Optical device

By removing the bottom filler from the edge coupler region between the optical package and the support substrate, the warping problem caused by the difference in thermal expansion coefficients is solved, improving the alignment accuracy and coupling efficiency between the fiber array unit and the edge coupler.

CN224985634UActive Publication Date: 2026-11-17TAIWAN SEMICONDUCTOR MANUFACTURING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202522383175.1
Authority / Receiving Office
CN · China
Patent Type
Utility models(China)
Current Assignee / Owner
Priority Date
2024-11-12
Filing Date
2025-11-10
Publication Date
2026-11-17
Estimated Expiration
2035-11-10

AI Technical Summary

Technical Problem

In the prior art, the difference in thermal expansion coefficients between the optical package and the supporting substrate leads to warping problems, affecting the stability and alignment accuracy of the device.

Method used

By removing the edge coupler region of the underfill between the optical package and the support substrate, the mechanical forces caused by the difference in thermal expansion coefficients are reduced, thereby reducing the risk of warpage.

Benefits of technology

It effectively reduces warpage between the optical package and the supporting substrate, improves the alignment accuracy and coupling efficiency between the fiber array unit and the edge coupler, and reduces insertion loss.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN224985634U_ABST
    Figure CN224985634U_ABST
Patent Text Reader

Abstract

The utility model relates to an optical device, including forming a plurality of first metal structures on a first region of a base surface of a die, and forming a second metal structure on a second region of the base surface of the die, wherein the second metal structure has a different shape than the first metal structure, and the second metal structure has a strip shape. The method can further include bonding the die to a support substrate with the plurality of first metal structures, and applying an underfill between the base surface of the die and the support substrate. In some embodiments, the underfill extends from a top surface of the support substrate to a sidewall of the die as viewed in cross-section.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of this utility model relate to an optical device. Background Technology

[0002] Electronic signaling and processing is a technology used for signal transmission and processing. In recent years, optical signaling and processing has been used in an increasing number of applications, especially for signal transmission in fiber optic applications.

[0003] Optical signaling and processing is often combined with electronic signaling and processing to provide fully-fledged applications. For example, optical fibers can be used for long-distance signal transmission, while electrical signals can be used for short-distance signal transmission, processing, and control. Therefore, devices integrating long-distance optical elements and short-distance electronic elements are formed for the conversion between optical and electrical signals, as well as the processing of both. The package can therefore include an optical (photonic) die containing the optical components and an electronic die containing the electronic components. Utility Model Content

[0004] This invention provides an optical device comprising: a substrate; an optical package bonded to the substrate via a plurality of connectors formed on a substrate surface of the optical package, wherein the optical package includes an optical die and an electronic die connected to the optical die, and the optical die includes at least one waveguide, and the electronic die includes at least one active device; an underfill disposed between the optical package and the substrate and surrounding the plurality of connectors, wherein the underfill is disposed in a first region on the substrate surface, and a second region of the optical package is adjacent to the first region, the second region being free of the underfill; and an optical fiber array unit optically connected to the optical package.

[0005] This invention provides an optical device comprising: a substrate; an optical package bonded to the substrate via a plurality of connectors formed on a substrate surface of the optical package, wherein the optical package includes an optical die and an electronic die connected to the optical die, and the optical die includes at least one waveguide, and the electronic die includes at least one active device; an underfill disposed between the optical package and the substrate and surrounding the plurality of connectors, wherein a first end of the underfill is located below the substrate surface of the optical package; and an optical fiber array unit optically connected to the optical package. Attached Figure Description

[0006] The best understanding of the present invention will be achieved by reading the following detailed description in conjunction with the accompanying drawings. It should be noted that, according to industry standard practice, the various features are not drawn to scale. In fact, for clarity of explanation, the dimensions of the various features may be arbitrarily increased or decreased.

[0007] Figures 1 to 8 An optical package comprising an edge coupler is shown according to some embodiments.

[0008] Figure 9 The diagram illustrates the formation of connection bumps, such as C4 solder bumps, and edge dams on the substrate of an optical package according to some embodiments.

[0009] Figure 10 The geometry of a connecting bump array, such as a C4 solder bump array, an edge dam geometry, and the positioning of a polymer layer on the surface of an optical packaging substrate are shown according to some embodiments.

[0010] Figure 11 The image illustrates the positioning of an optical package, including connecting bumps and edge dams, on a support substrate prior to bonding, according to some embodiments.

[0011] Figure 12 The diagram illustrates how an optical package is bonded to a support substrate via a connection engagement provided by a connection bump, according to some embodiments.

[0012] Figure 13 The illustration shows, according to some embodiments, the application of underfill to the space between a support substrate and an optical package, wherein an edge dam prevents the underfill from reaching one side of the package, which is connected to a fiber array unit (FAU) via an edge coupler.

[0013] Figure 14 The diagram illustrates how, according to some embodiments, fiber array units (FAUs) are connected to an edge coupler of an optical package. Detailed Implementation

[0014] This utility model provides numerous different embodiments or examples for implementing various features of the utility model. Specific examples of components and arrangements are described below to simplify the utility model. Of course, these are merely examples and are not intended to be limiting. For example, the following description of a first feature formed on or on a second feature may include embodiments in which the first and second features are formed in direct contact, and may also include embodiments in which additional features may be formed between the first and second features, thereby preventing direct contact between the first and second features. Furthermore, reference numerals and / or letters may be repeated in various embodiments of the utility model. Such repetition is for the purpose of brevity and clarity, and does not itself indicate a relationship between the various embodiments and / or configurations discussed.

[0015] Furthermore, for ease of explanation, this invention may use spatially relative terms such as "beneath," "below," "lower," "above," "upper," and similar expressions to describe the relationship between one device or feature shown in the figures and another device or feature. These spatially relative terms are intended to encompass different orientations of the device during use or operation, in addition to those shown in the figures. The device may have other orientations (rotated 90 degrees or in other orientations), and the spatially relative descriptive terms used in this invention can be interpreted accordingly.

[0016] Embodiments will now be discussed with reference to certain examples, in which an edge dam structure is used to control the application of underfill material between the bonded optical package and the support substrate. The coefficient of thermal expansion (CTE) of the optical package may differ from that of the support substrate to which the optical package is bonded. In some cases, due to this difference in CTE, mechanical forces may arise between the optical package and the support substrate when one of the structures experiences dimensional changes due to thermal temperature, as the optical package and support substrate are connected by the underfill. This can cause warpage in one or both of the support substrate and the optical package. It has been determined that some of the negative effects of device warpage can be minimized by removing the underfill from the area where the device's edge coupler connects to the fiber array unit (FUA).

[0017] However, the embodiments presented are intended to be illustrative and are not intended to limit the ideas presented to the precise embodiments described. Rather, the ideas presented can be incorporated into a wide range of embodiments, and all such embodiments can be included within the overall scope of the disclosure.

[0018] Reference Figure 1 It shows the optical interposer 100 (see Figure 5 The initial structure of ) according to some embodiments. Figure 1 In the specific embodiment shown, the optical interposer 100 is a photonic integrated circuit (PIC) and at this stage includes a first substrate 101, a first insulating layer 103, and material 105 for the first active layer 201 of the first optical element 203 (in... Figure 1 Not shown separately in the text, but in the following section about Figure 2 (Further explanation and discussion follow). In one embodiment, at the start of the manufacturing process of the optical interposer 100, the first substrate 101, the first insulating layer 103, and the material 105 of the first active layer 201 for the first optical element 203 may together be part of a silicon-on-insulator (SOI) substrate. First, consider the first substrate 101, which may be a semiconductor material such as silicon or germanium, a dielectric material such as glass, or any other suitable material that allows support for the overlay device.

[0019] The first insulating layer 103 may be a dielectric layer separating the first substrate 101 from the overlying first active layer 201, and in some embodiments, it may also be part of a covering material surrounding a subsequently fabricated first optical element 203 (discussed further below). In one embodiment, the first insulating layer 103 may be silicon oxide, silicon nitride, germanium oxide, germanium nitride, combinations thereof, or the like, formed using methods such as implantation (e.g., forming a buried oxide (BOX) layer), or deposited onto the first substrate 101 using methods such as chemical vapor deposition, atomic layer deposition, physical vapor deposition, combinations thereof, or the like. However, any suitable materials and fabrication methods may be used.

[0020] The material 105 of the first active layer 201 is initially (before patterning) a conformal material layer for initiating the fabrication of the first active layer 201 of the first optical element 203. In one embodiment, the material 105 of the first active layer 201 may be a translucent material that can be used as a core material for the desired first optical element 203, such as a semiconductor material like silicon, germanium, silicon-germanium, combinations thereof, or the like, while in other embodiments, the material 105 of the first active layer 201 may be a dielectric material such as silicon nitride or the like, although in other embodiments, the material 105 of the first active layer 201 may be a III-V material, lithium niobate material, or a polymer. In embodiments where the material 105 of the first active layer 201 is deposited, the material 105 of the first active layer 201 may be deposited using methods such as epitaxial growth, chemical vapor deposition, atomic layer deposition, physical vapor deposition, combinations thereof, or similar methods. In other embodiments where the first insulating layer 103 is formed using an implantation method, the material 105 of the first active layer 201 may initially be part of the first substrate 101 prior to the implantation process for forming the first insulating layer 103. However, the material 105 of the first active layer 201 can be formed using any suitable material and manufacturing method.

[0021] Figure 2 As shown, once the material 105 of the first active layer 201 is ready, the first optical element 203 of the first active layer 201 is fabricated using the material 105 of the first active layer 201. In embodiments, the first optical element 203 of the first active layer 201 may include, for example, optical waveguides (e.g., ridge waveguides, rib waveguides, buried channel waveguides, diffused waveguides, etc.), couplers (e.g., grating couplers, edge couplers of narrowed waveguides with widths between about 1 nanometer and about 200 nanometers, etc.), directional couplers, optical modulators (e.g., Mach-Zehnder silicon-photonic switches, microelectromechanical switches, microring resonators, etc.), amplifiers, multiplexers, demultiplexers, photoelectric converters (e.g., PN junctions), electro-optic converters, lasers, combinations thereof, or the like. However, any suitable first optical element 203 may be used.

[0022] To begin forming the first active layer 201 of the first optical element 203 from initial material, the material 105 of the first active layer 201 may be patterned into the desired shape of the first active layer 201 of the first optical element 203. In one embodiment, the material 105 of the first active layer 201 may be patterned using, for example, one or more photolithographic masking and etching processes. However, any suitable method may be used to pattern the material 105 of the first active layer 201. For some first optical elements 203, such as waveguides or edge couplers, the patterning process may be all or at least most of the manufacturing processes used to form these first optical elements 203.

[0023] Figure 3 As shown, for components requiring further manufacturing processes, such as Mach-Zehnder silicon photonic switches utilizing resistance heating elements, additional processing can be performed before or after material patterning of the first active layer 201. For example, implantation processes, additional deposition and patterning processes of different materials (e.g., resistance heating elements, III-V materials for converters), combinations of all these processes, or similar processes can be used to assist in the further fabrication of various desired first optical elements 203. In one particular embodiment, such as Figure 3 Specifically, in some embodiments, epitaxial deposition of a semiconductor material 301 (e.g., germanium for electro / optical signal modulation and conversion) can be performed on a patterned portion of the material 105 of the first active layer 201. In such embodiments, the semiconductor material 301 can be epitaxially grown to assist in the fabrication of a photodiode, for example, for a photoelectric converter. All these fabrication processes and all suitable first optical elements 203 can be fabricated, and all these combinations are fully intended to be included within the scope of the embodiments.

[0024] In one embodiment, the first optical element 203 includes one or more edge couplers 203a. The edge coupler 203a may also be referred to as a butt coupling, wherein the optical fiber is horizontally aligned with the edge coupler 203a leading to the waveguide. The edge coupler 203a enables high coupling efficiency, wide bandwidth, and polarization independence. In some embodiments, the fiber-to-chip edge coupler may include an inverse taper structure. In some embodiments, corresponding to the direction of light propagation, an inverse taper refers to a tapered waveguide whose width gradually increases along the mode propagation direction, meaning that the narrow end of the taper is close to the optical fiber, while the wide end is connected to the photonic waveguide.

[0025] The mode distribution is determined by the mode order and waveguide structure. For a specific mode, a waveguide with a specific cross-sectional area can support the entire mode. The fundamental transverse electric (TE) mode is primarily used in optical communication applications to transmit information. In some embodiments, the size of the silicon photonic waveguide is chosen to support the fundamental TE mode propagating with negligible loss. For example, the waveguide may have dimensions of approximately 200 nm high and approximately 500 nm wide. Waveguides with too small a cross-sectional area cannot support the entire fundamental mode, so the mode will be partially distributed in the outer region of the waveguide. Conversely, waveguides with too large a cross-sectional area will easily excite undesirable higher-order modes. A tapered shape is suitable for mode switching because the gradually changing cross-sectional area supports mode switching and mode size changes.

[0026] In some embodiments, the narrow end of the inverted tapered edge coupler 203a may have a smaller cross-sectional area than the intended mode size, thus failing to fully confine the incident mode, and a significant proportion of the electromagnetic field is distributed around the tapered tip. As the tapered width increases, it can support the entire mode and completely confine the electromagnetic field within the tapered shape. Overall, in some embodiments, the inverted tapered edge coupler 203a, with its narrow tip aligned with the fiber core, can convert large modes from the fiber into compressed guided modes in the photonic waveguide.

[0027] In some embodiments, the basic structure of the edge coupler is, for example Figure 3 The edge coupler 203a depicted is based on a single inverse taper and can switch modes over a length of several hundred micrometers. In some embodiments, the taper may have a linear profile. Common variations of the linear taper include multi-segment tapers, parabolic tapers, or quadratic tapers, as well as exponential inverse tapers.

[0028] In some embodiments, at least the tip of the edge coupler 203a may be composed of silicon nitride (SiN). In some embodiments, the edge coupler 203a composed of silicon nitride (SiN) may be formed on the same layer as the first optical element 203 composed of the material 105 of the first active layer 201, which in some examples is silicon (Si). To provide an edge coupler 203a composed of a different constituent material than the other first optical elements 203, a protective mask protecting the other first optical elements 203 may first be formed. An etching process may then be performed to remove the material 105 of the first active layer 201 from the area where the edge coupler 203a will be formed. In a subsequent process sequence, the material of the edge coupler 203a may then be formed and patterned to provide the edge coupler 203a, which may include at least one silicon nitride (SiN) tip.

[0029] Figure 4As shown, once the individual first optical elements 203 of the first active layer 201 are formed, a second insulating layer 401 can be deposited to cover the first optical elements 203 and provide additional encapsulation material. In one embodiment, the second insulating layer 401 may be a dielectric layer separating the individual elements of the first active layer 201 from each other and from the upper structure, and may additionally serve as another portion of the encapsulation material surrounding the first optical elements 203. In one embodiment, the second insulating layer 401 may be silicon oxide, silicon nitride, germanium oxide, germanium nitride, combinations thereof, or the like, formed using chemical vapor deposition, atomic layer deposition, physical vapor deposition, combinations thereof, or the like. Once the material of the second insulating layer 401 is deposited, it can be planarized using, for example, a chemical mechanical polishing process to planarize the top surface of the second insulating layer 401 (in embodiments where the second insulating layer 401 is intended to completely cover the first optical element 203), or to planarize the second insulating layer 401 with the top surface of the first optical element 203. However, any suitable materials and manufacturing methods can be used.

[0030] Figure 5 As shown, once the first optical element 203 of the first active layer 201 and the edge coupler 203a are fabricated, and the second insulating layer 401 is formed, a first metallization layer 501 is formed to electrically connect the first active layer 201 of the first optical element 203 to the control circuit, to each other, and to the subsequently attached device. Figure 5 Not shown in the text, but below regarding Figure 6 (Further explanation and description). In one embodiment, the first metallization layer 501 is formed of alternating layers of dielectric and conductive materials and can be formed by any suitable process (e.g., deposition, damascene, dual damascene, etc.). In a particular embodiment, there may be multiple metallization layers for interconnecting various first optical elements 203, but the exact number of first metallization layers 501 depends on the design of the optical interposer 100.

[0031] Furthermore, during the fabrication of the first metallization layer 501, one or more second optical elements 503 may be formed as part of the first metallization layer 501. In some embodiments, the second optical elements 503 of the first metallization layer 501 may include elements such as couplers (e.g., edge couplers, grating couplers, etc.) for connecting external signals, optical waveguides (e.g., ridge waveguides, rib waveguides, buried channel waveguides, diffused waveguides, etc.), optical modulators (e.g., Mach-Zehnder silicon photonic switches, microelectromechanical switches, microring resonators, etc.), amplifiers, multiplexers, demultiplexers, photoelectric converters (e.g., PN junctions), electro-optic converters, lasers, combinations thereof, or the like. However, any suitable optical element may be used for one or more second optical elements 503.

[0032] In one embodiment, one or more second optical elements 503 may be formed by initially depositing a material for one or more second optical elements 503. In one embodiment, the material for one or more second optical elements 503 may be a dielectric material, such as silicon nitride, silicon oxide, combinations thereof, or the like, or a semiconductor material, such as silicon, deposited using deposition methods such as chemical vapor deposition, atomic layer deposition, physical vapor deposition, combinations thereof, or the like. However, any suitable material and any suitable deposition method may be used.

[0033] Once the material for one or more second optical elements 503 is deposited or otherwise formed, the material can be patterned into the desired shape of the one or more second optical elements 503. In one embodiment, the material for one or more second optical elements 503 can be patterned using, for example, one or more photolithographic masking and etching processes. However, any suitable method can be used to pattern the material for one or more second optical elements 503.

[0034] For certain elements of one or more second optical elements 503, such as waveguides or edge couplers, the patterning process may be all or at least most of the fabrication used to form these elements. Furthermore, for elements requiring further fabrication processes, such as Mach-Zehnder silicon photonic switches utilizing resistance-heated elements, additional processing may be performed before or after patterning the material of one or more second optical elements 503. For example, implantation processes, additional deposition and patterning processes of different materials, combinations of all these processes, or the like, may be used to aid in the further fabrication of various desired one or more second optical elements 503. All these fabrication processes and all suitable one or more second optical elements 503 can be fabricated, and all these combinations are fully intended to be included within the scope of the embodiments.

[0035] Once one or more second optical elements 503 of the first metallization layer 501 have been fabricated, a first bonding layer 505 is formed on the first metallization layer 501. In one embodiment, the first bonding layer 505 may be used for dielectric-to-dielectric and metal-to-metal bonding. According to some embodiments, the first bonding layer 505 is formed of a first dielectric material 509, such as silicon oxide, silicon nitride, or the like. The first dielectric material 509 may be deposited using any suitable method, such as CVD, high-density plasma chemical vapor deposition (HDPCVD), PVD, atomic layer deposition (ALD), or the like. However, any suitable materials and deposition processes may be used.

[0036] Once the first dielectric material 509 is formed, a first opening is formed in the first dielectric material 509 to expose the conductive portion of the underlying layer, preparing for the formation of a first bonding pad 507 within the first bonding layer 505. Once the first opening is formed in the first dielectric material 509, a seed layer and electroplated metal can be filled into the first opening to form the first bonding pad 507 within the first dielectric material 509. The seed layer can be deposited over the top surface of the first dielectric material 509, the exposed conductive portion of the underlying layer, and the sidewalls of the opening and the second opening. The seed layer may contain a copper layer. The seed layer can be deposited using processes such as sputtering, evaporation, or plasma-enhanced chemical vapor deposition (PECVD), depending on the desired material. Electroplated metal can be deposited on the seed layer using electroplating processes such as electroplating or electroless plating. The electroplated metal may contain copper, copper alloys, or the like. The electroplated metal can serve as a filler material. Prior to the seed layer, a barrier layer (not shown separately) may be deposited over the top surface of the first dielectric material 509 and the sidewalls of the opening and the second opening. The barrier layer may comprise titanium, titanium nitride, tantalum, tantalum nitride, or the like.

[0037] After filling the first opening, a planarization process, such as CMP, is performed to remove excess portions of the seed layer and electroplated metal, forming a first bonding pad 507 within the first bonding layer 505. In some embodiments, bonding pad vias (not shown separately) may also be used to connect the first bonding pad 507 to the underlying conductive portion, and the first bonding pad 507 to the first metallization layer 501 via the underlying conductive portion.

[0038] Furthermore, the first bonding layer 505 may also include one or more third optical elements 511 integrated within the first bonding layer 505. In this embodiment, before depositing the first dielectric material 509, one or more third optical elements 511 may be fabricated using methods and materials similar to those used for one or more second optical elements 503 (as described above), such as waveguides and other structures at least partially formed by deposition and patterning processes. However, any suitable structure, material, and manufacturing method may be used.

[0039] Figure 5 The components depicted provide a photonic integrated circuit (PIC) 800. A photonic integrated circuit (PIC) 800 is an electronic component that contains photonic elements that work with light (photons). Figures 10 to 14 The photonic integrated circuit (PIC) 800 depicted in the image contains more detail than Figure 5 The figures are depicted sparingly only to simplify the illustrations and to discuss other elements of the present invention. This does not imply the scope of the above description, but is not intended to imply the inclusion of other elements of the present invention. Figures 10 to 14 Elements in the photonic integrated circuit 800 depicted in the diagram are omitted.

[0040] Figure 6 The bonding of a first semiconductor device 601 to a first bonding layer 505 of an optical interposer 100 is illustrated. In some embodiments, the first semiconductor device 601 is an electronic integrated circuit (EIC, e.g., a device without optical components) 600 and may have a semiconductor substrate 603, an active device layer 605, an overlying interconnect structure 607, a second bonding layer 609, and associated second bonding pads 611. In one embodiment, the semiconductor substrate 603 may be similar to the first substrate 101 (e.g., a semiconductor material such as silicon or silicon-germanium), the active device 605 may be a transistor, capacitor, resistor, etc., formed on the semiconductor substrate 603, the interconnect structure 607 may be similar to the first metallization layer 501 (without optical components), the second bonding layer 609 may be similar to the first bonding layer 505, and the second bonding pad 611 may be similar to the first bonding pad 507. However, any suitable device may be used.

[0041] In one embodiment, the first semiconductor device 601 may be configured to work in conjunction with the optical interposer 100 to achieve the desired functionality. In some embodiments, the first semiconductor device 601 may be a high-bandwidth memory (HBM) module, xPU, logic die, 3DIC die, CPU, GPU, SoC die, MEMS die, a combination thereof, or similar device. Any suitable device with any appropriate functionality may be used, all of which are fully encompassed within the scope of this embodiment.

[0042] In one embodiment, the first semiconductor device 601 and the first bonding layer 505 may be bonded using dielectric-to-dielectric and metal-to-metal bonding processes. In specific embodiments using dielectric-to-dielectric and metal-to-metal bonding processes, the process may be initiated by activating the surfaces of the second bonding layer 609 and the first bonding layer 505. Activating the top surfaces of the first bonding layer 505 and the second bonding layer 609 may include dry processing, wet processing, plasma processing, exposure to inert gas plasma, exposure to H2, exposure to N2, exposure to O2, combinations thereof, or similar processes, as examples. In embodiments using wet processing, RCA cleaning may be used, for example. In another embodiment, the activation process may include other types of processing. The activation process facilitates the bonding of the first bonding layer 505 and the second bonding layer 609.

[0043] After the activation process, the optical interposer 100 and the first semiconductor device 601 can be cleaned using, for example, chemical rinsing. The first semiconductor device 601 is then aligned and positioned to physically contact the optical interposer 100. Heat treatment and contact pressure are then applied to the optical interposer 100 and the first semiconductor device 601 to bond the optical interposer 100 to the laser die. For example, the optical interposer 100 and the first semiconductor device 601 can be subjected to pressures of about 200 kPa or less, and temperatures between about 25°C and about 250°C to melt the optical interposer 100 and the first semiconductor device 601. The optical interposer 100 and the first semiconductor device 601 can then be subjected to temperatures at or above the eutectic point of the materials of the first bonding pad 507 and the second bonding pad 611, for example, between about 150°C and about 650°C, to melt the metal. In this way, the optical interposer 100 and the first semiconductor device 601 form a dielectric-to-dielectric and metal-to-metal bonding device. In some embodiments, the bonded grains are subsequently baked, annealed, pressed, or otherwise treated to strengthen or complete the bond.

[0044] Furthermore, while specific processes for initiating and strengthening the bond have been described, these descriptions are illustrative and not intended to limit the embodiments. Instead, any suitable combination of baking, annealing, pressing, or other processes may be utilized. All of these processes are fully intended to be included within the scope of the embodiments.

[0045] Figure 6 It is also shown that once the first semiconductor device 601 has been bonded, a first gap filler material 613 (referred to in some embodiments as an oxide-containing filler material) is deposited to fill the space around the first semiconductor device 601 and provide additional support. In one embodiment, the first gap filler material 613 may be a material such as silicon oxide, silicon nitride, silicon oxynitride, combinations thereof, or the like, deposited to fill and overfill the space around the first semiconductor device 601. However, any suitable material and deposition method may be used.

[0046] Once the first gap filler 613 has been deposited, it can be planarized to expose the first semiconductor device 601. In one embodiment, the planarization process may be a chemical mechanical planarization process, a polishing process, or a similar process. However, any suitable planarization process may be used.

[0047] Figures 10 to 14 The electronic integrated circuit (EIC) 600 described in the text contains more than Figure 6 The drawings depict fewer details, intended only to simplify the illustrations in order to discuss other elements of the present invention. It is not intended to replicate the foregoing description. Figures 10 to 14 The elements in the electronic integrated circuit 600 depicted in the diagram are omitted from their existence.

[0048] The photonic integrated circuit (PIC) 800 and the first semiconductor device 601 providing the electronic integrated circuit (EIC) 600 can be elements of a compact universal photonic engine (COUPE) chip. The COUPE uses light instead of electricity to perform communication, allowing for faster data transmission and reduced energy consumption compared to traditional electronic-based systems. In some embodiments, the COUPE chip can combine the electronic integrated circuit (EIC) 600 with the photonic integrated circuit (PIC) 800 using System-on-Chip (SoIC) packaging technology. SoIC includes 3D chip-to-chip (3D IC) stacking technologies for integrating dies separated from system-on-chip (SoC) and system technology, such as dielectric-to-dielectric and metal-to-metal bonding processes. The photonic integrated circuit (PIC) 800 is designed to utilize the unique properties of light, offering advantages such as high bandwidth, low power consumption, and faster data transmission speeds compared to its electronic counterparts. These circuits typically include elements such as waveguides, couplers, lasers, light-emitting diodes (or other coherent light sources), modulators, detectors, and other optical components, such as mirrors and reflectors. However, any suitable components can be used. In contrast, the first semiconductor device 601 providing the electronic integrated circuit (EIC) 600 uses electronic devices, such as transistors and capacitors, on only a single chip. In some embodiments, the COUPE (Compact General Purpose Photonic Engine) can provide performance enhancements in high-performance computing (HPC) applications by enabling faster processing of large amounts of data. The COUPE uses light instead of electricity to perform communication, allowing for faster data transmission and reduced energy consumption compared to traditional electronic-based systems.

[0049] By incorporating an edge coupler 203a within the coupe chip, the edge coupler 203a provides a wider bandwidth and higher coupling efficiency than a grating coupler (GC). In some embodiments, the edge coupler 203a is an ideal approach to fulfilling the optical input / output (I / O) development roadmap in a coupe chip configuration that provides a photonic integrated circuit (PIC) 800 and a first semiconductor device 601 providing an electronic integrated circuit (EIC). Minimizing the warpage of the optical package 900 and the supporting substrate 950 is crucial for realizing the fiber array unit (FAU) 980 (such as...). Figure 14(As shown) Optimal alignment of the silicon nitride tip of the edge coupler 203a with the optical package 900 presents a challenge in order to provide high coupling efficiency and low insertion loss. The optical package 900 may also be referred to as a die. Furthermore, the die structure of the optical package may include an e-die structure (also referred to as an electronic integrated circuit 600) containing active devices, and a p-die structure (also referred to as a photonic integrated circuit (PIC) 800) containing waveguides.

[0050] Figure 7 The illustration shows the attachment of a semiconductor substrate 701 (referred to in some embodiments as a silicon substrate) to a first semiconductor device 601 and a first gap-filling material 613. In one embodiment, the semiconductor substrate 701 may be a support material transparent to the desired wavelength of light, such as silicon, and may be bonded using, for example, an adhesive (not in contact with...). Figure 7 (Shown separately) Attachment. However, in other embodiments, the semiconductor substrate 701 may be attached to the first semiconductor device 601 and the first gap filler material 613 using, for example, a bonding process. Any suitable method may be used to attach the semiconductor substrate 701.

[0051] Figure 8 The diagram illustrates the removal of a first substrate 101 and optionally a first insulating layer 103, thereby exposing a first active layer 201 of a first optical element 203 and an edge coupler 203a. In one embodiment, planarization processes such as chemical mechanical polishing, grinding, one or more etching processes, combinations thereof, or similar methods can be used to remove the first substrate 101 and the first insulating layer 103. However, any suitable method can be used to remove the first substrate 101 and / or the first insulating layer 103.

[0052] Once the first substrate 101 and the first insulating layer 103 are removed, the second active layer 801 of the fourth optical element 803 can be formed on the back side of the first active layer 201. In one embodiment, the second active layer 801 of the fourth optical element 803 can be used with the second optical element 503 of the first metallization layer 501 (as described above regarding...). Figure 5 Similar materials and similar processes are used to form the fourth optical element 803. For example, the second active layer 801 of the fourth optical element 803 may be formed by alternating cladding material layers (such as silicon oxide) and core material layers (such as silicon nitride), using deposition and patterning processes to form optical elements such as waveguides.

[0053] Figure 9The diagram illustrates the formation of a first through-device via (TDV) 901 and a third bonding layer 903 to form an optical package 900. In one embodiment, the first TDV 901 extends through the second active layer 801 and the first active layer 201 to provide a fast channel for power, data, and ground through the optical interposer 100. In one embodiment, the first TDV 901 can be formed by initially forming a TDV opening in the optical interposer 100. The TDV opening can be formed by applying and developing a suitable photoresist (not shown) and removing exposed portions of the second active layer 801 and the optical interposer 100.

[0054] Once the through-device via opening is formed in the optical interposer 100, the through-device via opening can be covered by a liner. The liner can be, for example, an oxide formed from tetraethylorthosilicate (TEOS) or silicon nitride, but any suitable dielectric material can be used instead. The liner can be formed using a plasma-enhanced chemical vapor deposition (PECVD) process, but other suitable processes such as physical vapor deposition or thermal processes can also be used.

[0055] Once a liner is formed on the sidewalls and bottom of the through-hole opening, a barrier layer (not shown separately) can be formed, and the remainder of the through-hole opening can be filled with a first conductive material. The first conductive material may contain copper, but other suitable materials such as aluminum, alloys, doped polysilicon, or combinations thereof may also be used. The first conductive material can be formed by electroplating copper onto a seed layer (not shown), filling and overfilling the through-hole opening. Once the through-hole opening is filled, excess liner, barrier layer, seed layer, and first conductive material outside the through-hole opening can be removed by a planarization process such as chemical mechanical polishing (CMP), but any suitable removal process may be used.

[0056] Optionally, in some embodiments, once the first through-hole 901 is formed, a second metallization layer electrically connected to the first through-hole 901 can be formed (in... Figure 9 (Not shown separately). In one embodiment, the second metallization layer may be formed as described above with respect to the first metallization layer 501, for example, by using a damascene process, a dual damascene process, or a similar process to form alternating dielectric and conductive material layers. In other embodiments, the second metallization layer may be formed and shaped using an electroplating process, and then the conductive material may be covered with a dielectric material. However, any suitable structure and manufacturing method may be used.

[0057] A third bonding layer 903 is formed to provide an electrical connection between the optical interposer 100 and subsequently attached devices. In one embodiment, the third bonding layer 903 may be similar to the first bonding layer 505, for example having a third bonding pad 909 (similar to the first bonding pad 507) or even a fifth optical element 911 (similar to the third optical element 511). However, any suitable device may be used.

[0058] Figure 9 An embodiment is shown in which a connection bump 913 (such as a C4 solder bump) and an edge dam 915 are formed on an optical package 900 substrate. The connection bump 913 provides a conductive area for the third bonding pad 909 and other external devices (such as a support substrate 950, etc.). Figures 11 to 14 Contact between (shown). In some embodiments, the connection bump 913 provides multiple connections between the optical package 900 and the support substrate 950. The connection bump 913 may be a conductive bump (e.g., a C4 bump, a ball grid array, a microbump, etc.) or a conductive post using materials such as solder and copper. In embodiments where the connection bump 913 is a contact bump, the connection bump 913 may contain materials such as tin, or other suitable materials such as silver, lead-free tin, or copper.

[0059] An example process sequence for forming the connecting bump 913 may begin by forming a polymer layer 914 on the optical package 900 substrate. In some embodiments, the polymer layer 914 may be a polyimide or other type of insulating material. For example, the polymer layer 914 may be an organic polyimide film, such as a photosensitive polyimide film. Polyimide (sometimes abbreviated as PI) is a polymer containing imide groups. An example polyimide may be produced by the condensation of pyromelliticdianhydride and 4,4'-oxydianiline.

[0060] In some embodiments, a conductive barrier layer (not shown) may be formed on the contact point (e.g., the third bonding pad 909) prior to the formation of the polymer layer 914. In some embodiments, the conductive barrier layer is adapted to adhere to the passivation layer and bond with the solder bump. Depending on the selected bumping process, this layer may be deposited locally (e.g., vapor-deposited bumps) or may be a continuous thin film. In some embodiments, this layer is also referred to as ball-limiting metallurgy (BLM) and under-bump metallurgy (UBM). Note that the conductive barrier layer may be omitted.

[0061] In some examples, a polymer layer 914 (e.g., photosensitive polyimide) may be deposited on a third bonding layer 903 (and optionally a conductive barrier layer) having a third bonding pad 909. The polymer layer 914 may be deposited using a deposition method such as spin coating. In embodiments, the thickness of the polymer layer 914 may be between about 5 and 10 micrometers; however, other sizes are also contemplated in this invention.

[0062] In some embodiments, when the polymer layer 914 is composed of photosensitive polyimide, the material is exposed and developed to form vias corresponding to the third bonding pad 909. The via openings may be arranged in an array on the first side 916 of the optical package 900. The via openings are then filled with connecting bumps 913. Figure 10 and Figure 11 As shown, the connecting bump 913 provides a connection between the optical package 900 and the support substrate 950.

[0063] Reference Figure 9 In some embodiments, the polymer layer 914 may also be patterned and developed to remove portions of the polymer layer 914 (e.g., a photosensitive polyimide layer) from the second side 918 of the optical package 900. The second side 918 of the optical package 900 provides the bonding location for the fiber optic array unit (FAU) 980, which will be referred to later. Figure 14 The following description is provided. It has been determined that, in some examples, the polymer layer 914 may shrink. When attached to the substrate surface of the optical package 900, the shrinkage of the polymer layer 914 can exert stress on the optical package 900, causing it to warp. Warping of the optical package 900 is one of the factors affecting the alignment of the edge coupler 203a integrated into the optical package 900 with the fiber optic array unit (FAU) fibers subsequently connected, where misalignment of the fiber optic array unit (FAU) can lead to insertion loss. It has been determined that removing the polymer layer 914 from the second side 918 of the optical package 900 can reduce the aforementioned stress exerted by the polymer layer 914 on the second side 918 of the optical package 900.

[0064] In some embodiments, the polymer layer 914 may also be patterned and developed to form trenches that will be later filled to provide an edge dam 915. In some embodiments, the trenches may have a rectangular geometry that extends across the entire width of the via opening array. The trench width of the edge dam 915 is selected to continuously extend across the width of the optical package 900 region where the connecting bump 913 is located. The edge dam 915 will be formed within the trench patterned into the polymer layer 914. The pattern used to form the trench in the polymer layer 914 that defines the location of the edge dam 915 may also be removed from the back surface 920 of the optical package 900. However, a portion of the polymer layer 914 may remain in the second side 918 of the optical package near the outer sidewall of the edge dam 915.

[0065] The edge dam 915 is positioned to prevent the bottom filler 975, which is subsequently applied to fill the space between the first side 916 of the optical package 900 and the support substrate 950, from reaching the second side 918 of the optical package 900, as described below. Figure 13 As stated above. It has been determined that when the bottom filler 975 is present in the substrate 950 (the substrate is in... Figure 10 and Figure 11 When the optical package 900 and the substrate 950 are connected, mechanical stress may occur due to the difference in the coefficients of thermal expansion (TECs) between them. This mechanical stress caused by the difference in thermal expansion between the optical package 900 and the substrate 950 can lead to warping of the optical package 900. Warping of the second side 918 of the optical package 900 may negatively affect the alignment of the edge coupler 203a in the optical package 900 with the fiber 985 of the fiber array unit (FAU) 980, such as... Figure 14 As shown. By preventing the bottom filler from filling the space between the substrate 950 and the optical package 900 at the second side 918 of the optical package 900, the warping effect caused by the difference in the coefficients of thermal expansion (TECs) of the optical package 900 and the substrate 950 can be eliminated or at least minimized.

[0066] It is worth noting that the patterning of the via openings for connecting the bumps 913, the grooves for the edge dams 915, and the removal of the polymer layer 914 from the second side 918 of the optical package 900 can be performed using the same photolithographic pattern. However, other embodiments are also contemplated, in which at least one of the via openings for connecting the bumps 913, the grooves for the edge dams 915, and the removal of the polymer layer 914 from the second side 918 of the optical package 900 is patterned using a separate photomask.

[0067] Figure 9 The diagram illustrates the formation of a connecting bump 913 and an edge dam 915. The connecting bump 913 can be formed using an electroplating process, a bumping process, or a combination of electroplating and bumping processes. In some embodiments, the connecting bump 913 may include solder bumps, such as C4 solder bumps, or the connecting bump 913 may include a copper pillar 912. In some embodiments, the connecting bump 913 may include a copper pillar 912 in conjunction with a solder cap 917. Although Figures 9 to 14 A connecting bump 913 including a copper pillar 912 and a solder cap 917 is shown, but the copper pillar 912 may be omitted, and a solder element may be formed on the third bonding pad 909 as a C4 solder bump.

[0068] In embodiments where the connection bump 913 is a solder bump (such as a C4 solder bump), the material layer of the connection bump 913 can be locally deposited, such as an evaporated bump, or it can be a continuous thin film, such as when current needs to be transferred through the wafer during electroplating. The solder bump can be composed of tin solder. In some embodiments, when a barrier layer (such as a ball-confined metallization (BLM) layer) is present, the size and shape of the connection bump 913 after reflow are determined by the size of the barrier layer beneath the bump.

[0069] In some embodiments where the connector bump 913 is a C4 solder bump composed of solder bumps, the connector bump 913 can be formed by initially forming a tin layer on the third bonding pad 909 using common methods such as vapor deposition, electroplating, printing (e.g., screen printing), solder transfer, and balling. Other methods include solder spraying and pick-and-place solder transfer processes. After the bump manufacturing step, the wafer undergoes reflow, i.e., thermal cycling in a protective environment to melt the solder bumps and form spherical balls.

[0070] The connecting bump 913, including the copper pillar 912, is an alternative to a connecting bump 913 composed entirely of solder (such as the C4 solder bump described above). The structure and manufacturing process of the pillar-shaped bump 913 including the copper pillar 912 may include the integration of photolithography and electroplating processes. In some embodiments, a thin metal seed layer (not shown) may be sputtered onto the third bonding pad 909. In some examples, the thin metal seed layer may include an adhesive layer (typically Ti or TiW) and a copper seed layer as an electroplated conductive layer. The thin metal seed layer may be formed before or after the processing of the polymer layer 914, such as the deposition and patterning of the polymer layer 914. In some embodiments, the thin metal seed layer may be omitted.

[0071] In some embodiments, the copper pillar 912 may be electroplated onto the third bonding pad 909, for example, onto a thin metal seed layer (if present) present on the third bonding pad 909. In some examples, the electroplating may include an electroplating bath composed of copper sulfate, with copper chloride as an additive. Other additives are also added to the bath to improve deposit properties, such as grain refinement and ductility. In some embodiments, these additives include thiourea, polyethylene glycol, and surfactants. In some embodiments, a pyrophosphate bath may be used for high-speed electrodeposition of copper (Cu). In some embodiments, the combination of a high-speed electroplating bath and pulsed current allows for the deposition of different morphologies of copper (Cu) at high deposition rates of several micrometers (µm) per minute.

[0072] In some embodiments, a solder cap 917 may be formed on top of the copper pillar 912, followed by reflow to form a solder bump or cap. The solder cap 917 may include tin solder and may be formed using the solder bumping method described above.

[0073] In some embodiments, the connecting bumps 913 may be arranged in an array on a first side 916 of the optical package 900. The array may include rows and columns of connecting bumps 913.

[0074] Figure 9 An edge dam 915 is also depicted at the interface between the first side 916 and the second side 918 of the substrate surface of the optical package 900. In some embodiments, the location of the edge dam 915 defines a keep-out zone (KOZ) 925 between the edge dam 915 and the waveguide connection surface present on the sidewall 919 of the optical package 900 (first in...). Figure 13 (as depicted in the image). In some embodiments, the edge dam 915 has a continuous width (e.g., as shown in the image). Figure 10 The metal pillars (shown) extend across the entire width of the array of connecting bumps 913. In some examples, the edge dam 915 may be made of copper. However, other metal compositions, such as aluminum or tungsten, may also be used for the edge dam.

[0075] Edge dam 915 is deposited in grooves previously patterned in polymer layer 914 by means of electroplating or similar methods. Similar to the electroplating process used to form connecting bump 913 elements, such as the copper pillar 912 described above, the deposition surface of edge dam 915 may include a seed layer formed prior to the electroplating process for forming edge dam 915. In some examples, the thin metal seed layer may include an adhesive layer, typically Ti or TiW, and a copper seed layer as a conductive layer for electroplating. Edge dam 915 can be formed using an electroplating process similar to that described above for forming copper pillar 912 connecting bump 913 elements. However, edge dam 915 does not include solder cap 917. Edge dam 915 may have a strip geometry. The electroplating process is configured to provide an edge dam 915 with a height greater than the height of polymer layer 914. As further explained below, edge dam 915 does not directly contact support substrate 950. The height of the edge dam 915 is chosen to extend toward the support substrate 950, but does not contact the support substrate 950 when the support substrate 950 and the optical package 900 are bonded together, such as... Figure 12 As shown. This provides that the edge dam 915 will act as a barrier, or dam, to prevent the bottom filler 975 applied to the first side 916 of the optical package 900 from extending to the second side 918 of the optical package 900 to the extent that it fills the space between the optical package 900 and the support substrate 950 on the second side 918, as shown. Figure 13 Further details are provided below.

[0076] In some embodiments, the edge dam 915 and the connecting bump 913 may be formed using the same process sequence. In some embodiments, the edge dam 915 and the connecting bump 913 may be formed using different process sequences. In embodiments where the edge dam 915 and the connecting bump 913 are formed using different process sequences, a block mask, such as a photoresist mask, may be used to individually process the areas in the optical package 900 where the connecting bump 913 and the edge dam 915 may be formed.

[0077] Figure 10 yes Figure 9 The top-to-bottom view of the side cross-section of the structure depicted in the figure. Figure 10 The polymer layer 914 is further shown to be present on the first side 916 of the optical package 900, but removed from the second side 918 of the optical package 900. The second side of the optical package 900 includes the sidewall 919 of the optical package 900, as shown. Figure 14 As shown, a fiber array unit (FAU) will subsequently be attached thereto. The polymer layer 914 is shrinkable and has a different coefficient of thermal expansion than the optical package 900. Dimensional variations between the polymer layer 914 and the optical package 900 can result in mechanical stress, causing the optical package 900 to warp. Warping of the optical package 900 can affect the alignment of the fiber of the fiber array unit with the edge coupler 203a of the optical package 900, adversely leading to insertion loss. Removing the polymer layer 914 from the second side 918 of the optical package 900 creates an exposed back surface 920 of the optical package 900 on the second side 918. By removing the polymer layer 914 from the second side 918 of the optical package 900, the mechanical stress caused by polymer shrinkage can be removed from the portion of the optical package 900 that contacts the fiber array unit (FAU) 980, as... Figure 14 As shown, the fiber array unit (FAU) 980 will then be attached. Preventing the optical package 900 from warping on the side in contact with the fiber array unit (FAU) advantageously minimizes insertion loss.

[0078] Figure 10The geometry of the array of connection bumps 913 and the geometry of the edge dam 915 are also shown. The connection bumps 913 exist only on the first side 916 of the optical package 900. The second side 918 of the optical package 900 does not include the connection bumps 913. Removing the connection bumps 913 from the second side 918 of the optical package 900 is another mechanism to minimize edge warping of the optical package 900. The array of connection bumps 913 may include columns and rows of connection bumps 913, providing a quadrilateral geometry, such as a rectangle. The spacing of the connection bumps 913 may be equal to 150 micrometers. The width W1 of the edge dam 915 is chosen to continuously extend across the width W2 of the array of connection bumps 913. The edge dam 915 is positioned to prevent subsequently formed underfill 975 (used to fill the space between the first side 916 of the optical package 900 and the support substrate 950) from filling into the second side 918 of the optical package 900, which will be referenced below. Figure 13 A more detailed description will follow.

[0079] Figure 11 and Figure 12 This illustrates that after forming the connecting bumps 913 and the edge dams 915, the optical package 900 can be bonded to the support substrate 950. Figure 11 The optical package 900, including connecting bumps 913 and edge dams 915, is positioned on a support substrate 950 prior to bonding. Figure 12 The optical package 900 is shown to be bonded to the support substrate 950 via a connection engagement 930 (e.g., solder engagement) provided by a connection bump 913.

[0080] The support substrate 950 can be used to couple the optical package 900 with other devices to form, for example, a die-on-wafer-on-substrate (CoWoS) structure. ® The support substrate 950 may be a printed circuit board (PCB). The support substrate 950 may include a plurality of substrate contact pads 951. The support substrate 950 may also contain doped or undoped bulk silicon, or an active layer of a silicon-on-insulator (SOI) substrate. Typically, an SOI substrate contains a layer of semiconductor material, such as silicon, germanium, silicon-germanium, SOI, silicon-germanium-on-insulator (SGOI), or combinations thereof. Other substrates that may be used include multilayer substrates, gradient substrates, or mixed-orientation substrates.

[0081] Optionally, active devices (not shown separately) may be added to the support substrate 950. Active devices may include various active and passive devices, such as capacitors, resistors, inductors, etc., and may be used to generate the structural and functional requirements necessary for the design of the support substrate 950. Active devices may be formed inside or on the surface of the support substrate 950 using any suitable method.

[0082] In some embodiments, the bonding between the optical package 900 and the support substrate 950 may include aligning solder elements (e.g., solder caps 917) connected to the connection bumps 913 of the optical package 900 with the substrate contact pads 951 of the support substrate 950, such as... Figure 11 As shown.

[0083] After alignment, the connecting bump 913 can be brought into contact with the substrate contact pad 951 under temperature and pressure to achieve bonding between the two via a bonding joint 930. In some embodiments, the solder of the solder cap 917 can be used, for example, a reflow process to provide bonding between the copper pillar 912 and the substrate contact pad 951. For example, the optical package 900 and the support substrate 950 can be fused to the support substrate 950 via the bonding joint 930 under a pressure of about 200 kPa or less and a temperature between about 25 degrees Celsius and about 250 degrees Celsius. In some embodiments, the optical package 900 and the support substrate 950 can subsequently be baked, annealed, pressurized, or otherwise treated to strengthen or complete the bonding. In some embodiments, the solder cap 917 can be omitted. In some embodiments, the bonding between the two copper pillar structures can be provided by thermal compression bonding (TCB).

[0084] After the optical package 900 is bonded to the support substrate 950, there is a space between the base surface of the optical package 900 and the upper surface of the support substrate 950. After bonding, the edge dam 915 extends from the base surface of the optical package 900 toward the upper surface of the support substrate 950, but the edge dam 915 is separated from the upper surface of the support substrate 950, for example, it does not directly contact the upper surface of the support substrate 950.

[0085] Figure 13 An embodiment is shown in which underfill 975 is applied to the space between a support substrate 950 and an optical package 900, wherein an edge dam 915 blocks the underfill 975 from filling the sides of the package, such as the second side 918 of the interface between the edge coupler 203a and the fiber optic unit (FAU) 980. The underfill 975 may be a thermosetting epoxy resin or polymer, applied to the connection joints 930 to protect them and reinforce the solder joints. The underfill 975 may flow using capillary action and may be heat-cured.

[0086] Underfill 975 may be introduced initially from a first side 916 of the optical package 900 and fill the space between the substrate surface of the optical package 900, the connection joint 930, and the upper surface of the support substrate 950. The underfill 975 spans the entire width of the first side 916, contacting an edge dam 915. The edge dam 915 prevents excessive inflow of the underfill 975 into the space between the optical package 900 and the support substrate 950 on the second side 918 of the optical package 900, thus avoiding filling the space on the second side 918. In some embodiments, the underfill 975 may completely fill the space between the optical package 900, the support substrate 950, and the connection joint 930. In some embodiments, after filling the space on the first side 916, the underfill 975 may extend to the first sidewall of the optical package 900. However, the underfill 975 will not appear on the second side 918, nor will it extend to the second sidewall of the optical package.

[0087] By blocking the bottom filler 975 from filling the space on the second side 918 of the optical package 900, the edge dam 915 forms an exclusion zone (KOZ) 925 on the second side 918 of the optical package 900. The exclusion zone (KOZ) 925 is a space, such as an air-filled void, between the bottom of the optical package 900 and the upper surface of the support substrate 950. More specifically, the bottom filler 975 material fills from the first side 916 of the optical package 900 to the edge dam 915, but does not fill the exclusion zone (KOZ) 925. In some embodiments, such as Figure 14 As shown, the space of the exclusion zone (KOZ) 925 extends from the edge dam 915 to the sidewall 919 of the optical package 900, where the fiber array unit (FAU) is subsequently connected. In some embodiments, the length of the exclusion zone (KOZ) 925 can range from 100 micrometers to 5000 micrometers.

[0088] In some embodiments, by forming an exclusion zone (KOZ) 925 on the second side 918 of the optical package 900, for example to prevent the underfill 795 from filling the space between the support substrate 950 and the optical package 900 on the second side 918, the method and structure described herein can avoid warping of the optical package 900 due to the difference in the coefficient of thermal expansion (CTE) between the support substrate 950 and the optical package 900. The coefficient of thermal expansion (CTE) of the optical package 900 may differ from that of the support substrate 950 to which it is bonded. In some cases, due to the difference in the coefficient of thermal expansion, mechanical forces may be generated between the optical package 900 and the support substrate 950 when one of the structures undergoes dimensional changes due to thermal temperature. This may cause one or both of the support substrate 950 and the optical package 900 to warp. It has been determined that removing the underfill 975 from the area of ​​the edge coupler 203a connected to the optical fiber in the device, for example by forming an exclusion zone (KOZ) 925 on the second side 918 of the optical package 900, can minimize some of the negative effects of device warping. For example, it has been determined that warping of the optical package 900 may cause misalignment between the edge coupler 203a and the fiber array unit (FAU) 980. Methods and structures for removing at least the underfill 975 from the space between the optical package 900 and the support substrate 950 on the second side 918 of the optical package 900 can minimize or substantially eliminate misalignment between the edge coupler 203a and the fiber array unit (FAU) 980. Therefore, using an edge dam 915 to block the underfill 975 from filling the space between the optical package 900 and the support substrate 950 on the second side 918 of the optical package 900 can reduce or eliminate insertion loss caused by warping due to the difference in the coefficient of thermal expansion (TEC) between the optical package 900 and the support substrate 950.

[0089] Figure 14 A fiber optic array unit (FAU) 980 is shown connected to interface with an edge coupler 203a of an optical package 900. The FAU 980 is positioned abutting the sidewall 919 of the optical package 900 on a second side 918, where an exclusion zone (KOZ) 925 exists between the substrate surface of the optical package 900 and the upper surface of the support substrate 950. By using an edge dam 915 to block the bottom filler 975 from filling the space in the exclusion zone 925 on the same side (i.e., the second side 918) of the optical package 900 as the FAU 980, warpage is minimized to eliminate potential misalignment between the fiber 985 of the FAU 980 and the edge coupler 203a. By avoiding misalignment between the FAU 980 and the edge coupler 203a, incident loss of the transmitted signal between the fiber 985 and the edge coupler 203a is minimized and / or eliminated.

[0090] In some embodiments, one or more optical fibers 985 are signal-coupled to the optical package 900 via a fiber optic array unit (FAU) 980. In some embodiments, the total number of optical fibers 985 may range from 2 to 40, although more optical fibers 985 may be added. According to an alternative embodiment, there is only a single optical fiber 985 in the fiber optic array unit (FAU) 980. The optical fiber 985 may be optically coupled to an edge coupler 203a. In some embodiments, a refractive index matching adhesive is applied and cured to secure the fiber optic array unit (FAU) 980 to the optical package 900. The refractive index of the refractive index matching adhesive may range from about 1.4 to 1.5.

[0091] By utilizing the method and structure described in this invention, co-packaged optical packages and support substrates can be designed to control die warpage. The method and structure described in this invention employ an edge dam to form an exclusion zone (KOZ) near the junction of the fiber array unit (FAU) and the optical package, wherein the exclusion zone (KOZ) does not contain bottom filler that fills the space between the optical package substrate and the support substrate. By minimizing the warpage of the co-packaged optical package and support substrate, a larger alignment window is provided for coupling the fiber of the fiber array unit to the edge coupler of the optical package.

[0092] In one embodiment, a method of manufacturing an optical device includes: forming a connecting bump on a first side of an optical package substrate surface; forming a dam between a second side of the optical package substrate surface and the first side, the location of the dam defining an exclusion region between the dam and a coupler surface present on the sidewall of the optical package; bonding the optical package to a support substrate via the connecting bump; and applying an underfill from the first side of the optical package to fill the space between the optical package, the connecting bump, and the support substrate, wherein the underfill fills the space from the first side of the optical package to the dam, but does not fill the exclusion region on the second side of the optical package.

[0093] In one embodiment, a method of manufacturing an optical device includes: forming a plurality of first metal structures on a first region of a grain substrate surface and forming a second metal structure on a second region of a grain substrate surface, wherein the second metal structure has a different shape compared to the first metal structures and has a strip-like shape; bonding a grain to a support substrate through the plurality of first metal structures; and applying an underfiller between the grain substrate surface and the support substrate, wherein, in a cross-sectional view, the underfiller extends from a top surface of the support substrate to a sidewall of the grain. In one embodiment, applying the underfiller includes injecting the underfiller into the first region in a direction toward the second region. In one embodiment, the underfiller injected into the first region fills the space between the plurality of first metal structures. In one embodiment, the second metal structure prevents the underfiller from filling the second region. In one embodiment, the plurality of first metal structures have a height greater than the height of the second metal structure. In one embodiment, the grain includes a first portion comprising an active device and a second portion comprising a waveguide. In one embodiment, a polymer layer is located on the substrate surface of the first region of the grain, wherein the polymer layer is not present in at least a portion of the second region of the grain. In one embodiment, the second metal structure is a metal column having a continuous width that extends across the entire width of the plurality of first metal structures.

[0094] In another embodiment, an optical device includes: a substrate; an optical package bonded to the substrate via a plurality of connectors formed on the surface of an optical package substrate, wherein the optical package includes an optical die and an electronic die connected to the optical die, the optical die including at least one waveguide and the electronic die including at least one active device; an underfill disposed between the optical package and the substrate and surrounding the plurality of connectors, wherein the underfill is disposed in a first region on the substrate surface and a second region of the optical package adjacent to the first region and without underfill; and a fiber array unit optically connected to the optical package. In one embodiment, the second region is adjacent to a coupling surface of the fiber array unit. In one embodiment, the second region without the underfill has a length ranging from 100 micrometers to 5000 micrometers. In one embodiment, the optical device further includes a dam structure located between the first region containing the underfill and the second region without the underfill. In one embodiment, the dam structure has a columnar geometry extending across the entire width of the plurality of connectors located in the first region. In one embodiment, the dam structure comprises copper.

[0095] In another embodiment, an optical device includes a substrate; an optical package bonded to the substrate by a plurality of connectors formed on the surface of an optical package substrate, wherein the optical package includes an optical die and an electronic die connected to the optical die, the optical die including at least one waveguide and the electronic die including at least one active device; an underfill disposed between the optical package and the substrate and surrounding the plurality of connectors, wherein a first end of the underfill is located below the surface of the optical package substrate; and a fiber array unit optically connected to the optical package. In one embodiment, the optical package includes a first sidewall and a second sidewall, wherein at least a portion of the first sidewall is covered by the underfill, and the second sidewall is without underfill. In one embodiment, the second sidewall is adjacent to the fiber array unit optically connected to the optical package, the second sidewall of the optical package being located on the side of the optical package opposite to the first sidewall. In one embodiment, the plurality of connectors are present in a first region between the optical package and the substrate, the first region being adjacent to the first sidewall filled with underfill. In one embodiment, a portion of a second region between the optical package and the substrate is adjacent to the second sidewall and is without underfill. In one embodiment, a dam structure is present between the first region and a portion of the second region without underfill.

[0096] The foregoing has outlined the features of several embodiments to enable those skilled in the art to better understand the nature of this invention. Those skilled in the art should understand that they can readily use this invention as the basis for designing or modifying other processes and structures to achieve the same purposes and / or attain the same advantages as the embodiments described in this invention. Those skilled in the art should also recognize that such equivalent constructions do not depart from the spirit and scope of this invention, and that they can make various changes, substitutions, and modifications without departing from the spirit and scope of this invention.

Claims

1. An optical device, characterized by, Comprising: a substrate; an optical package bonded to the substrate by a plurality of connections formed on a base surface of the optical package, wherein the optical package comprises an optical die and an electronic die connected to the optical die, and the optical die comprises at least one waveguide, the electronic die comprises at least one active device; an underfill disposed between the optical package and the substrate and surrounding the plurality of connections, wherein the underfill is disposed within a first region of the base surface, and a second region of the optical package is adjacent to the first region, the second region being free of the underfill; and a fiber array unit optically connected to the optical package. wherein the second region is adjacent to a coupling surface of the fiber array unit.

2. The optical device of claim 1, wherein, wherein the second region free of the underfill has a length in a range of 100 microns to 5000 microns.

3. The optical device of claim 1, wherein, further comprising a dam structure located between the first region containing the underfill and the second region free of the underfill.

4. The optical device of claim 1, wherein, wherein the dam structure has a columnar geometry extending across an entire width of the plurality of connections located in the first region.

5. The optical device of claim 4, wherein, Comprising:

6. An optical device, characterized by a substrate; an optical package bonded to the substrate by a plurality of connections formed on a base surface of the optical package, wherein the optical package comprises an optical die and an electronic die connected to the optical die, and the optical die comprises at least one waveguide, the electronic die comprises at least one active device; an underfill disposed between the optical package and the substrate and surrounding the plurality of connections, wherein a first end of the underfill is located below the base surface of the optical package; and a fiber array unit optically connected to the optical package. wherein the optical package comprises a first sidewall and a second sidewall, wherein at least a portion of the first sidewall is covered by the underfill, and the second sidewall is free of the underfill.

7. The optical device of claim 6, wherein, wherein the second sidewall is adjacent to the fiber array unit optically connected to the optical package, the second sidewall of the optical package being located on an opposite side of the optical package from the first sidewall.

8. The optical device of claim 7, wherein, wherein the plurality of connections are present in a first region between the optical package and the substrate, the first region being adjacent to the first sidewall filled with the underfill.

9. The optical device of claim 7, wherein, wherein a portion of the optical package and the substrate adjacent to the second sidewall is free of the underfill.

10. The optical device of claim 9, wherein, wherein the second sidewall is adjacent to the fiber array unit optically connected to the optical package, the second sidewall of the optical package being located on an opposite side of the optical package from the first sidewall. wherein the plurality of connections are present in a first region between the optical package and the substrate, the first region being adjacent to the first sidewall filled with the underfill. wherein a portion of the optical package and the substrate adjacent to the second sidewall is free of the underfill.