Heterogeneous integrated photonic platform with nonlinear frequency conversion elements

Abstract

An apparatus includes first, second, and third elements fabricated on a common substrate. The first element includes an active waveguide structure including a first section supporting a first optical mode. The second element includes a passive waveguide structure supporting a second optical mode. The third element, at least partially butt-coupled to the second section, includes an intermediate waveguide structure supporting an intermediate optical mode. At least a portion of the second element is nonlinear, supporting frequency conversion. Gradually decreasing waveguide structures in at least one of the second and third elements facilitate efficient adiabatic conversion between the first optical mode and one of the intermediate optical modes. No adiabatic conversion occurs between any of the intermediate optical modes and the first optical mode. Photolithographic alignment marks are used to define the mutual alignment of the elements.

Description

Technical Field

[0001] This invention relates to semiconductor lasers. More specifically, certain embodiments of the invention relate to improved performance of heterogeneous integrated lasers using different materials, said heterogeneous integrated lasers being optically coupled to nonlinear elements for frequency conversion. Background Technology

[0002] A photonic integrated circuit (PIC), or integrated optical circuit, is a device that integrates multiple photonic functions and is therefore similar to an electronic integrated circuit. The main difference is that a photonic integrated circuit provides functionality for information signals applied to an optical carrier. The most commercially available material platform for photonic integrated circuits is indium phosphide (InP), which allows for the integration of various active and passive optical functions on the same chip. Although many current PICs are implemented on the InP platform, significant research has been conducted over the past decade on implementing PICs using silicon instead of InP due to some of the superior properties and processing capabilities of silicon, leveraging existing investments in electronic integrated circuits.

[0003] The biggest drawback of using silicon in PICs is that it is an indirect bandgap material, making it difficult to provide an electrical pump source. This problem is typically addressed by assembling a PIC comprising two or more chips made of different materials in a separate process. This approach is challenging due to the need for very fine alignment, which increases packaging costs and introduces scalability limitations. Another approach to addressing the bandgap problem is to combine two different materials and process them together, thereby eliminating the need for precise alignment during the bonding of larger workpieces or complete wafers of different materials and allowing for mass production. In this disclosure, we use the term "hybrid" to describe the first approach, which involves the precise assembly of separately processed portions, and we use the term "heterogeneous" to describe the latter approach, which involves combining two materials and then processing the resulting composite to define waveguides and other components of interest.

[0004] To transmit optical signals between different materials, heterojunction methods utilize tapered structures with gradually decreasing dimensions until the refractive indices of the efficient modes of the different materials match, resulting in efficient power transfer. This approach typically works well when the materials have similar refractive indices, such as silicon and InP. However, when there are significant differences in effective refractive indices, such as between SiN and GaAs or InP, the requirement for gradually decreasing dimensions becomes prohibitive, limiting efficient power transfer. Specifically, extremely small, gradually decreasing widths (on the order of tens of nanometers) may be necessary to provide good coupling. Achieving such dimensions is complex and can be prohibitively expensive.

[0005] While InP and silicon-based PICs address many current needs, they have several limitations; these include the fact that the operating wavelength range is limited by the material's absorption, which increases losses, and the limitation on the maximum light intensity that the PIC can handle, and therefore the limitation on optical power. To address these limitations, alternative waveguide materials such as SiN, LiNbO3, TiO2, Ta2O5, and AlN have been considered. Typically, these dielectric waveguides have high bandgap energies, which provide better high-power handling and transparency at shorter wavelengths; however, these materials also typically have low refractive indices. For example, SiN with a bandgap of ~5 eV has a refractive index of ~2, AlN has a bandgap of ~6 eV and a refractive index of approximately ~2, and SiO2 with a bandgap of ~8.9 eV has a refractive index of ~1.44. In comparison, both InP and GaAs have refractive indices >3. This makes gradually decreasing refractive indices challenging.

[0006] Alternative hybrid methods have the drawbacks mentioned above (i.e., the need for precise alignment) as well as corresponding complex encapsulation and scaling limitations.

[0007] A novel method for addressing the problems discussed above is proposed in US Patent No. 10,859,764B2, which employs a docking-coupled mode converter to allow the use of heterogeneous processes without requiring extremely small, gradually decreasing widths. This invention relates to a PIC employing docking coupling in this manner, and it includes a nonlinear element for converting the frequency of a directly injected light source. Various nonlinear effects can be utilized, and in some embodiments, second harmonic generation achieves the performance of a chip-scale light source at certain optical wavelengths, which may be challenging or even impossible to achieve using other methods.

[0008] One example is the chip-scale integration of a narrow-linewidth InP laser operating around 1550 nm (linewidths as low as 1 Hz or even lower have been demonstrated using InP and SiN), which is frequency-doubled to provide a similarly narrow-linewidth source emitting around 775 nm. The pump wavelength of such a source, based on InP as the active material, can be tunable over a very wide range, typically 1200 nm to 1800 nm, but it can be even wider. Due to the inherently higher losses affecting shorter wavelengths, it is difficult to achieve similar levels of performance with GaAs-based pumps (which typically support direct emission in the 630 nm to 1100 nm wavelength range, but can be even wider) using SiN waveguides.

[0009] The second example addresses the green-yellow gap in available sources emitting between 520 nm and 630 nm. Supporting this wavelength range with only directly emitting elements, such as commercially available LED sources (as opposed to laser sources), is extremely challenging due to the challenges associated with shifting the bandgap of GaN-based gain media to longer wavelengths. However, by frequency doubling the output of GaAs sources operating between 1040 nm and 1260 nm, it is possible to enter the wavelength “gap” region, where direct emission from GaAs occurs with very high efficiency.

[0010] Appropriately, by utilizing nonlinear type conversions other than frequency doubling, and using high-performance pumps integrated with nonlinear elements at the chip level, other wavelength ranges can be accessed in a similar manner.

[0011] In particular, the embodiments described below relate to the detailed design of optical coupling structures and waveguides, which are essential for creating high-performance lasers that rely on nonlinear elements for frequency conversion. Attached Figure Description

[0012] Figure 1 The illustration shows an apparatus according to an embodiment of the present invention, shown in cross-section.

[0013] Figure 2 A cross-sectional top view of an apparatus according to some embodiments of the present invention is shown.

[0014] Figure 3 Cross-sectional end views of three devices according to some embodiments of the present invention are shown.

[0015] Figure 4 Cross-sectional top views of two devices according to some embodiments of the present invention are shown.

[0016] Figure 5 Cross-sectional top views of two devices according to some embodiments of the present invention are shown.

[0017] Figure 6 Cross-sectional top views of three devices according to some embodiments of the present invention are shown.

[0018] Figure 7 The illustration shows lasers based on photonic integrated circuits (PICs) according to two embodiments of the present invention.

[0019] Figure 8 The illustration shows lasers based on photonic integrated circuits (PICs) according to three other embodiments of the present invention. Detailed Implementation

[0020] This paper describes an embodiment of a platform for realizing photonic integrated circuits using wafer bonding and deposition of different materials, wherein optical coupling is improved by using mode switching and docking coupling schemes, and wherein the output of the light source is efficiently coupled to a nonlinear element for frequency conversion.

[0021] In the following detailed description, reference is made to the accompanying drawings, which form a part of this specification, wherein like numerals always denote like parts, and wherein exemplary embodiments are illustrated by way of practicing the subject matter of this disclosure. It should be understood that other embodiments may be utilized, and structural or logical changes may be made without departing from the scope of this disclosure. Therefore, the following detailed description should not be taken in a limiting sense, and the scope of the embodiments is defined by the appended claims and their equivalents.

[0022] This description may use perspective-based descriptions, such as top / bottom, inside / outside, up / down, etc. Such descriptions are used only to facilitate discussion and are not intended to limit the application of the embodiments described herein to any particular orientation. The description may use the phrases "in one embodiment" or "in an embodiment," each of which may refer to one or more of the same or different embodiments. Furthermore, the terms "comprising," "including," "having," etc., used with respect to embodiments of this disclosure are synonymous.

[0023] For the purposes of this disclosure, the phrase "A and / or B" means (A), (B), or (A and B). For the purposes of this disclosure, the phrase "A, B, and / or C" means (A), (B), (C), (A and B), (A and C), (B and C), or (A, B, and C).

[0024] This document uses the term "coupling" along with its derivatives. "Coupling" can mean one or more of the following: "Coupling" can mean two or more elements in direct physical, electrical, or optical contact. However, "coupling" can also mean two or more elements in indirect contact with each other, but still cooperating or interacting with each other, and can mean one or more other elements are coupled or connected between the elements referred to as being coupled to each other. The term "direct coupling" means two or more elements in direct contact at least a portion of their surfaces. The term "butt coupling" is used in its usual sense here, meaning "termination" or axial coupling, where there is a minimum or zero axial offset between the elements in question. For example, in the case of forming some kind of thin interlayer between elements, such as a thin coating typically used to provide high reflectivity or anti-reflective properties, the axial offset can be slightly greater than zero. It should be noted that the axes of two waveguide structures or elements do not need to be collinear for them to be accurately described as butt coupling. In other words, the interface between the elements does not need to be perpendicular to either axis. The following discusses... Figure 2 The embodiments are examples of this possibility.

[0025] This document may use the terms “active device,” “active structure,” or other “active” elements, sections, or components. An active device or part of an active device is capable of directly generating, amplifying, attenuating, modulating, and / or detecting light. This contrasts with what we refer to as a “passive device,” whose primary function is to confine and guide light, and / or provide separation, combination, filtering, and / or other functions typically associated with passive devices. Some passive devices can provide functions that overlap with those of active devices, such as, for example, phase tuning achieved using thermal effects or similar effects that can provide modulation. An absolute distinction between “active” and “passive” should not be assumed purely based on material composition or device structure. For example, a silicon device may be considered active under certain modulation conditions or when detecting low-wavelength radiation, but passive in most other cases. Active or passive devices may include elements, sections, and / or sections with nonlinear properties. As known in the art and described below, nonlinear properties can be used for frequency conversion.

[0026] Figure 1 This is a schematic cross-sectional view of an integrated photonic device 100 that utilizes docking coupling and mode switching to achieve efficient coupling between dissimilar materials and employs nonlinear elements for frequency conversion. An exemplary cross-section includes a substrate 105, which can be any suitable substrate for semiconductor and dielectric processing, such as Si, InP, GaAs, quartz, sapphire, glass, GaN, silicon-on-insulator, or other materials known in the art. In the illustrated embodiment, a layer 104 of a second material is deposited, grown, transferred, bonded, or otherwise attached to the top surface of the substrate 105 using techniques known in the art. The primary purpose of layer 104 is to provide an optical cladding for layer 102 (described below) if necessary to form an optical waveguide centered on 102. Optical waveguides are typically implemented by confining light waves by placing a core of higher refractive index between two lower refractive index layers. In some embodiments, layer 104 is omitted, and the substrate 105 itself serves as the cladding.

[0027] Using techniques known in the art, layer 102 is deposited, grown, transferred, bonded, or otherwise attached to the top of layer 104 (if present) and / or the top of substrate 105. If layer 104 is present, the refractive index of layer 102 is higher than that of layer 104; or, if layer 104 is absent, the refractive index of layer 102 is higher than that of substrate 105. In one embodiment, the material of layer 102 may include, but is not limited to, SiN and SiNO. xOne or more of TiO2, Ta2O5, (doped) SiO2, LiNbO3, and AlN are used. In some embodiments, other common dielectric materials may be used for layer 102. In other embodiments, semiconductor materials may be used for layer 102. In some embodiments, the refractive index of layer 102 is between 1.44 and 2.5. One or both of layers 104 and 102 may be patterned, etched, or re-deposited, which is common in the art. Layer 102 is the core of a passive waveguide structure, referred to as the second element in the claims of this disclosure.

[0028] Layer 108 has a lower refractive index than layer 102. Layer 108 covers layer 102 and lies beneath layers 101 and 103 (described in detail below) used to planarize the top surface of layer 102. In some embodiments, the planarity of the top surface of layer 108 is provided by chemical mechanical polishing (CMP) or other etching, chemical, and / or mechanical polishing methods. In other embodiments, planarity is provided due to the inherent nature of the method of depositing layer 108, such as if the material of layer 108 is spin-coated glass, polymer, photoresist, or other suitable material. Planarization can be controlled to leave a layer of desired, typically very low thickness (e.g., ...) on top of layer 102. Figure 1 (as shown), or remove all material (not shown) above the top surface level of layer 102. With layer 108 remaining on top of layer 102, its target thickness ranges from 10 nm to several hundred nm. Due to the typical (common in the art) non-uniformity of planarization processes (e.g., CMP), the actual thickness of the entire wafer can differ from the target thickness by more than ±100 nm. In some embodiments, spin-coated material is used for planarization and then etched back, thereby improving the uniformity of the entire wafer compared to a typical CMP process.

[0029] Layer 101 is bonded to all or part of the top of the corresponding (108, 102) top surfaces. Layer 101 includes the core of an active waveguide structure, referred to as the first element in the claims of this disclosure.

[0030] The bonding can be a direct molecular bond, or additional materials can be used to facilitate bonding, such as, for example, metal layers or polymer films known in the art. Layer 101 forms what is commonly referred to as the active region, and it itself can be composed of one or more materials, including but not limited to GaAs and GaAs-based ternary and quaternary materials, InP and InP-based ternary and quaternary materials, or any other material suitable for providing direct light emission. In some embodiments, layer 101 is multilayered, including sublayers that together provide both optical and electrical confinement as well as electrical contacts, as known in the field of active devices. In yet another embodiment, layer 101 uses lower layers 102, 108, 104, and / or 105 to provide electrical and / or optical confinement as well as one or more electrical contacts. In some embodiments, the sublayers of layer 101 provide vertical confinement ( Figure 1 (top / bottom) and horizontal restriction (perpendicular to) Figure 1 The surface of the cut is provided by at least one etching, as is known in the field of active device manufacturing.

[0031] In some embodiments, layer 101 can be efficiently electrically pumped to generate light emission and gain. This invention achieves efficient optical coupling between waveguides formed in layers 101 and 102. The material of layer 102 can provide additional functionalities such as broadband transparency, high-intensity processing, phase shifting, combining, separating, filtering, and / or other functionalities known in the art through temperature, strain, or other tuning mechanisms. In some embodiments, one characteristic of layer 102 is a sufficiently large nonlinearity that can be used for frequency conversion. This nonlinearity may be an inherent property of the material forming layer 102, but it can be enhanced by the implementation of electrode 109, such as... Figure 1 One of the electrodes 109 shown. Typically, at least two such electrodes are used, as in [the diagram]. Figure 3-6 This will be explained in more detail with the help of [the relevant parties]. Electrodes can also be used to implement phase-tuning elements, such as, for example, thermal tuning, to further optimize performance. In some embodiments, electrodes can also be used to implement phase or intensity modulation. In some embodiments, nonlinearity provides efficient second harmonic generation. In other embodiments, nonlinearity provides efficient third harmonic generation. In other embodiments, different frequency conversion schemes are utilized, including but not limited to four-wave mixing, optical parametric oscillation, Raman and / or Brillouin effects.

[0032] Layer 103 promotes efficient coupling, and in the presence of layer 106, layer 106 promotes efficient coupling. Layer 103 is the core of the intermediate waveguide structure, which is referred to as the intermediate element in the claims of this disclosure.

[0033] Optional layer 106 primarily serves as an anti-reflective or high-reflective coating at the interface between layers 101 and 103. Layer 103 serves as an intermediate waveguide layer, which in some embodiments receives the distribution of optical modes supported by the waveguide for which layer 101 provides the core (shown by line 150), efficiently captures it as mode distribution 151, progressively transfers it to mode distribution 152, and finally to mode distribution 153. Mode distribution 153 is efficiently coupled to the waveguide for which layer 102 provides the core. The mode distribution 153 is then efficiently frequency-converted into mode 154 operating at different frequencies using nonlinearities in layer 102. In some embodiments, the frequency of mode 154 is substantially twice the frequency of mode 153. In other embodiments, the frequency of mode 154 is substantially three times the frequency of mode 153. In other embodiments, different relationships are generated between the frequencies of modes 153 and 154.

[0034] By utilizing a gradually decreasing structure fabricated in layers 102 and / or 103, the refractive index of layer 103 can be designed to facilitate efficient coupling of mode distribution 150 and efficiently transform modes into modes having mode distribution 153. In some embodiments, the refractive index of layer 103 is between 1.5 and 2. The thickness of layer 103 is an optimized parameter, and in some embodiments, it is between 400 nm and 2000 nm. Prior to this invention, i.e., without the intermediate layer 103, the requirement for a gradually decreasing width would be problematic, as described above. However, the use of the intermediate layer 103 significantly reduces the stringent requirement for a gradually decreasing width, thereby allowing efficient transfer between very high refractive index materials (such as, for example, GaAs in layer 101) and low refractive index materials (such as SiN in layer 102).

[0035] By observing the mode distribution, the differences between the optical modes supported by the waveguides in layers 101 and 102 may be obvious or not obvious, but the mode overlap is less than 100%, and the vertical offset between modes 150 and 153 (in Figure 1 The loss due to imperfect mode overlap may (in the absence of intermediate layer 103) result in significant optical loss. In some cases, a loss of up to 1 dB may be considered acceptable, but a loss greater than 1 dB is unacceptable. In other cases, a loss level of 3 dB may be the criterion for selection. The function of layer 103 is to keep the optical loss due to imperfect mode overlap below a level determined to be acceptable in a given application.

[0036] The upper cladding 107 of the waveguide implemented in 103 and / or 102 can be ambient air (meaning no actual cladding material is deposited), or it can be as follows: Figure 1Any other suitable material intentionally deposited as shown, including but not limited to polymers, SiO2, SiNx, SiNOx, etc. In some embodiments, the same material is used for layers 107 and 108. In some embodiments (not shown), the cladding function of layer 107 can be provided by multiple depositions, for example, one material provides cladding for mold 153 guided by a core formed in layer 102, and another material provides cladding for mold 151 guided by a core formed in layer 103. In all cases, the refractive index of the cladding material is lower than that of the material providing the core for mold guidance. In yet another embodiment (not shown), layer 103 can provide cladding function to layers 102 as well as molds 153 and 154 because its refractive index is designed to be low.

[0037] Layer 109 is a conductive material deposited to provide an electric field on at least a portion of layer 102 and to enhance or make its nonlinear effects more efficient, as will be... Figure 3-6 This will be described in more detail with the help of [unclear]. In some embodiments, layer 109 is absent, and material 102 inherently provides efficient nonlinear effects without the need for an applied electric field. In other embodiments, additional conductive electrodes are deposited to provide a tuner function to optimize the nonlinear effects (not shown here, but by means of [unclear]). Figure 6 (Description provided).

[0038] There are one or more photolithographic alignment marks (not shown in this cross-sectional view, but see, for example) Figure 2 (240 in the diagram, and described below) to facilitate precise alignment between the layers formed during various processing steps.

[0039] In some embodiments, layer 108 is absent, and layer 101 is bonded, and layer 103 is deposited on top of the non-planar layer 102. In such embodiments, there is no planarization step.

[0040] Figure 2 A top view of an integrated photonic device 200 according to an embodiment of the present invention is depicted, magnified to a transition region where the boundaries between different materials are angled to control transmission and back reflection. Functional layers 201 to 207 (unless explicitly defined differently) correspond to... Figure 1 The functional layers described are 101 to 107.

[0041] An optical mode 250 (defined by at least one laterally etched portion, not shown), supported by active layer 201, is guided into layer 203 via an optional coating layer 206 (if present). Layer 203 supports optical mode 252 and serves to convert the mode into a highly efficient coupling layer 202 to support mode 253. Layer 206 can provide high reflectivity, or alternatively, when designed as an anti-reflective coating, can promote reduced reflection. To facilitate coupling between modes supported by layers 201 and 202, the dimensions of layer 202 gradually decrease towards layer 201, as indicated by the relatively small width of tip 211 relative to the width of layer 202 shown on the far left of the figure. It has been calculated that the requirement for the gradually decreasing dimensions is significantly relaxed to several hundred nanometers due to the presence of layer 203. For example, for tip widths greater than 100 nanometers, a coupling efficiency of 70% or greater between 201 and 202 can be achieved even if the refractive index difference between 201 and 202 is greater than 1. Conversely, without layer 203, layer 201 would have to be progressively smaller so that its mode could be directly coupled into layer 202, and for similar coupling efficiency, the progressively smaller size of layer 201 (not shown) would have to be much smaller than 100 nanometers. In another embodiment, a cone (not shown) is formed in layer 203 instead of layer 202. In yet another embodiment, for efficient coupling, cones can be formed in both layers 202 and 203. In some embodiments, the cones in layers 202 and 203 can be multi-level cones, meaning they utilize more than one etch depth to facilitate more efficient coupling.

[0042] Additionally, in some embodiments, one or more interfaces between layers 201, 206 and / or 203 are angled to reduce the corresponding back reflections.

[0043] Angle 220 is the angle between the tangent to the wave propagation direction inside structure 201 and the exit surface of 201 (or its interface junctions 206 and / or 203 if layer 206 is absent). Angle 220 is primarily used to control the back reflection of the mode supported by layer 201 as it reaches the interface junctions 206 and / or 203, and its precise value is the result of numerical or experimental optimization. In some embodiments, it ranges from 0° to 80°.

[0044] Angle 230 is the angle between the wave propagation direction within structure 201 and the angle of the waveguide formed by 203. This angle is an optimization parameter for the coupling efficiency between the modes supported by layers 201 and 203, and is related to the selection of angle 220 and / or the refractive index of the materials used in layers 201 and 203 and their respective claddings. In some embodiments, it ranges from 0° to 80°.

[0045] Precise vertical alignment between the axis defined by the wave propagation direction inside structure 201 and the center of waveguide 203 at the interface to 206 and / or 201. Figure 2 The "up / down" in the text refers to optimization parameters, where this offset can be positive (e.g., up / down). Figure 2 (upper) and negative (upper) Figure 2 (in the lower part) and / or essentially equal to 0 (no offset). This optimization is performed directly with numerical software to maximize the performance of the transition while optimizing the angle 230.

[0046] Prior to this invention, the requirement for a gradually decreasing width for direct transmission between layers 101 / 201 and 102 / 202 would have been problematic, i.e., without the intermediate layer 103 / 203. However, the use of an intermediate layer 103 / 203, butt-coupled to layers 101 / 201 (although with an angled interface in some embodiments), significantly reduces the stringent requirement for a gradually decreasing width, thereby allowing efficient transmission between very high refractive index materials (e.g., GaAs- and / or InP-based layers in 101 / 201) and low refractive index materials (e.g., SiN, LiNbO3, or similar materials in layer 102 / 202). Layer 203 may comprise a dielectric, polymer, and / or any other suitable material. At the butt-coupled interface, no adiabatic transition occurs between the optical modes supported by elements 101 / 201 and 103 / 203. Layers 103 / 203 and / or 102 / 202 may include bends (not shown) to control the routing of the guided light waves. One or more photolithographic alignment marks 240 are present to facilitate precise alignment between layers formed during various processing steps.

[0047] Figure 3 Three embodiments of the invention are depicted, shown in cross-sectional end views 300, 330, and 360 in regions where nonlinear interactions are optimized.

[0048] The nonlinearity can be an inherent property of the material forming layer 302, and optimization can involve designing the geometry of layer 302 to maximize the effect. The nonlinearity can also be designed, or alternatively, by providing electrodes 309a and 309b to apply an electric field and induce a stronger nonlinearity. In some embodiments (not shown), electrodes 309a and 309b are absent, and the inherent material properties of layer 302 provide an efficient nonlinear effect.

[0049] Layer 302, which provides the core function for the passive waveguide formed therein, is surrounded by cladding material 307, which may include multiple sublayers, such as, for example, by means of Figure 1Layers 104, 107, and 108 are described. In some embodiments, a cladding layer exists on top of substrate 305. In other embodiments not shown, substrate 305 may serve as at least a portion of the cladding layer for a waveguide formed in layer 302.

[0050] In the embodiment shown in view 300, the two electrodes are placed laterally across the width of the waveguide to generate an electric field when a voltage is applied between the electrodes. In the embodiment shown in view 330, the two electrodes are placed vertically across the depth of the waveguide to generate an electric field when a voltage is applied between the electrodes. In some embodiments, the substrate 305 is conductive and is used to increase the electric field strength at waveguide 302. In the embodiment shown in view 360, the two electrodes are placed vertically to generate an electric field when a voltage is applied between the electrodes, similar to the case in view 330, but in 360, both electrodes have contacts positioned on the same side of the wafer (as shown at the top).

[0051] Various other embodiments can be envisioned that provide similar functionality, with the goal of applying an optimized electric field to the core of the passive waveguide defined across layer 102, including utilizing the conductivity in layer 102.

[0052] Figure 4 Two embodiments of the invention are depicted, shown in top views 400 and 450 in regions where nonlinear interactions are optimized. In these embodiments, periodic electrodes are used to enhance nonlinearity and provide quasi-phase matching. As those skilled in the art will appreciate, the pitch of the periodic electrodes should conform to quasi-phase matching and / or resonance conditions at the fundamental frequency and transition frequency (in the case of a resonator) to optimize frequency transition.

[0053] In the embodiment shown in view 400, two sets of electrodes are utilized. One set of electrodes 409a, shown in white, provides the anode, and another set of electrodes 409b, shown in black, provides the cathode. The sets are positioned such that one type of electrode is on one side of waveguide 402, for example, above waveguide 402 as shown, and the other type, shown in black, is on the opposite side (below waveguide 402 as shown).

[0054] In the embodiment shown in view 450, two sets of electrodes are also utilized, but arranged differently. One set of electrodes 409a provides the anode, and the other set of electrodes 409b provides the cathode, but these two sets of electrodes are positioned in a staggered, periodic configuration. In the illustrated embodiment, the staggering exists on both sides (top and bottom) of the waveguide 402, but in other cases, it may be on only one side. The pitch between each set of electrodes is again chosen to optimize the frequency conversion.

[0055] Figure 5Two embodiments of the invention, shown in top-down views 500 and 550, are depicted in regions where their nonlinear interactions are optimized. In these embodiments, the interaction length between the optical pump signal and the frequency conversion signal is increased, resulting in more efficient nonlinear frequency generation, as will be described below.

[0056] In the embodiment shown in view 500, the interaction length is increased by designing a longer waveguide structure 502, which is typically arranged in a helical manner. In regions optimized for frequency conversion, the waveguide structure can be implemented to utilize… Figure 3 and 4 Any of the electrode configurations shown. In some embodiments, waveguide crossings are avoided, and a double helix as depicted in view 500 is utilized. Optical pump signal (λ) p The signal enters the waveguide structure 502 at the input port 505, and as it propagates along the waveguide structure 502, a frequency-converted signal (λ) is generated using the nonlinearity of the waveguide material. NL The pump signal and the frequency conversion signal both exist at output 506, and their precise power ratio is defined by the efficiency of the nonlinear process. In other embodiments (not shown), crossovers can be used to increase the density of the structure (total length to total area). In other embodiments, other geometries are designed to be most suitable for manufacturing.

[0057] In the embodiment shown in view 550, a resonant structure (such as, for example, a ring resonator, a racetrack resonator, or a similar structure) is used to increase the interaction between the pump and frequency-converting signals by utilizing the effective cavity length of the resonator and / or the power accumulation within the resonator. Various resonator configurations (such as, for example, a drop-down resonator 570 (as shown in view 550), an all-pass resonator (not shown), multiple coupling regions designed with a single ring (more than two) and / or multiple rings (with one or more coupling points)) can be utilized to optimize performance in terms of conversion efficiency. The waveguide structure of the resonator 570 in the region optimized for frequency conversion can be implemented using... Figure 3 and Figure 4 Any electrode configuration shown. Optical pump signal (λ) p The signal enters the waveguide structure 552 at input port 555 and, as it propagates along the waveguide structure 552, is at least partially coupled to the resonator 570. The precise coupling strength is an optimized parameter for device efficiency, and the uncoupled portion exits the structure at output 556. Utilizing the nonlinearity and power accumulation within the resonator, a frequency-converted signal (λ) is generated within the resonator 570. NLThe frequency-converted signal is efficiently coupled to bus waveguide 553 and exits the structure at output port 557. A portion of the pump signal may also be present at output port 557, but output port 557 is primarily used as the output of the frequency-converted signal and is optimized for this purpose. Similarly, output port 556 may also have a portion of the frequency-converted signal, but output port 556 is primarily used as the output of any remaining unconverted pump signal and is optimized for this purpose. Optimization is typically achieved by adjusting the wavelength-sensitive coupling ratio of the resonator.

[0058] Figure 4 Either of the two periodic electrode pattern cases shown can be used for optimization. Figure 5 The increased interaction length introduces nonlinear effects, although only one is explicitly shown in view 550. Arrows with short dashed lines indicate optical pumping (λ). p The direction of the frequency conversion signal (λ) is defined by a direct-emission laser with integrated nonlinear elements, and the arrow with a long dashed line indicates the frequency conversion signal (λ). NL The direction of the coupling structure can be optimized to promote higher efficiency, as is known to those skilled in the art in the case of the resonator method.

[0059] Figure 6 Three embodiments of the invention are depicted, shown in top views 600, 630, and 660 in regions where nonlinear interactions are optimized. In these embodiments, tuner elements are used to optimize nonlinear effects in the passive waveguide, resulting in more efficient nonlinear frequencies with improved phase-matching conditions.

[0060] In view 600, a layer 602 is shown (e.g., by means of, for example) Figure 3 The device, as described in layer 302, includes an optimized waveguide and at least one tuner element 610 that implements phase-matching control. Such a device may include, via... Figure 3-5 Additional electrodes described (not shown).

[0061] In view 630, a device with a helical geometry (similar to the geometry shown in view 500) is shown, having an optimized waveguide implemented in layer 632 and at least one tuner element 640 that implements phase-matching control. Such a device may include, by means of… Figure 3-5 Additional electrodes described (not shown).

[0062] In view 660, a device with a resonant structure (similar to the resonant structure shown in view 550) is illustrated. This device has an optimized waveguide implemented in layer 662 and at least one tuner element 670, which implements phase-matching control. Such a device may include, by means of… Figure 3-5 Additional electrodes described (not shown).

[0063] In all cases, the tuner can be optimized in terms of layout, length, and / or placement to provide better efficiency or other desired characteristics, as is known in the field of designing photon-based tuners. In some embodiments, thermal tuning effects are utilized.

[0064] Figure 7 Views 700 and 750 depict two embodiments of a laser utilizing the photonic integrated circuit (PIC) of the present invention. In both embodiments, a standalone laser is coupled to a resonator-based nonlinear element. A standalone laser is defined as a laser that emits laser light without any additional components, meaning it includes one or more gain elements forming a laser cavity and a mirror, wherein in some embodiments, the mirror can also be used to select a single longitudinal mode to create a single-frequency laser. In other embodiments, a grating is formed in the gain element to realize a single-frequency (DFB) laser. In other embodiments, the mirror is tunable, thereby forming a widely tunable laser.

[0065] View 700 illustrates an embodiment that utilizes a standalone laser coupled to a nonlinear resonator structure to provide nonlinear frequency conversion. In the case of critical coupling, no power output is to the right side of the resonator in the orientation shown in the figure; however, in most cases, a small fraction of the power incident on the resonator is transmitted to the right because the extinction ratio is not infinite. The signal coupled to the resonator is frequency-converted and coupled out at the second coupling port of the resonator (output to the lower right as shown). Similar functionality can be provided by using only one coupling port (not shown) to the resonator or by three or more coupling points (not shown) to the resonator. A phase tuner is an optional component that can be included to promote higher performance of the PIC.

[0066] In view 750, the overall arrangement is similar to that in view 700; however, in this case, due to intentional defects or general manufacturing flaws, there is backscattering from the pump wavelength of the resonator, in either case causing the laser injection to lock onto the resonator. This injection locking has been shown to improve the laser linewidth if the resonator quality factor is sufficiently high. High-Q resonators supported by modern semiconductor processes have demonstrated linewidth improvements of several orders of magnitude or even greater. General operation is similar to the arrangement shown in view 700, while the arrangement shown in view 750 generally has a lower linewidth. A phase tuner is an optional component that can be included to promote higher performance of the PIC.

[0067] Figure 8 Views 800, 830, and 860 depict three embodiments of a laser based on a photonic integrated circuit (PIC) according to the present invention. Figure 7 The illustrated embodiments provide a contrast, in which the nonlinear element is part of the laser cavity. In all embodiments, the phase tuner is an optional element that can contribute to higher performance of the PIC.

[0068] In view 800, a nonlinear resonator is used as one of the mirrors due to intentional defects in the resonator or general manufacturing flaws that cause back reflection of the gain element. The second mirror required to complete the cavity can be implemented in a variety of ways, including gratings, ring mirrors, ring resonators, and / or various combinations thereof.

[0069] In view 830, two separate mirrors are used to form the cavity together with the gain. Both mirrors can be designed to provide optimal response, for example, to accommodate the high reflectivity (λ) of the optical pump. p and lower reflectivity (λ) for frequency-converted signals. NL In this way, the laser threshold can be optimized while simultaneously improving the extraction efficiency of the frequency-converted signal. A nonlinear element is inserted between the two mirrors to provide efficient frequency conversion. These mirrors can be tunable to provide wavelength tuning for PIC-based lasers.

[0070] In view 860, an arrangement similar to that in the device shown in view 830 is used, but multiple nonlinear resonator elements are used to support broadband tuning and / or higher conversion efficiency.

[0071] It should be understood that these illustrative embodiments teach only some examples of PIC-based lasers utilizing the present invention, and many other similar arrangements are conceivable. Furthermore, such a PIC-based laser can be combined with a number of other components to provide additional functionality or better performance, such as various filtering elements, amplifiers, monitoring photodiodes, modulators, and / or other photonic components.

[0072] Embodiments of the present invention offer numerous advantages. This integrated platform enables scalable fabrication of PICs made from a variety of materials, thereby providing higher performance and / or the ability to operate at wavelengths challenging for locally direct-emission lasers through frequency conversion using nonlinearity. Furthermore, this platform is capable of handling high optical power compared to typical Si-based or InP-based PICs.

[0073] Previous methods typically employ a gradually decreasing structure to transfer the optical mode from an active III / V semiconductor device (101) to a passive device (102), where the width of the compound semiconductor region is adiabatically reduced to a submicron size. However, as the refractive index difference increases, the required gradually decreasing width rapidly shrinks to tens of nanometers. This invention deploys a docking coupling scheme to eliminate interference with the compound semiconductor waveguide (…). Figure 1 The need for very small, gradually decreasing dimensions in 101 (of which) simplifies the manufacture of this structure.

[0074] Other methods have relied on attaching pre-fabricated active optical devices to the die of a passive waveguide. This requires extremely high alignment accuracy, often exceeding the capabilities of typical die-bonded devices. This limits the overall processing power of the process and the performance of the optical coupling.

[0075] This invention utilizes wafer bonding, typically involving a capping sheet of compound semiconductor material on a carrier wafer having a dielectric waveguide (e.g., by means of...). Figure 1 The described process flow and subsequent semiconductor manufacturing processes are as known in the art. It enables precise optical alignment between active and passive waveguides via conventional photolithography steps, eliminating the need for precise physical alignment. This photolithography-based alignment allows for scalable manufacturing using wafer-level techniques.

[0076] By using a docking coupling method combined with a mode converter (intermediate waveguide) to facilitate efficient optical transmission between different materials, this mode converter eliminates the need for a narrow, tapered tip that is challenging to solve and manufacture using existing technology tools.

[0077] It should be understood that the optical coupling between modes in the active and passive layers is mutual, such that... Figure 1 For example, this structure can be configured to facilitate optical transmission from region 101 to region 102, and also facilitate the reverse transmission from region 102 to region 101. It should be understood that multiple such transitions can be implemented on a properly configured PIC, without limitation on their number or orientation.

[0078] In some embodiments, the active device can utilize the substrate for more efficient heat dissipation because it is in direct contact with the substrate without a dielectric in between.

[0079] In some embodiments, the active device generates a hybrid waveguide structure having a dielectric layer, which can be used, for example, to create a wavelength-selective component formed within a laser cavity for, for example, a distributed feedback (DFB) laser or similar component.

[0080] Embodiments of the optical devices described herein can be incorporated into a variety of other devices and systems, including but not limited to various computing and / or consumer electronics devices / appliances, communication systems, sensors, and sensing systems.

[0081] It should be understood that this disclosure teaches only a few examples of illustrative embodiments, and many variations of the invention can be readily devised by those skilled in the art upon reading this disclosure, and the scope of the invention is determined by the appended claims.

Claims

1. An integrated photonic device, comprising: Fabrication of the first, second, and third elements on a common substrate; Wherein, the first element includes an active waveguide structure supporting a first optical mode, the second element includes a passive waveguide structure supporting a second optical mode, and the third element, which is at least partially docked to the first element, includes an intermediate waveguide structure supporting an intermediate optical mode. In this embodiment, at least a portion of the second element is nonlinear and supports frequency conversion; The gradually decreasing waveguide structure in at least one of the second and third elements facilitates efficient thermal switching between the second optical mode and the intermediate optical mode. Wherein, no adiabatic transformation occurs between any intermediate optical mode and the first optical mode; and Photolithographic alignment marks are used to define the mutual alignment of the first, second, and third elements, which facilitates precise alignment between layers formed during the processing steps of manufacturing the first, second, and third elements; The nonlinear portion of the second element includes a spiral-based delay structure configured to enhance frequency conversion, and the spiral-based delay structure in the second element includes at least two electrodes. The nonlinear portion of the second element includes at least two electrodes configured to maximize the frequency conversion caused by the electric field applied by the two electrodes; and - Wherein, the electrode is vertically offset from the core of the passive waveguide in the passive waveguide structure in a plane perpendicular to the bottom surface of the common substrate, and at least one of the electrodes utilizes the substrate conductivity to optimize the electric field distribution, or - Wherein, the nonlinear portion of the second element includes at least two sets of electrodes, the electrodes in each set being separated by a pitch selected to optimize frequency conversion, and at least one set of electrodes in the set interleaved with the other set of electrodes in the set.

2. The integrated photonic device according to claim 1, in, The third element is directly overlaid on the planarized top surface achieved through planarization prior to the deposition of the third element.

3. The integrated photonic device according to claim 1, in, The interface between the first and third elements is at an angle, which is optimized to minimize reflection.

4. The integrated photonic device according to claim 1, in, The nonlinear portion of the second element has waveguide dimensions optimized for maximizing the frequency conversion.

5. The integrated photonic device according to claim 1, in, The electrode is laterally offset from the core of the passive waveguide in the passive waveguide structure in a plane parallel to the bottom surface of the common substrate.

6. The integrated photonic device according to claim 1, in, The nonlinear portion of the second element includes a resonant structure configured to enhance frequency conversion.

7. The integrated photonic device according to claim 6, in, The resonant structure in the second element includes at least two electrodes.

8. The integrated photonic device according to claim 1, in, The nonlinear portion of the second element includes at least one tuner element configured to adjust phase matching and enhance frequency conversion.

9. The integrated photonic device according to claim 6, in, The resonant structure in the nonlinear portion of the second element includes at least one tuner element.

10. The integrated photonic device according to claim 1, in, The spiral-based delay structure in the nonlinear portion of the second element includes a tuner element.

11. The integrated photonic device according to claim 1, in, The first element includes a GaAs laser; The second element includes a SiN waveguide structure; The third element comprises a material with a refractive index lower than SiN and provides efficient optical coupling between the first and second elements; and At least a portion of the output from the GaAs laser is frequency-doubled at the second element.

12. The integrated photonic device according to claim 11, in, The nonlinear structure of the second element includes a resonant structure configured to enhance the frequency harmonics.

13. The integrated photonic device according to claim 1: in, The first element includes an InP laser; The second element includes a SiN waveguide structure; The third element comprises a material with a refractive index lower than SiN that provides efficient optical coupling between the first and third elements; and At least a portion of the output from the InP laser is frequency-doubled at the second element.

Images