to FIGS . 8-11 are combinable with the MOM capacitor configurations described with reference to FIGS .

[0156] 3 and 4, and with the slow-wave structures described with reference to FIG . 5. For example, FIG. 12 is a cross-sectional view that schematically illustrates an asymmetric transmission line structure combining the MOM capacitor configuration of FIG. 3 with the asymmetric ground electrode configuration of FIG. 10, in accordance with an embodiment that is described herein. In FIG. 12, signal electrode 42 comprises MOM capacitor 61 formed between adj acent metal layers Mi and M2 (as in FIG. 3) , and ground electrode 44 comprises fewer metal layers than ground electrode 40 (as in FIG. 10) . As another example, although the asymmetric configurations of FIGS . 10 and 11 are shown without slow-wave structures for clarity of illustration, in other embodiments, slow-wave structures 66, 68, and optionally 70 and / or 72, are disposed on the Mi layers of the electrodes in combination with the asymmetric configurations of FIGS . 10 and 11.SEMICONDUCTOR WAVEGUIDE

[0157] FIG. 13 is a cross-sectional view that schematically illustrates a transmission line structure with a semiconductor waveguide layer disposed beneath a TFLN waveguide, in accordance with an embodiment that is described herein.

[0158] In the embodiment of FIG. 13, the transmission line structure comprises ground electrode 40, signal electrode 42, and ground electrode 44, with TFLN waveguide 36 positioned between ground electrode 40 and signal electrode 42. A semiconductor waveguide layer 82 is disposed directly beneath TFLN waveguide 36, for example disposed within oxide layer 58 .

[0159] Semiconductor waveguide layer 82 enhances velocity matching between the modulation signal and the optical signal propagating through TFLN waveguide 36. Semiconductor waveguide layer 82 modifies the effective refractive index experienced by the optical signal, thereby adjusting the optical propagation velocity to better match the propagation velocity of the modulation signal along the transmission line .

[0160] In certain embodiments, semiconductor waveguide layer 82 comprises silicon. In other embodiments, semiconductor waveguide layer 82 comprises other semiconductor materials .

[0161] In the embodiment of FIG. 13, all metal layers of each electrode (40, 42, 44 ) are electrically connected through the respective via layers . In other embodiments, signal electrode 42 comprises a MOM capacitor 61 formed between two metal layers, as described above with reference to FIGS . 3 and 4. Semiconductor waveguide layer 82 is combinable with any of the MOM capacitor configurations andasymmetric transmission line configurations described herein .

[0162] Semiconductor waveguide layer 82 is also combinable with a Metal-Insulator-Metal (MIM) capacitor on ground electrode 40, as described below with reference to FIG. 14.

[0163] MIM CAPACITOR ON GROUND ELECTRODE FIG. 14 is a cross-sectional view that schematically illustrates a transmission line structure with a metalinsulator-metal capacitor on a ground electrode and a semiconductor waveguide layer disposed beneath a TFLN waveguide, in accordance with an embodiment that is described herein.

[0164] In the embodiment of FIG. 14, the transmission line structure comprises ground electrode 40, signal electrode 42, and ground electrode 44, with TFLN waveguide 36 positioned between ground electrode 40 and signal electrode 42. Semiconductor waveguide layer 82 is disposed directly beneath TFLN waveguide 36, as described above with reference to FIG. 13.

[0165] Ground electrode 44 comprises a plurality of electrically connected metal layers and is disposed adj acent to signal electrode 42 on a side opposite ground electrode 40. Signal electrode 42 also comprises a plurality of electrically connected metal layers . In contrast to the embodiments of FIGS . 3 and 4, signal electrode 42 in FIG. 14 does not comprise a MOM capacitor, i . e . , all metal layers of signal electrode 42 are electrically connected through the respective via layers .

[0166] In contrast to ground electrode 40 of the previous embodiments, in the embodiment illustrated in FIG. 14 ground electrode 40 comprises a Metal-Insulator-Metal (MIM)capacitor 84. MIM capacitor 84 provides a separate electrical path for a bias control signal, enabling precise modulation bias control for the Mach-Zehnder interferometer . In the illustrated embodiment, MIM capacitor 84 comprises a capacitor structure inserted between two metal layers of ground electrode 40. The inserted capacitor structure comprises two metal plates with an insulating dielectric (e . g. , silicon dioxide) therebetween, positioned between two of the metal layers (e . g. , Mi and M2) of ground electrode 40.

[0167] In other embodiments, MIM capacitor 84 comprises a finger-type electrode structure . FIG. 15 is a sectional view (corresponding to perpendicular view to the cross-sectional views in the previous figures) that schematically illustrates a finger-type electrode structure for MIM capacitor 84, in accordance with an embodiment that is described herein. As shown in FIG. 15, MIM capacitor 84 comprises a first electrode portion 86 and a second electrode portion 88 arranged in an interdigitated configuration. The interdigitated configuration is within one or more metal layers of ground electrode 40. First electrode portion 86 comprises a plurality of fingers extending in a first direction, and second electrode portion 88 comprises a plurality of fingers extending in a second direction opposite the first direction. The fingers of first electrode portion 86 and second electrode portion 88 are interleaved within the same metal layer, creating alternating parallel conductive elements separated by gaps filled with the oxide material . The capacitance of MIM capacitor 84 is determined by the lateral fringe electric field between the interleaved fingers, and is proportional to the number of fingers and their length. In certainembodiments, the finger pattern is replicated across multiple metal layers of ground electrode 40 and connected by vias to increase the overall capacitance .

[0168] Although FIG. 14 shows semiconductor waveguide layer 82 in combination with MIM capacitor 84, in other embodiments, MIM capacitor 84 is used without semiconductor waveguide layer 82. Similarly, semiconductor waveguide layer 82 is usable without MIM capacitor 84, as described above with reference to FIG. 13. It is also noted that although FIG. 14 shows MIM capacitor 84 on ground electrode 40 (the active side adj acent to TFLN waveguide 36) and FIG.

[0169] 15 was described as MIM capacitor 84 being within one or more metal layers of ground electrode 40, in other embodiments, MIM capacitor 84 is disposed on ground electrode 44 or on both ground electrodes 40 and 44.

[0170] TEST STRUCTURE AND COMPACT DOE

[0171] By way of introduction, electro-optic modulators such as electro-optic modulator 10 of FIG. 1 occupy approximately 1 cm in length and 0.5 mm in width. For a die size of 1 cm x 1 cm, only approximately 20 design-of-experiment (DOE) structures using a full GSGSG configuration are accommodated. The embodiments described below with reference to FIGS . 16-18 address this challenge by providing a compact test structure that reduces the width of each DOE structure while enabling accurate characterization of modulator performance .

[0172] FIG. 16 is a cross-sectional view that schematically illustrates a test structure with two waveguides positioned at different distances from a signal electrode, in accordance with an embodiment that is described herein. Here the transmission line structure is arranged in a GSGconfiguration comprising ground electrode 40, signal electrode 42, and ground electrode 44. First TFLN waveguide 36 is positioned between ground electrode 40 and signal electrode 42 at a first distance GL from signal electrode 42. A second TFLN waveguide 38 is positioned between signal electrode 42 and ground electrode 44 at a second distance GR from signal electrode 42. Second distance GR is greater than first distance GL . In certain embodiments, the second distance GR is much greater than first distance GL (e . g. , an order of magnitude greater, i . e . , GR >> GL) .

[0173] In the embodiment of FIG. 16, all metal layers of each electrode (40, 42, 44 ) are electrically connected through the respective via layers . In other embodiments, signal electrode 42 comprises a MOM capacitor 61 formed between two metal layers, as described above with reference to FIGS . 3 and 4. In certain embodiments, one or more of the electrodes in the test structure comprises a single metal layer . It is noted that the asymmetric waveguide placement ( GR >> GL) described herein is particularly beneficial for characterizing asymmetric transmission line designs (e . g. , as described with reference to FIGS . 8-12 ) . For symmetric transmission line designs, the test structure is configured with GR equal to GL, as there is no asymmetry to characterize . In the embodiment of FIG. 16, the distance between ground electrode 40 and signal electrode 42 is smaller than the distance between signal electrode 42 and ground electrode 44, forming an asymmetric test structure distinct from the asymmetric transmission line configurations of FIGS . 8-12.

[0174] Because second TFLN waveguide 38 is positioned at a greater distance GR from signal electrode 42 than first TFLN waveguide 36, the electric field experienced by secondTFLN waveguide 38 is weaker than the electric field experienced by first TFLN waveguide 36. As a result, the modulation strength of second TFLN waveguide 38 is substantially less than the modulation strength of first TFLN waveguide 36. For example, when GR is approximately ten times GL, the single-arm half-wave voltage Vnon the side of second TFLN waveguide 38 is approximately fifteen times larger than the single-arm half-wave voltage on the side of first TFLN waveguide 36, rendering the modulation of second TFLN waveguide 38 negligible .

[0175] This configuration enables efficient characterization of modulator performance . By measuring the half-wave voltage of the test structure of FIG. 16, the half-wave voltage of a full GSGSG electro-optic modulator (such as electro-optic modulator 10 of FIG. 1 ) is predicted by applying a factor of two . Additionally, since the transmission line structure of FIG. 16 is configured identically to the transmission line structures described above with reference to FIGS . 8-12, the characteristic bandwidth of the full GSGSG configuration is also accurately determined from the test structure .

[0176] The GSG test structure of FIG. 16 occupies approximately 60% of the width of a corresponding GSGSG configuration, reducing the size of each test structure by approximately 40% in width. This enables the placement of a greater number of DOE structures on a single die . For example, for a die size of 1 cm x 1 cm, approximately 35 DOE structures are accommodated using the GSG test structure of FIG. 16, compared to approximately 20 DOE structures using a full GSGSG configuration.

[0177] FIG. 17 is a cross-sectional view that schematically illustrates a test structure similar to FIG. 16 with fewermetal layers on a second ground electrode, in accordance with an embodiment that is described herein.

[0178] In the embodiment of FIG. 17, the transmission line structure is identical to that of FIG. 16, except that ground electrode 44 comprises fewer metal layers than ground electrode 40. In the illustrated embodiment, one or more bottom metal layers and corresponding via layers are removed from ground electrode 44 , as described above with reference to FIG. 10. This configuration enables characterization of asymmetric transmission line designs in a compact test structure format . Although FIG. 17 illustrates the asymmetric configuration of FIG. 10, the test structure of FIGS . 16-18 is applicable to any of the asymmetric configurations described above with reference to FIGS . 8-12, including configurations with shifted ground electrodes, removed slow-wave structures, or combinations thereof . In certain embodiments, signal electrode 42 of the test structure of FIG. 17 comprises MOM capacitor 61 formed between two metal layers, for example as described above with reference to FIGS . 3 and 4.

[0179] FIG. 18 is a top view that schematically illustrates the test structure of FIG. 16 in a GSG configuration, in accordance with an embodiment that is described herein.

[0180] As shown in FIG. 18, the test structure comprises input waveguide 12 that receives an optical signal and directs it to input splitter 14. Input splitter 14 divides the optical signal into two paths, directing portions of the optical signal into first TFLN waveguide 16 and second TFLN waveguide 18. First TFLN waveguide 16 is positioned between ground electrode 20 and signal electrode 22 . Second TFLN waveguide 18 is positioned between signal electrode 22 and ground electrode 24. Second TFLN waveguide 18 ispositioned at a greater distance from signal electrode 22 than first TFLN waveguide 16, corresponding to the asymmetric waveguide placement described above with reference to FIG. 16. After passing through the modulation region, first TFLN waveguide 16 and second TFLN waveguide 18 converge at an output combiner 30, which combines the optical signals . The combined optical signal exits the test structure through an output waveguide 32 .

[0181] First TFLN waveguide 16 and second TFLN waveguide 18 of FIG. 18 correspond to first TFLN waveguide 36 and second TFLN waveguide 38 of FIGS . 16 and 17, respectively. Ground electrode 20 and ground electrode 24 of FIG. 18 correspond to ground electrode 40 and ground electrode 44 of FIGS . 16 and 17, respectively. Signal electrode 22 of FIG. 18 corresponds to signal electrode 42 of FIGS . 16 and 17. It is noted that unlike the electro-optic modulator 10 of FIG.

[0182] 1, the test structure of FIG. 18 does not include on-chip termination or phase shifter elements, as these are not required for characterizing electro-optic bandwidth and half-wave voltage . In some embodiments, however, one or more of these elements may be included, depending on the specific testing setup or test plan.

[0183] INTEGRATION AND FABRICATION

[0184] The transmission line structures and modulator configurations described above with reference to FIGS . 1-18 are designed to be compatible with traditional CMOS foundry processes that provide multiple metal layers . This compatibility enables integration of the TFLN-based modulators with other optical and optoelectronic components on a single die, forming a complete integrated electrooptic transceiver . The embodiments described below withreference to FIGS . 19 and 20 illustrate such integration, in which TFLN is employed exclusively for high-speed electro-optic modulation, while all other functionalities are implemented in a material stack through a one-step die-to-wafer or waf er-to-waf er bonding process . In contrast to other approaches, such as those that use transfer printing to attach TFLN to a silicon photonics (SiPho) platform, or that fabricate TFLN and SiPho chips separately and join them as discrete elements, the embodiments described herein integrate the TFLN layer directly onto the die through die-to-wafer or waf er-to-waf er bonding, enabling the TFLN modulation region and the other functional regions on the die to share a common plurality of metal layers . This shared metallization enables efficient electrical signal routing between the modulation region and other functional regions on the die, and supports co-integration with electronic integrated circuits (EICs) through flip-chip bonding.

[0185] FIG. 19 is a cross-sectional view that schematically illustrates an integrated die comprising a TFLN modulation region, a heater region, and a photodetector region, in accordance with an embodiment that is described herein.

[0186] The integrated die, generally designated 200, is built upon a silicon substrate 60, with an oxide layer 58 disposed above silicon substrate 60. The various functional regions and metal layer stacks are embedded in an oxide material 59 disposed above oxide layer 58. As described in previous sections of this document, in certain embodiments oxide material 59 and oxide layer 58 are portions of a continuous oxide structure, whereas in other embodiments, oxide material 59 and oxide layer 58 are separate oxide structures formed at different stages of the fabrication process .FIG. 19 illustrates a cross-section through a portion of a TFLN region 100, indicated by a dashed boundary. Within TFLN region 100, a TFLN layer 56 is disposed above oxide layer 58, and a TFLN waveguide 36 is formed on or as an extension of TFLN layer 56. A semiconductor waveguide layer 82 is disposed within oxide layer 58, directly beneath TFLN waveguide 36 ( for example in a manner similarly described above with reference to FIG. 14 ) . A ground electrode 40 is formed from a metal layer stack (Mi, Vi, M2, . . . Vn-i, Mn) on one side of TFLN waveguide 36, and a signal electrode 42 is formed from a metal layer stack (Mi, Vi, M2, . . . Vn-i, Mn) on the other side of TFLN waveguide 36. TFLN waveguide 36 is positioned between ground electrode 40 and signal electrode 42 (similar to as described in the other crosssection figures) . It should be understood that TFLN region 100 comprises additional metal stacks and TFLN waveguides to support the one or more modulators of the transmitter section shown in FIG. 20, as described below. Furthermore, in certain embodiments, semiconductor waveguide layer 82 is not present beneath TFLN waveguide 36 in the modulation region, but is disposed beneath the TFLN waveguide in a mode transition region where optical signals are coupled between SiPho waveguides and the TFLN waveguides .

[0187] To the left of TFLN region 100, a heater stack 102 comprises a metal layer stack (Mi, Vi, M2, . . . Vn-i, Mn) with a heating element 104 disposed below Mi . Heating element 104 is configured for thermal tuning of optical components and for providing resistance to electrical components on the die . Heating element 104 may be implemented in various ways, including, for example, as a resistive heater formed from a doped silicon region, a metal thin-film resistor,or a dielectric resistor disposed beneath Mi of heater stack 102 .

[0188] Although FIG. 19 shows a single heating element 104, in other embodiments, the die comprises multiple heating elements for thermal tuning of different optical components and for providing resistance to different electrical components .

[0189] Adj acent to heater stack 102, a silicon nitride (SiN) waveguide 106 and a silicon (Si) element 108 are disposed within oxide layer 58. Although SiN waveguide 106 and Si element 108 are shown in FIG. 19 as fully etched profiles, in other embodiments, they are partially etched. Similarly, although TFLN waveguide 36 is shown in FIG. 19 as partially etched, in other embodiments, TFLN waveguide 36 is fully etched .

[0190] SiN waveguides and partially etched Si waveguides are used for low-loss optical routing on the die . SiN and Si waveguides are also used for passive optical components, including power combination and splitting, interferometric mixing (e . g. , optical hybrids) , polarization manipulation, and adiabatic tapers for layer transition and edge coupling. In other embodiments, doped Si waveguides are used for active optical components, including phase shifting, thermal tuning, and variable optical attenuation.

[0191] Although FIG. 19 shows a single SiN waveguide 106, in other embodiments, the die comprises multiple SiN waveguides for routing and processing optical signals between different components on the die .

[0192] To the right of TFLN region 100, a photodetector stack 110 comprises a metal layer stack (Mi, Vi, M2, . . . Vn-i, Mn) . A germanium (Ge) element 112 is disposed above a silicon (Si) element 114. Ge element 112 is connected to Mi ofphotodetector stack 110 through a contact (or via) 116. Si element 114 is connected to Mi of photodetector stack 110 through a contact (or via) 118. Ge element 112 and Si element 114 together form a photodetector configured to convert optical signals into electrical signals .

[0193] The architecture of FIG. 19 demonstrates that the TFLN layer (and TFLN region 100) is employed exclusively for high-speed electro-optic modulation, while all other optical and optoelectronic functionalities are implemented in a standard SiPho material stack. These functionalities comprise low-loss waveguide routing (via SiN waveguide 106 and Si element 108 ) , passive optical functions such as power splitting, combining, interferometric mixing, polarization manipulation, and edge coupling (via SiN and Si waveguides) , thermal tuning (via heating element 104 ) , active optical functions such as phase shifting and variable optical attenuation (via doped Si waveguides) , and photodetection (via Ge element 112 and Si element 114 ) . Each material layer is employed where it provides maximum functional benefit, resulting in a hybrid platform that combines the high-speed electro-optic capabilities of TFLN with the mature device ecosystem of SiPho .

[0194] The SiPho material stack comprises one or more silicon photonic device layers . In certain embodiments, the SiPho material stack comprises a single silicon photonic device layer . In other embodiments, the SiPho material stack comprises multiple silicon photonic device layers, such as silicon, silicon nitride, and germanium layers .

[0195] The TFLN layer is integrated into the die through a one-step die-to-wafer or waf er-to-waf er bonding process . Optical signals are coupled into and out of TFLN region 100 through optimized Si / TFLN hybrid adiabatic tapers, enablinglow-loss and broadband mode conversion between the SiPho waveguides and the TFLN waveguides .

[0196] The process for fabricating integrated die 200 comprises several steps . In certain embodiments, the steps include : ( 1 ) fabricating a SiPho wafer comprising device layers (e . g. , Si, SiN, Ge, and doped regions with oxide cladding) below the metal layers ; (2 ) opening a bonding window on the SiPho wafer; (3) bonding TFLN dies or wafer to the SiPho wafer at the bonding window openings, either through direct bonding or through an adhesive layer; (4 ) removing the silicon handle and buried oxide layers from the TFLN dies or wafer through one or more etching processes; (5) partially or fully etching the TFLN layer to form waveguide structures for optical signal guiding; ( 6) depositing a passivation layer followed by a chemicalmechanical polishing (CMP) process to create a planarized surface; and (7 ) fabricating the metal layers to form heaters, transmission lines, vias, MOM / MIM capacitors, and electrical signal routing traces . In certain embodiments of the fabrication process, the TFLN bonding is performed prior to metallization. In other embodiments, the TFLN bonding is performed after metallization.

[0197] The architecture of FIG. 19 demonstrates that the TFLN layer (and TFLN region 100) is integrated into the die through a single bonding step, enabling seamless integration of the TFLN modulation region with the SiPho components . The shared plurality of metal layers supports efficient electrical signal routing and integration of support circuitry across the entire die .

[0198] FIG. 20 is a block diagram that schematically illustrates a die layout of an integrated electro-optic transceiver comprising a transmitter section and a receiversection on the integrated die of FIG. 19, in accordance with an embodiment that is described herein.

[0199] As shown in FIG . 20, an integrated electro-optic transceiver 300 is formed on a die 120. In the illustrated embodiment, die 120 is a SiPho die, and comprises a transmitter section 122 and a receiver section 130.

[0200] Transmitter section 122 comprises TFLN region 100 containing four Mach-Zehnder Modulators (MZMs) 124a, 124b, 124c, and 124d. Each MZM 124a-124d comprises two TFLN waveguide arms, each waveguide arm positioned between a signal electrode and a ground electrode, as described in previous sections of the present disclosure . A coherent light source 128, such as a multi-wavelength laser or a tunable laser, provides optical signals to transmitter section 122. The optical signals are split and routed via splitter stages 129a, 129b, 129c to MZMs 124a-124d for modulation. Specifically, splitter stage 129a splits the optical signal from coherent light source 128 into two paths . Splitter stages 129b and 129c further split each path, directing optical signals to MZMs 124a, 124b and MZMs 124c, 124d, respectively. The four MZMs 124a-124d are configured as a coherent In-phase / Quadrature ( IQ) modulator, enabling coherent optical transmission . In further detail, the output optical signals of MZMs 124a-124d are routed through respective I and Q branches, each branch including a phase shifter 135a, 135b configured to independently control the optical phase of the signal propagating through the branch. Phase shifters 135a and 135b correspond to phase shifter 35 of FIG. 1. The modulated optical signals exit transmitter section 122 at a transmitter output 138. It is noted that although not shown in FIG. 20 for clarity of illustration, phase shifters andon-chip terminations are within each modulator, similar to as illustrated and described with reference to FIG. 1.

[0201] Although FIG. 20 shows four MZMs in transmitter section 122, in other embodiments, transmitter section 122 comprises a different number of modulators, including a single modulator . Furthermore, the one or more modulators in transmitter section 122 are configurable for modulation formats other than coherent IQ modulation. The number and arrangement of the modulators may be adjusted to support different modulation formats for various system requirements .

[0202] Each MZM 124a-124d is coupled to a respective RF driver 126a, 126b, 126c, and 126d. In certain embodiments, the coupling of MZMs 124a-124d to drivers 126a-126 is via flip-chip bonding. Drivers 126a-126d provide the modulation signals to the signal electrodes of the respective MZMs . The plurality of metal layers of SiPho die 120 is configured to support co-integration of drivers 126a-126d with MZMs 124a-124d. The co-integration of drivers 126a-126d with MZMs 124a-124d through flip-chip bonding provides shorter electrical interconnection paths, lower parasitic inductance, and higher bandwidth compared to wire-bonding approaches, which is advantageous for high-speed modulation signals .

[0203] The output optical signals of MZMs 124a-124d are combined and routed via combiner stages 131a, 131b, 131c to transmitter (Tx) output 138. Specifically, combiner stages 131a and 131b combine the output optical signals of MZMs 124a-124b and MZMs 124c-124d, respectively, to form in-phase ( I ) and quadrature (Q) signal components . Combiner stage 131c combines the I and Q signal components into a combined modulated optical signal that exits thetransceiver at Tx output 138. In certain embodiments, combiner stage 131c comprises a Polarization Splitter-Rotator (PSR) that rotates and combines optical signals into orthogonal polarization states . In certain embodiments, Tx output 138 is coupled to an optical fiber for transmission.

[0204] Coherent light source 128 provides optical signals to both transmitter section 122 and receiver section 130. A splitter stage 127 splits the output of coherent light source 128 into a first path directed to transmitter section 122 (via splitter stage 129a) and a second path directed to receiver section 130 (via a splitter stage 143) .

[0205] Receiver section 130 receives optical signals at a receiver input 140. The received optical signals are split and directed via a splitting stage 141 to hybrid mixers 132a and 132b . In certain embodiments, splitting stage 141 comprises a PSR that separates the received optical signals into orthogonal polarization components directed to respective hybrid mixers 132a and 132b . Hybrid mixers 132a and 132b combine the received optical signals with a local oscillator signal from coherent light source 128 (received via splitting stage 143) for coherent detection. The outputs of hybrid mixers 132a and 132b are directed to photodetectors 134a, 134b, 134c, 134d, 134e, 134f, 134g, and 134h. Photodetectors 134a-134h convert the optical signals into electrical signals . The electrical signals from photodetectors 134a-134h are amplified by Transimpedance Amplifiers (TIAs) 136a, 136b, 136c, and 136d. The plurality of metal layers of SiPho die 120 is configured to support co-integration of TIAs 136a-136d with photodetectors 134a-134h, for example through flip-chipbonding. In certain embodiments, drivers 126a-126d and TIAs 136a-136d are implemented within one or more Electronic Integrated Circuits (EICs) co-integrated with SiPho die 120. In certain embodiments, a single EIC comprises both the drivers and the TIAs . In other embodiments, separate EICs are used for the drivers and the TIAs .

[0206] Although FIG. 20 shows eight photodetectors in receiver section 130, in other embodiments, receiver section 130 comprises a different number of photodetectors, including a single photodetector . Similarly, although FIG.

[0207] 20 shows four TIAs in receiver section 130, in other embodiments, receiver section 130 comprises a different number of TIAs, including a single TIA. The number of photodetectors and TIAs is determined by the receiver architecture and system requirements, such as the number of polarization and phase channels, the detection scheme (e . g. , coherent or direct detection) , and the desired signal processing configuration.

[0208] The co-integration of transmitter section 122 and receiver section 130 on a single SiPho die 120 enables a fully integrated, single-die optical transceiver that incorporates the high-performance TFLN modulation region together with the SiPho and electronic device layers implemented within a CMOS-compatible platform. The singledie approach offers various benefits, including simplified optical input / output assembly, improved module-level yield, and reduced input / output count that minimizes optical coupling losses . The performance of the transceiver is optimized for a multilayer metal stack, enabling efficient signal routing and integration of support circuitry while complying with CMOS foundry design rules .ADDITIONAL CONSIDERATIONS

[0209] The metal layers described herein may comprise any conductive material suitable for use in CMOS foundry processes . In some embodiments, the metal layers comprise copper (Cu) , which is commonly used in lower and intermediate metal layers due to its low resistivity and suitability for high-speed signal routing. In some embodiments, the top metal layer comprises aluminum (Al) , which is often used for power distribution and pad bonding. In some embodiments, the metal layers comprise tungsten (W) , cobalt (Co) , ruthenium (Ru) , or other conductive materials . The via layers may comprise tungsten or other conductive materials suitable for forming vertical interconnects between metal layers . Barrier or liner layers, such as titanium (Ti) , titanium nitride (TiN) , tantalum (Ta) , or tantalum nitride (TaN) , may be disposed between the metal layers and the surrounding dielectric material . The specific materials used for the metal layers, via layers, and barrier layers are determined by the foundry process and are not limited to the examples described herein.

[0210] The metal layers, via layers, and other structural elements depicted in the figures are represented schematically as rectangles of various sizes and shapes . The relative dimensions, proportions, and spacing between different layers and different stacks are chosen for clarity of illustration and are not drawn to scale . The actual dimensions and proportions of these elements are determined by the specific CMOS foundry design rules and process parameters, and are not limited to the representations shown in the figures .As used herein, the terms "first", "second", and similar ordinal terms are used for identification purposes to distinguish between elements, and do not imply any particular order, sequence, preference, or priority unless explicitly stated otherwise .

[0211] As used herein, directional terms such as "left", "right", "above", "below", "upper", and similar terms are used for reference purposes to describe the relative positions of elements as illustrated in the figures, and are not intended to be limiting. The actual orientation of the described elements may vary depending on the implementation, and such directional terms should be interpreted accordingly.

[0212] It will be appreciated that the embodiments described above are cited by way of example, and that the present invention is not limited to what has been particularly shown and described hereinabove . Rather, the scope of the present invention includes both combinations and subcombinations of the various features described hereinabove, as well as variations and modifications thereof which would occur to persons skilled in the art upon reading the foregoing description and which are not disclosed in the prior art . Documents incorporated by reference in the present patent application are to be considered an integral part of the application except that to the extent any terms are defined in these incorporated documents in a manner that conflicts with the definitions made explicitly or implicitly in the present specification, only the definitions in the present specification should be considered .

Claims

1. CLAIMS1. An electro-optic modulator comprising:a thin-film lithium niobate (TFLN) waveguide configured to guide an optical signal; anda transmission line comprising:a signal electrode formed from a plurality of metal layers and configured to receive a modulation signal that modulates the optical signal, the signal electrode comprising a metal-oxide- metal (MOM) capacitor formed between two metal layers of the plurality of metal layers, and a ground electrode formed from one or more metal layers and disposed adj acent to the signal electrode on a first side of the signal electrode, wherein the TFLN waveguide is positioned between the signal electrode and the ground electrode .

2. The electro-optic modulator of claim 1, wherein the two metal layers of the MOM capacitor are adj acent metal layers .

3. The electro-optic modulator of claim 1, wherein the two metal layers of the MOM capacitor are non-adj acent metal layers .

4. The electro-optic modulator of claim 3, wherein one or more intermediate metal layers between the two non-adj acent metal layers are omitted from the signal electrode .

5. The electro-optic modulator of claim 4, wherein the ground electrode comprises the one or more intermediate metal layers .

6. The electro-optic modulator of claim 1, further comprising a second ground electrode disposed adj acent to the signal electrode on a second side of the signal electrode opposite the first side, wherein the ground electrode comprises a first number of metal layers and the second ground electrode comprises a second number of metal layers less than the first number of metal layers .

7. The electro-optic modulator of claim 1, further comprising a second ground electrode disposed adj acent to the signal electrode on a second side of the signal electrode opposite the first side, wherein the second ground electrode is disposed at a greater distance from the signal electrode than the ground electrode .

8. The electro-optic modulator of claim 1, further comprising one or more slow-wave structures on one or more metal layers of the plurality of metal layers, the one or more slow-wave structures configured to provide velocity matching between the modulation signal and the optical signal .

9. The electro-optic modulator of claim 8, wherein the one or more slow-wave structures comprise T-rails .

10. The electro-optic modulator of claim 8, wherein the one or more slow-wave structures comprise a corrugated coplanar waveguide (C-CPW) structure .

11. The electro-optic modulator of claim 1, wherein the MOM capacitor is configured to provide velocity matching between the modulation signal and the optical signal .

12. The electro-optic modulator of claim 1, wherein the MOM capacitor is configured to provide a separate electrical path for a bias control signal .

13. The electro-optic modulator of claim 1, further comprising a second signal electrode, wherein the transmission line is configured in a GSGSG arrangement for differential drive operation.

14. An electro-optic modulator comprising:a thin-film lithium niobate (TFLN) waveguide configured to guide an optical signal;a transmission line comprising:a signal electrode formed from a plurality of metal layers and configured to receive a modulation signal that modulates the optical signal, a first ground electrode formed from one or more metal layers and disposed adj acent to the signal electrode on a first side of the signal electrode, wherein the first ground electrode has a first structural configuration, and wherein the TFLN waveguide is positioned between the signal electrode and the first ground electrode, anda second ground electrode formed from one or more metal layers and disposed adj acent to the signal electrode on a second side of the signal electrode opposite the first side, wherein the second ground electrode has a second structural configuration different from the first structural configuration.

15. The electro-optic modulator of claim 14, wherein the second structural configuration comprises one or more slow-wave structures .

16. The electro-optic modulator of claim 14, wherein the first structural configuration is defined by the oneor more metal layers of the first ground electrode, wherein the second structural configuration is defined by the one or more metal layers of the second ground electrode, wherein the one or more metal layers of the first ground electrode consists of a first number of metal layers, wherein the one or more metal layers of the second ground electrode consists of a second number of metal layers less than the first number of metal layers .

17. The electro-optic modulator of claim 14, wherein the first structural configuration is defined by a first distance between the first ground electrode and the signal electrode, and wherein the second structural configuration is defined by a second distance between the second ground electrode and the signal electrode, and wherein the second distance is greater than the first distance .

18. The electro-optic modulator of claim 14, wherein the signal electrode comprises a metal-oxide-metal (MOM) capacitor formed between two metal layers of the plurality of metal layers .

19. An electro-optic modulator comprising:a thin-film lithium niobate (TFLN) waveguide configured to guide an optical signal; anda transmission line comprising:a signal electrode formed from a plurality of electrically connected metal layers and configured to receive a modulation signal that modulates the optical signal, and a ground electrode formed from one or more metal layers and disposed adj acent to the signal electrode, wherein the TFLN waveguide is positioned between the signal electrode and theground electrode, and wherein the ground electrode comprises a metal-insulator-metal (MIM) capacitor .

20. The electro-optic modulator of claim 19, wherein the MIM capacitor includes a capacitor structure inserted between two metal layers of the ground electrode .

21. The electro-optic modulator of claim 19, wherein the MIM capacitor comprises a finger-type electrode structure .

22. The electro-optic modulator of claim 19, further comprising a semiconductor waveguide layer disposed beneath the TFLN waveguide .

23. The electro-optic modulator of claim 19, further comprising a second ground electrode formed from a plurality of electrically connected metal layers and disposed adj acent to the signal electrode on a side opposite the ground electrode .

24. An electro-optic modulator comprising:a thin-film lithium niobate (TFLN) waveguide configured to guide an optical signal;a semiconductor waveguide layer disposed beneath the TFLN waveguide; anda transmission line comprising:a signal electrode formed from a plurality of electrically connected metal layers and configured to receive a modulation signal that modulates the optical signal, and a ground electrode formed from one or more metal layers and disposed adj acent to the signal electrode, wherein the TFLN waveguide ispositioned between the signal electrode and the ground electrode .

25. The electro-optic modulator of claim 24, wherein the semiconductor waveguide layer is configured to enhance velocity matching between the modulation signal and the optical signal .

26. The electro-optic modulator of claim 24, wherein the semiconductor waveguide layer comprises silicon.

27. The electro-optic modulator of claim 24, wherein the signal electrode comprises a metal-oxide-metal (MOM) capacitor formed between two metal layers of the plurality of metal layers .

28. The electro-optic modulator of claim 24, wherein the ground electrode comprises a metal-insulator-metal (MIM) capacitor .

29. An electro-optic modulator comprising:a thin-film lithium niobate (TFLN) waveguide configured to guide an optical signal;a plurality of metal layers;a signal electrode formed from a first part of the plurality of metal layers and configured to receive a modulation signal that modulates the optical signal, the signal electrode comprising a metal- oxide-metal (MOM) capacitor formed between two metal layers of the plurality of metal layers; a first ground electrode formed from a second part of the plurality of metal layers and disposed adj acent to the signal electrode on a first side of the signal electrode, wherein the TFLN waveguide ispositioned between the signal electrode and the ground electrode; anda second ground electrode formed from a third part of the plurality of metal layers and disposed adj acent to the signal electrode on a second side of the signal electrode opposite the first side .

30. The electro-optic modulator of claim 29, wherein the first ground electrode has a first structural configuration, and wherein the second ground electrode has a second structural configuration different from the first structural configuration.

31. The electro-optic modulator of claim 29, wherein the first ground electrode comprises a first number of metal layers of the plurality of metal layers and the second ground electrode comprises a second number of metal layers of the plurality metal layers less than the first number of metal layers .

32. The electro-optic modulator of claim 29, wherein the second ground electrode is disposed at a greater distance from the signal electrode than the first ground electrode .

33. An apparatus comprising:a first thin-film lithium niobate (TFLN) waveguide configured to guide a first optical signal; a second TFLN waveguide configured to guide a second optical signal; anda transmission line arranged in a ground-signal-ground (GSG) configuration, the transmission line comprising :a signal electrode formed from one or more metal layers and configured to receive a modulation signal ,a first ground electrode formed from one or more metal layers and disposed adj acent to the signal electrode on a first side of the signal electrode, wherein the first TFLN waveguide is positioned between the signal electrode and the first ground electrode at a first distance from the signal electrode, anda second ground electrode formed from one or more metal layers and disposed adj acent to the signal electrode on a second side of the signal electrode opposite the first side, wherein the second TFLN waveguide is positioned between the signal electrode and the second ground electrode at a second distance from the signal electrode, wherein the second distance is greater than the first distance .

34. The apparatus of claim 33, wherein the second distance is approximately ten times the first distance .

35. The apparatus of claim 33, wherein the signal electrode is formed from a plurality of metal layers and comprises a metal-oxide-metal (MOM) capacitor formed between two metal layers of the plurality of metal layers .

36. The apparatus of claim 33, wherein the first ground electrode has first structural configuration, and wherein the second ground electrode has a second structural configuration different from the first structural configuration .

37. The apparatus of claim 33, wherein the second ground electrode is disposed at a greater distance from the signal electrode than the first ground electrode .

38. An integrated electro-optic transceiver comprising :a die comprising a plurality of metal layers;a thin-film lithium niobate (TFLN) layer bonded to the die ;a transmitter section comprising one or more modulators, each modulator comprising:a TFLN waveguide associated with the TFLN layer, a signal electrode formed from a first part of the plurality of metal layers and configured to receive a modulation signal that modulates an optical signal propagating through the TFLN waveguide, anda ground electrode formed from a second part of the plurality of metal layers and disposed adj acent to the signal electrode, wherein the TFLN waveguide is positioned between the signal electrode and the ground electrode; and a receiver section comprising one or more photodetectors formed in the die .

39. The integrated electro-optic transceiver of claim 38, wherein the die is a silicon photonics (SiPho) die .

40. The integrated electro-optic transceiver of claim 38, wherein the TFLN layer is bonded to the die by a one-step die-to-wafer or waf er-to-waf er bonding process .

41. The integrated electro-optic transceiver of claim 38, wherein the transmitter section comprises one or more modulators .

42. The integrated electro-optic transceiver of claim 38, wherein the one or more photodetectors comprise germanium disposed on silicon.

43. The integrated electro-optic transceiver of claim 38, further comprising one or more heating elements formed in the die and configured for thermal tuning.

44. The integrated electro-optic transceiver of claim 38, wherein the die further comprises at least one of : i) one or more silicon nitride (SiN) waveguides, or ii) one or more silicon (Si) waveguides .

45. The integrated electro-optic transceiver of claim 38, wherein the signal electrode comprises a metal-oxide-metal (MOM) capacitor formed between two metal layers of the plurality of metal layers .

46. The integrated electro-optic transceiver of claim 38, wherein the plurality of metal layers is configured to support co-integration with one or more electronic integrated circuits (EICs) comprising a plurality of RF drivers and a plurality of transimpedance amplifiers (TIAs) .

47. A method of fabricating an integrated electro-optic transceiver, the method comprising:obtaining a die comprising a plurality of metal layers; bonding a thin-film lithium niobate (TFLN) layer to the die ;forming one or more modulators in a transmitter section, each modulator comprising a TFLN waveguide associated with the TFLN layer, a signal electrode formed from a first part of the plurality of metal layers, and a ground electrode formed from a secondpart of the plurality of metal layers, wherein the TFLN waveguide is positioned between the signal electrode and the ground electrode; andforming one or more photodetectors in a receiver section of the die . .

48. The method of claim 47, wherein the die is a silicon photonics (SiPho) die .

49. The method of claim 47, wherein obtaining the die comprises obtaining a wafer comprising a silicon substrate, one or more silicon photonic device layers, and the plurality of metal layers .

50. The method of claim 47, wherein bonding the TFLN layer comprises a one-step die-to-wafer or waf er-to-waf er bonding process .

51. The method of claim 47, further comprising forming a metal-oxide-metal (MOM) capacitor in the signal electrode between two metal layers of the plurality of metal layers .

52. The method of claim 47, wherein forming the one or more photodetectors comprises forming germanium on silicon.

53. The method of claim 47, further comprising forming one or more heating elements in the die for thermal tuning.

54. The method of claim 47, wherein bonding the TFLN layer to the die is performed prior to forming the plurality of metal layers .

55. The method of claim 47, wherein the plurality of metal layers is formed prior to bonding the TFLN layer to the die .

56. The method of claim 47, further comprising: integrating one or more electronic integrated circuits (EICs) in the die through flip-chip bonding.

57. The method of claim 56, wherein each of the one or more EICs includes a plurality of RE drivers and a plurality of transimpedance amplifiers (TIAs) .

Images