Optical Engine and its Fabrication Method

By using a fabrication method with virtual substrates and conductive pillars in the optical engine, the problems of heat dissipation, warpage, size, and signal loss in the optical engine were solved, achieving high-density I/O connections and excellent heat dissipation performance, thereby improving process yield and product reliability.

CN123124844APending Publication Date: 2026-11-17NAT CENT FOR ADVANCED PACKAGING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611365580.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-03
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing optical engines suffer from poor heat dissipation performance of EIC, large module warpage, large TMV size leading to large module volume and low IO density, and high high-speed signal loss.

Method used

By employing a virtual substrate and electronic integrated circuit fabrication method, a first conductive pillar is prefabricated on the virtual substrate, and a through-hole is formed in the encapsulation layer. Combined with the process sequence of "first mounting, then thinning", the virtual substrate can be removed and the electronic integrated circuit can be thinned, shortening the signal transmission path, increasing the I/O density and suppressing warpage.

Benefits of technology

It significantly improves the IO density of the optical engine, enhances heat dissipation, reduces high-speed signal loss, solves the module warpage problem, and improves process yield and product reliability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123124844A_ABST
    Figure CN123124844A_ABST
Patent Text Reader

Abstract

This application relates to the field of optical communication technology and discloses an optical engine and its fabrication method. The fabrication method includes: providing a virtual substrate and an electronic integrated circuit (EIC), wherein a first conductive pillar is disposed on the virtual substrate and a second conductive pillar is disposed on the electronic integrated circuit; the thickness of the virtual substrate is less than the thickness of the electronic integrated circuit; mounting the virtual substrate and the electronic integrated circuit on the same side of a substrate; encapsulating to form an encapsulation layer covering the electronic integrated circuit, the virtual substrate, and the first and second conductive pillars; thinning the encapsulation layer from a first surface; removing the substrate; and thinning the encapsulation layer from a second surface to remove the virtual substrate, exposing the second end of the first conductive pillar from the encapsulation layer, and thinning the electronic integrated circuit. Compared with related technologies, this application can solve the problems of poor heat dissipation performance of the EIC, large module warpage, large TMV size leading to large module volume and low IO density, and high high-speed signal loss in related technologies.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of optical communication technology, and in particular to an optical engine and its fabrication method. Background Technology

[0002] Co-Packaged Optics (CPO), or Optical I / O, has gained widespread attention from manufacturers in recent years. CPO switches, due to their advantages such as low cost, low power consumption, low latency, and low loss, are widely recognized as the future trend and will replace traditional pluggable switches. In CPO switches or optical I / O systems, the optical engine, as a core component, has become a key research focus in this field in recent years.

[0003] Currently, mainstream CPO optical engines adopt a 3D fan-out package structure that stacks photonic integrated circuits (PICs) on electronic integrated circuits (EICs). Among them, EICs include chips such as transimpedance amplifiers (TIAs), drivers, analog-to-digital converters (ADCs), digital-to-analog converters (DACs), and microcontroller units (MCUs), while PICs include photonic integrated circuits such as silicon photonic chips.

[0004] However, existing optical engines suffer from poor EIC heat dissipation, large module warpage, large TMV (Through Mold Via) size leading to large module size and low IO density, and high high-speed signal loss. Summary of the Invention

[0005] This application provides an optical engine and its fabrication method to solve the problems of poor heat dissipation performance of EIC, large module warpage, large TMV size leading to large module volume and low IO density, and high high-speed signal loss in related technologies.

[0006] In a first aspect, this application provides a method for fabricating an optical engine, comprising: A virtual substrate and an electronic integrated circuit are provided. A first conductive pillar is disposed on the virtual substrate, and a second conductive pillar is disposed on the electronic integrated circuit. The thickness of the virtual substrate is less than the thickness of the electronic integrated circuit. The virtual substrate and the electronic integrated circuit are mounted on the same side of the substrate; The package is encapsulated to form an encapsulation layer covering the electronic integrated circuit, the virtual substrate, the first conductive pillar, and the second conductive pillar; the encapsulation layer has a first surface and a second surface facing away from each other, and the first surface is facing away from the substrate; The encapsulation layer is thinned from the first surface to expose the first end of the first conductive post and the second conductive post from the encapsulation layer; Remove the substrate; The encapsulation layer is thinned from the second surface to remove the dummy substrate, exposing the second end of the first conductive pillar from the encapsulation layer, and the electronic integrated circuit is thinned.

[0007] Secondly, this application provides an optical engine, comprising: A chip unit includes an encapsulation layer, a first conductive post disposed in the encapsulation layer, and an electronic integrated circuit; the two ends of the first conductive post are exposed from a first side and a second side of the encapsulation layer, respectively; the electronic integrated circuit has a second conductive post disposed thereon and is exposed from the first side of the encapsulation layer. A photonic integrated circuit is mounted on the first side of the packaging layer and has a second wiring layer integrated on its surface. The second wiring layer is located between the photonic integrated circuit and the packaging layer and is electrically connected to the first conductive post and the second conductive post. The thickness of the chip unit is less than or equal to 100 μm.

[0008] Thirdly, this application provides an optical engine, which is manufactured based on the above-described preparation method provided in this application.

[0009] This application simplifies the process flow and shortens the packaging cycle by making the thickness of the virtual substrate smaller than that of the electronic integrated circuit (IC), requiring only one back-side thinning step to simultaneously remove the virtual substrate, expose the second end of the first conductive pillar, and thin the IC to the target thickness (e.g., less than or equal to 100 μm). Specifically, by pre-fabricating the first conductive pillar on the virtual substrate, the first conductive pillar can have a smaller size and smaller center-to-center spacing compared to traditional TMVs, thereby achieving more vertical interconnect channels per unit area, significantly improving I / O density, and enabling module miniaturization. The virtual substrate is completely removed during the thinning process, not occupying the thickness space of the final product, further facilitating module miniaturization. Simultaneously, the height of the first conductive pillar decreases as the IC is thinned, shortening the signal transmission path, suppressing parasitic effects, and thus reducing high-speed signal transmission loss. Furthermore, this application adopts a process sequence of "mounting first, then thinning". The electronic integrated circuit maintains its original thickness during mounting and has sufficient mechanical strength. After packaging is completed, thinning is performed from the back side, which fundamentally avoids the chip cracking problem caused by thinning before mounting in conventional processes, and significantly improves process yield and product reliability.

[0010] Furthermore, by thinning the chip unit to less than or equal to 100μm, the overall thickness of the chip unit enters the low warp range where warp decreases linearly with thickness, effectively suppressing module warp; at the same time, the reduction in the thickness of the electronic integrated circuit shortens the heat conduction path from its interior to the back side, improving the heat dissipation performance of the EIC. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the specific embodiments or related technologies of this application, the drawings used in the description of the specific embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0012] Figure 1 This is a partial cross-sectional view of the optical engine of the related technology; Figure 2 This is a flowchart of a method for fabricating an optical engine according to an embodiment of this application; Figure 3 This is a schematic diagram of the structure of the virtual substrate and the first conductive pillar according to an embodiment of this application; Figure 4 This is a schematic diagram of the structure of the electronic integrated circuit and the second conductive pillar according to an embodiment of this application; Figure 5 This is a flowchart of a method for preparing a virtual substrate according to an embodiment of this application; Figure 6 This is a process diagram of mounting a virtual substrate and an electronic integrated circuit onto a substrate according to an embodiment of this application; Figure 7 This is a process diagram after completing step S103 according to the embodiment of this application; Figure 8 This is a process diagram after completing step S104 according to the embodiment of this application; Figure 9 This is a process diagram showing the fabrication of the dielectric layer and the under-bump metal layer according to an embodiment of this application; Figure 10 This is a process diagram showing the completion of the first temporary substrate bonding according to an embodiment of this application; Figure 11 This is a process diagram after completing step S105 according to the embodiment of this application; Figure 12 This is a process diagram after completing step S106 according to the embodiments of this application; Figure 13 This is a graph showing the relationship between the thickness of the bottom chip and the amount of warpage. Figure 14 for Figure 13Enlarged view of a local area in the middle; Figure 15 This is a process diagram after step S108 is completed according to the embodiment of this application; Figure 16 This is a process diagram after completing step S110 according to the embodiments of this application; Figure 17 This is a process diagram after completing step S111 according to the embodiments of this application; Figure 18 This is a process diagram showing the completion of photonic integrated circuit mounting according to an embodiment of this application; Figure 19 This is a structural diagram of the final optical engine obtained after removing the second temporary substrate according to an embodiment of this application.

[0013] Figure label: 100, Virtual substrate; 110, First conductive pillar; 111, First end; 112, Second end; 120, First interposer; 200, Electronic integrated circuit; 210, Second conductive pillar; 220, Second interposer; 300, Substrate; 310, Release layer; 320, PI layer; 321, Alignment mark; 400, Encapsulation layer; 410, First surface; 420, Second surface; 430, Dielectric layer; 440, Under-bump metal layer; 450, Solder ball; 500, First redistribution layer; 510, Thermal bump; 520, Second dielectric layer; 530, Metal wiring layer; 600, Photonic integrated circuit; 610, Second redistribution layer; 620, Underfill; 710, First temporary substrate; 720, Third interposer; 730, Second temporary substrate; 740, Fourth interposer. Detailed Implementation

[0014] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the application and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the present application are shown in the drawings, not the entire structure.

[0015] In the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessarily obscuring the concepts of this application. Various structural schematic diagrams according to embodiments of this application are shown in the accompanying drawings. These drawings are not to scale, and some details are enlarged for clarity, and some details may be omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed. In the context of this application, when a layer / element is referred to as being "on" another layer / element, the layer / element may be directly on the other layer / element, or there may be an intermediate layer / element between them. Additionally, if a layer / element is "on" another layer / element in one orientation, then when the orientation is reversed, the layer / element may be "below" the other layer / element.

[0016] In related technologies, such as Figure 1 As shown, an optical engine includes a chip unit 010 and a photonic integrated circuit 020. The chip unit 010 includes a packaging layer 011, a through-mold via (TMV) 012 disposed in the packaging layer 011, an electronic integrated circuit 015, and a first redistribution layer 013 and a second redistribution layer 014 disposed on both sides of the packaging layer 011. The first redistribution layer 013 and the second redistribution layer 014 are electrically connected through the through-mold via 012 and are electrically connected to the electronic integrated circuit 015. The photonic integrated circuit 020 is mounted on the chip unit 010 and is electrically connected to the first redistribution layer 013.

[0017] The optical engines of the relevant technologies have at least the following problems: Firstly, it has poor heat dissipation performance. Electronic integrated circuits generate a lot of heat when they are working. However, because there is a layer of low thermal conductivity DAF film (Die Attach Film) on the bottom of the electronic integrated circuit, the heat is mainly conducted to the wiring layer below through the molding compound and DAF film. The heat dissipation path is long and the thermal resistance is high, which leads to excessive temperature rise of the electronic integrated circuit, affecting the reliability and performance of the system.

[0018] Secondly, the module warpage is significant. Because electronic integrated circuits involve various materials with mismatched coefficients of thermal expansion (CTE) (such as silicon, molding compound, copper, and DAF film), significant stress is generated during temperature changes, leading to substantial warpage of the entire module even at room temperature. For large-sized chips (such as 15mm × 13mm and above), the warpage problem is particularly severe, directly affecting optical coupling accuracy and subsequent soldering yield.

[0019] Third, the module size is large and the IO density is low. The TMV of related technologies is directly fabricated in the molding layer. Due to the material properties of the molding layer and the limitations of the photolithography aspect ratio, the critical dimension (CD) of the TMV is large, and the center-to-center spacing between adjacent TMVs is also large. This results in the TMV consuming a large chip area, which is not conducive to the miniaturization of the module and cannot support the application requirements of high channel number (i.e., high IO density).

[0020] Fourth, high-speed signal transmission loss is high. Due to the large height of the TMV and the long signal transmission path, the parasitic capacitance and inductance effects of the TMV itself are significant, resulting in increased high-frequency signal loss, making it difficult to meet the signal integrity requirements of high-speed optical modules (such as 800G and above).

[0021] To resolve at least one of the above technical issues, please refer to Figure 2 This application provides a method for fabricating an optical engine, comprising: Step S101: Provide a virtual substrate and an electronic integrated circuit.

[0022] The aforementioned virtual substrate serves as a temporary carrier during the manufacturing process. Its surface is used to prefabricate the first conductive pillar, and it is completely removed in subsequent processes through thinning, ultimately not remaining in the optical engine product structure. The virtual substrate is preferably made of silicon, but glass, ceramics, or other rigid materials can also be used.

[0023] Specifically, please refer to Figure 3 The virtual substrate 100 has a first conductive pillar 110 on its surface. This first conductive pillar 110 is a metal pillar structure (preferably a copper pillar). The first conductive pillar 110 is used to be embedded in the encapsulation layer after the subsequent encapsulation layer formation step, forming a through-mold via (TMV) to achieve vertical interconnection between wiring layers on both sides of the encapsulation layer. Additionally, a first interposer layer 120 is also provided on the surface of the virtual substrate 100 opposite to the first conductive pillar 110, for bonding to the substrate in subsequent steps. The first interposer layer 120 is a bonding dielectric layer, preferably made of at least one insulating material selected from silicon oxide, silicon nitride, polyimide (PI), or benzocyclobutene (BCB). In the subsequent substrate removal step, the first interposer layer 120 is removed along with the virtual substrate 100 and is not retained in the final product structure.

[0024] In some embodiments, the first conductive pillar 110 may be formed on the surface of the virtual substrate 100 using processes such as semiconductor photolithography and electroplating.

[0025] By forming the first conductive pillar 110 on the surface of the virtual substrate 100, the critical dimension (CD) of the first conductive pillar 110 can be controlled below 10 μm, and the center-to-center spacing between adjacent first conductive pillars 110 can be controlled below 30 μm. Compared with related technologies that fabricate TMVs directly in the encapsulation layer, the first conductive pillar 110 of this application has a smaller critical dimension and a smaller center-to-center spacing, thereby realizing more vertical interconnect channels per unit area and thus improving IO density.

[0026] It's important to note that I / O density refers to the number of input / output (IO) pins or vertical interconnect channels that can be arranged per unit area. I / O density is determined by the pitch between adjacent vertical interconnect channels; that is, the smaller the pitch, the more vertical interconnect channels can be arranged per unit area, and the higher the I / O density; conversely, the larger the pitch, the fewer vertical interconnect channels can be arranged per unit area, and the lower the I / O density.

[0027] Electronic integrated circuits are chips responsible for electrical signal processing in optical engines, including but not limited to one or more of the following: transimpedance amplifiers (TIA), drivers, analog-to-digital converters (ADC), digital-to-analog converters (DAC), microcontroller units (MCU), clock recovery circuits (CDR), and serial-to-parallel converters (SerDes).

[0028] Specifically, please refer to Figure 4The electronic integrated circuit 200 is provided with a second conductive post 210, which is used to realize the electrical connection between the electronic integrated circuit 200 and external circuits, such as providing power and ground to the electronic integrated circuit 200, and inputting and outputting high-speed signals between the electronic integrated circuit 200 and the photonic integrated circuit 600. Exemplarily, there are two second conductive posts 210, one of which serves as a power supply pin to provide the operating voltage to the electronic integrated circuit 200; the other serves as a ground pin to provide a ground connection to the electronic integrated circuit 200, thereby enabling the electronic integrated circuit 200 to obtain the power and ground required for normal operation. However, this application is not limited to this. In practical applications, the number of second conductive posts 210 can be set to other numbers according to the functional requirements of the electronic integrated circuit 200. For example, it may also include second conductive posts 210 for inputting and outputting high-speed electrical signals between the electronic integrated circuit 200 and the photonic integrated circuit 600. Those skilled in the art should understand that the number and arrangement of the second conductive posts 210 can be flexibly designed according to the model, function, and package layout requirements of the electronic integrated circuit 200. In addition, a second interposer 220 is disposed on the surface of the electronic integrated circuit 200 opposite to the second conductive post 210 for bonding to the substrate in subsequent steps. The second interposer 220 is a bonding dielectric layer, preferably made of at least one insulating material selected from silicon oxide, silicon nitride, polyimide (PI), or benzocyclobutene (BCB). During the subsequent thinning step of the packaging layer, the second interposer 220 is removed during the thinning of the back side of the electronic integrated circuit 200 and is not retained in the final product structure.

[0029] The thickness of the virtual substrate 100 is less than the thickness of the electronic integrated circuit 200. This allows for the removal of the virtual substrate 100 and the exposure of the first conductive post 110 from the encapsulation layer during subsequent thinning of the encapsulation layer from its second surface (the surface opposite to the first conductive post 110). Specifically, since the thickness of the virtual substrate 100 is less than the thickness of the electronic integrated circuit 200, when the grinding amount reaches the thickness of the virtual substrate 100, the virtual substrate 100 is completely removed, and the second end of the first conductive post 110 (the end closer to the second surface) is exposed from the encapsulation layer. At this point, only a portion of the electronic integrated circuit 200 equivalent to the thickness of the virtual substrate 100 is ground away; that is, the electronic integrated circuit 200 is thinned, and its remaining thickness is the target thickness to which the electronic integrated circuit 200 is expected to be thinned. In other words, by making the thickness of the virtual substrate 100 less than the thickness of the electronic integrated circuit 200, the removal of the virtual substrate 100, the exposure of the second end of the first conductive post 110, and the thinning of the electronic integrated circuit 200 can be achieved simultaneously in the same thinning step. It is easy to understand that the virtual substrate 100 serves only as a temporary carrier in the process and is not retained in the final product structure, so it does not increase the thickness and volume of the final product.

[0030] In some embodiments, please refer to Figure 5 The method for preparing the virtual substrate 100 includes: Step S201: Provide a complete circular virtual base; A full-circuit virtual substrate refers to a virtual substrate that has not yet been cut and is in the form of a complete wafer. Its material is, for example, silicon, and its size is, for example, 6 inches, 8 inches, or 12 inches. Full-circuit virtual substrates are used for mass production of the first conductive pillar 110 to improve production efficiency. The material of the virtual substrate 100 is not limited to silicon; it can also be other rigid materials such as glass and ceramics.

[0031] Step S202: Form the first conductive pillar 110 on the surface of the circular virtual substrate.

[0032] Specifically, in some embodiments, photoresist can be coated onto the surface of a circular virtual substrate using photolithography and electroplating processes. An opening pattern corresponding to the first conductive pillar 110 is formed through exposure and development. A seed layer (such as Ti / Cu) is then formed by sputtering, and copper is filled into the openings through electroplating to form the first conductive pillar 110. Afterward, the photoresist is removed. In other embodiments, the first conductive pillar 110 can also be formed by other methods such as chemical plating or physical vapor deposition (PVD). The material of the first conductive pillar 110 is not limited to copper; it can also be other conductive metals such as gold, silver, and nickel, or their alloys.

[0033] Step S203: Thin the circular virtual substrate on which the first conductive pillar 110 is formed.

[0034] The above-mentioned thinning step is used to make the thickness of the thinned circular virtual substrate less than the thickness of the electronic integrated circuit 200.

[0035] Step S204: Cut the thinned circular virtual substrate to obtain virtual substrate 100 (e.g., Figure 3 (As shown).

[0036] After completing step S204 above, multiple independent virtual substrates 100 can be obtained, each virtual substrate 100 having one or more first conductive pillars 110 disposed thereon. In other words, a virtual substrate 100 refers to an independent unit required for packaging a single optical engine, obtained by cutting a complete circular virtual substrate. Each virtual substrate 100 corresponds to one optical engine.

[0037] In some embodiments, a second conductive pillar 210 can be fabricated on a wafer-level electronic integrated circuit first, and then the electronic integrated circuit 200 can be obtained by dicing. Specifically, the fabrication method of the electronic integrated circuit 200 includes: First, an electronic integrated circuit wafer is provided, on which multiple electronic integrated circuit units are formed, and each electronic integrated circuit unit has a second conductive post 210 disposed on its active surface (front side).

[0038] Then, the electronic integrated circuit wafer is cut to obtain multiple independent electronic integrated circuits 200.

[0039] In some embodiments, before dicing the electronic integrated circuit wafer, the wafer can be thinned to achieve a preset thickness for the electronic integrated circuit 200. This preset thickness is, for example, 300 μm to 800 μm. Maintaining this thickness when mounted on the substrate provides sufficient mechanical strength and surface flatness, facilitating subsequent mounting, packaging, and thinning processes, and avoiding chip cracking issues caused by excessive thinness. During subsequent thinning of the packaging layer, the electronic integrated circuit 200 is ground and thinned from its back side to a desired thickness, for example, less than 100 μm. In other words, before mounting, the electronic integrated circuit 200 is a chip that has not been thinned to an ultra-thin thickness. At this point, the electronic integrated circuit 200 is at the aforementioned preset thickness, which is greater than the desired thickness. This allows the electronic integrated circuit 200 to maintain a relatively thick state during mounting to ensure yield, achieving ultra-thinness in the final product to obtain excellent warp suppression and heat dissipation performance, balancing process feasibility and product performance.

[0040] In step S102, the virtual substrate 100 and the electronic integrated circuit 200 are mounted on the same side of the substrate 300.

[0041] Specifically, please refer to Figure 6In step (1), a substrate 300 is first provided. The substrate 300 is a temporary carrier, preferably a glass substrate, which has high flatness and thermal stability and can provide stable support in subsequent processes. A release layer 310 is coated on the surface of the substrate 300. The release layer 310 is preferably a photo-release layer or a laser-detachable bonding material, which is used to make the substrate 300 easy to peel off by light or heat in subsequent processes, while being able to withstand the high temperature and mechanical stress of the packaging and thinning processes.

[0042] Please see Figure 6 In step (2), an alignment mark 321 is prepared on the surface of the release layer 310. Specifically, a polyimide (PI) layer 320 is coated on the release layer 310, and the alignment mark 321 is formed in the PI layer 320 by photolithography. The alignment mark 321 is used to achieve precise positioning and alignment between the virtual substrate 100, the electronic integrated circuit 200 and other components in subsequent mounting, photolithography and bonding processes.

[0043] Please see Figure 6 In step (3), the virtual substrate 100 and the electronic integrated circuit 200 are mounted on the surface of the release layer 310. The virtual substrate 100 and the electronic integrated circuit 200 are arranged side by side, and the spacing between the virtual substrate 100 and the electronic integrated circuit 200 is set according to design requirements. Figure 3 As shown, a first intermediary layer 120 is also disposed on the surface of the virtual substrate 100 opposite to the first conductive post 110, such as... Figure 4 As shown, a second interposer 220 is also provided on the surface of the electronic integrated circuit 200 opposite to the second conductive post 210. The virtual substrate 100 is bonded to the release layer 310 through the first interposer 120; the electronic integrated circuit 200 is bonded to the release layer 310 through the second interposer 220.

[0044] In the above manner, the virtual substrate 100 and the electronic integrated circuit 200 are firmly bonded to the same side of the substrate 300, and both face away from the substrate 300. That is, the surface on the virtual substrate 100 where the first conductive pillar 110 is formed faces away from the substrate 300, and the surface on the electronic integrated circuit 200 where the second conductive pillar 210 is formed faces away from the substrate 300. Both the first conductive pillar 110 and the second conductive pillar 210 protrude in a direction away from the substrate 300.

[0045] Step S103: Perform packaging to form a packaging layer covering the electronic integrated circuit 200, the virtual substrate 100, the first conductive pillar 110, and the second conductive pillar 210.

[0046] like Figure 7As shown, after the virtual substrate 100 and the electronic integrated circuit 200 are mounted, an encapsulation step is performed. The virtual substrate 100, the electronic integrated circuit 200, the first conductive pillar 110, and the second conductive pillar 210 are encapsulated using an encapsulation material to form an encapsulation layer 400. This encapsulation material is, for example, epoxy molding compound (EMC). However, in practical applications, the material of the encapsulation layer 400 is not limited to epoxy molding compound; it can also be other thermosetting or thermoplastic encapsulation materials, such as curing adhesives, polyimide, etc. The encapsulation layer 400 covers the virtual substrate 100 and the electronic integrated circuit 200, as well as the entire surface of the first conductive pillar 110 and the second conductive pillar 210. The encapsulation layer 400 has a first surface 410 and a second surface 420 facing away from each other, wherein the first surface 410 faces away from the substrate 300 (i.e.,...). Figure 7 The middle encapsulation layer 400 faces upwards, and the second surface 420 faces the substrate 300 (i.e., Figure 7 (Surface of the middle encapsulation layer 400 facing downwards).

[0047] Step S104, as follows Figure 8 As shown, the encapsulation layer 400 is thinned from the first surface 410 so that the first end 111 of the first conductive post 110 and the second conductive post 210 are exposed from the encapsulation layer 400.

[0048] Specifically, the thinning step described above involves thinning the packaging layer 400 from the side away from the substrate 300. The thinning method can be mechanical polishing, chemical mechanical polishing (CMP), or a combination of both. By polishing away a portion of the material above the packaging layer 400, the first end 111 of the first conductive pillar 110 (i.e., the end away from the virtual substrate 100) and the end of the second conductive pillar 210 are exposed from the surface of the packaging layer 400. In other words, the endpoint of the thinning step is that the first end 111 of the first conductive pillar 110 and the end faces of the second conductive pillar 210 are precisely exposed on the surface of the packaging layer 400 away from the substrate 300, so that an electrical connection can be subsequently formed with the RDL (i.e., the second redistribution layer, described in detail later) on the photonic integrated circuit side.

[0049] In some embodiments, after completing step S104 above, as Figure 9 As shown, the preparation method of this application further includes: A first dielectric layer 430 with an opening is formed on the surface of the encapsulation layer 400 opposite to the substrate 300; An under bump metallization (UBM) layer 440 is fabricated, at least a portion of which penetrates the first dielectric layer 430 through the aforementioned opening and is electrically connected to the first conductive post 110 or the second conductive post 210.

[0050] The first dielectric layer 430 is an insulating material layer, preferably made of at least one of polyimide (PI), benzocyclobutene (BCB), silicon oxide, or silicon nitride. It provides insulation protection for the encapsulation layer 400, preventing unnecessary electrical coupling between the subsequently formed wiring layer and the conductive structures in the encapsulation layer 400. By forming openings in the first dielectric layer 430, areas requiring electrical connection (such as the first end 111 of the first conductive post 110 or the end face of the second conductive post 210) can be selectively exposed, thereby enabling electrical lead-out.

[0051] One end of the under-bump metal layer 440 is directly contacted and electrically connected to the first end 111 of the first conductive post 110 or the end face of the second conductive post 210 through an opening in the first dielectric layer 430. The other end is used to electrically connect to the RDL (i.e., the second redistribution layer) of the subsequently mounted photonic integrated circuit via solder balls. In some embodiments, the under-bump metal layer 440 is, for example, a multilayer metal stack structure, such as Ti / Ni / Au or Ti / Cu / Ni / Au, to provide a solderable surface, enabling the solder balls to spread evenly and solder firmly. At the same time, it can also prevent the tin (Sn) in the solder from interdiffusion with the copper in the first conductive post 110 or the second conductive post 210 to form brittle intermetallic compounds (IMC), ensuring the long-term reliability of the solder joint.

[0052] It should be noted that the conductive structures in the encapsulation layer 400 (such as the first end 111 of the first conductive pillar 110 or the end face of the second conductive pillar 210) can also be electrically connected to the RDL (i.e., the second redistribution layer) on the photonic integrated circuit side using other structures or methods, and are not limited to the aforementioned under-bump metal layer 440 and solder balls. Those skilled in the art can flexibly select the above or other suitable connection methods according to actual process conditions and performance requirements. These methods should all be considered equivalent substitutions or variations of the technical solution of this application and are all within the protection scope of this application.

[0053] In some embodiments, after completing step S104 above, as Figure 10 As shown, the preparation method of this application further includes: Provide a first temporary substrate 710; The first temporary substrate 710 is bonded to the surface of the encapsulation layer 400 opposite to the substrate 300.

[0054] Specifically, the first temporary substrate 710 is, for example, a rigid carrier plate, preferably made of glass, silicon, or ceramic. It provides mechanical support and protection for the encapsulation layer 400 and its internal virtual substrate 100, electronic integrated circuit 200, and first conductive pillar 110 in subsequent processes, preventing warping, deformation, or breakage during substrate 300 removal and subsequent thinning and rewiring processes. The first temporary substrate 710 can be bonded to the surface of the encapsulation layer 400 away from the substrate 300 via a third interposer 720. This third interposer 720 is a bonding dielectric layer, preferably made of at least one of temporary bonding adhesive, photosensitive polymer, or thermoplastic material. It enables a strong bond between the first temporary substrate 710 and the encapsulation layer 400 under certain temperature, pressure, or light conditions, and allows the first temporary substrate 710 to be released after subsequent processes by heating, laser irradiation, or chemical dissolution.

[0055] By bonding the first temporary substrate 710, support can be provided from the other side when the substrate 300 is subsequently removed, preventing the encapsulation layer 400 from warping or breaking due to loss of support after the substrate 300 is removed.

[0056] Step S105, as follows Figure 11 As shown, substrate 300 is removed.

[0057] Specifically, the release layer 310 can be made to lose its adhesiveness by means of heating, laser irradiation, or chemical dissolution, thereby peeling the substrate 300 from the PI layer 320. In some embodiments, the release layer 310 is a photorelease layer, which is decomposed or loses its adhesiveness by irradiating the back side of the substrate 300 (i.e., the surface opposite to the encapsulation layer 400) with a laser, thereby easily peeling off the substrate 300. In other embodiments, the release layer 310 is a thermosensitive material, which is softened or decomposed by heating to a preset temperature, thereby achieving the peeling off of the substrate 300.

[0058] Step S106, as follows Figure 12 As shown, from the second surface 420 (e.g.) Figure 7 (As shown) The encapsulation layer 400 is thinned to remove the virtual substrate 100, so that the second end 112 of the first conductive post 110 is exposed from the encapsulation layer 400, and the electronic integrated circuit 200 is thinned.

[0059] Specifically, since the thickness of the dummy substrate 100 is less than the thickness of the electronic integrated circuit 200, the dummy substrate 100 is ground away first when the encapsulation layer 400 is ground from the second surface 420. After the dummy substrate 100 is completely removed, the second end 112 of the first conductive post 110 (i.e., the end closest to the dummy substrate 100) is exposed from the encapsulation layer 400. During this process, the back surface of the electronic integrated circuit 200 is gradually thinned. The endpoint of this thinning step is that the electronic integrated circuit 200 is thinned to the desired thickness, and the back surface of the electronic integrated circuit 200 is flush with the end face of the second end 112 of the first conductive post 110. At this point, the thinned back surface of the encapsulation layer 400 becomes a flat, coplanar surface encompassing the back surface of the electronic integrated circuit 200 and the end face of the first conductive post 110.

[0060] As can be seen from the above, through the above-mentioned "first mounting, then thinning" process design, the electronic integrated circuit 200 is kept in a relatively thick state during mounting (for example, a preset thickness in the range of 300μm to 800μm). After mounting, it is thinned to the desired thickness (for example, less than 100μm) through the above-mentioned thinning step (i.e., step S106).

[0061] Figure 13 This is a graph showing the relationship between the thickness of the bottom chip and the amount of warpage. Figure 14 for Figure 13 A magnified view of a local area. (e.g.) Figure 13 and Figure 14 As shown, the horizontal axis represents the thickness t (unit: μm) of the bottom chip, and the vertical axis represents the warpage (unit: μm) of the module (the overall structure of the optical engine) at room temperature. The bottom chip refers to the overall structure formed by the encapsulation layer 400, the electronic integrated circuit 200, and the first conductive pillar 110 after completing step S106. The overall thickness of this bottom chip is t, which is the distance between the surfaces of the encapsulation layer 400 near and away from the first temporary substrate 710. Warpage refers to the bending deformation of the module at room temperature due to the mismatch of the coefficients of thermal expansion (CTE) of the various layers. A larger warpage indicates poorer module dimensional stability and lower optical coupling accuracy and encapsulation yield.

[0062] Figure 13 The multiple curves in the figure correspond to different silicon percentages A (i.e., the percentage of silicon material area to the total area of ​​the bottom chip) in the bottom chip. The values ​​of A are, for example, 20%, 30%, 40%, 50%, 60%, 70%, and 80%. Each curve reflects the change of warpage with the thickness t of the bottom chip under the corresponding silicon percentage condition. Figure 14 for Figure 13 A magnified view of the chip thickness t in the middle and bottom range of 0 to 100 μm is provided to more clearly show the warping trend within the ultra-thin thickness range.

[0063] from Figure 13 and Figure 14 It can be seen that regardless of the silicon percentage A, the warpage initially increases and then decreases with increasing bottom chip thickness t. Specifically, as the bottom chip thickness t increases from 0, the warpage gradually rises, reaching a peak in the range of approximately 300μm to 400μm; after t continues to increase beyond the peak point, the warpage gradually decreases. Therefore, the module warpage is most severe when the bottom chip thickness t is in the range of 300μm to 400μm.

[0064] More importantly, from Figure 14 It is clearly evident that when the thickness t of the bottom chip is less than or equal to 100 μm, the warpage decreases approximately linearly with decreasing thickness t. In other words, within the ultra-thin range of t ≤ 100 μm, reducing the thickness of the bottom chip can linearly suppress warpage. This indicates that this application, by thinning the electronic integrated circuit 200 to below 100 μm, brings the overall thickness of the bottom chip into this "low warpage range," thereby fundamentally solving the warpage problem in large-size chip 3D stacking.

[0065] In addition, from Figure 14 It can also be seen that when the bottom chip thickness t≤100μm, the warpage difference corresponding to different silicon ratios A is small, indicating that the warpage sensitivity to silicon ratio is reduced in this ultra-thin range, the process window is wider, and it is easier to achieve stable packaging yield.

[0066] Therefore, by thinning the electronic integrated circuit 200 to the desired thickness, this application can significantly reduce module warpage and improve optical coupling efficiency and packaging yield. Simultaneously, the significantly reduced thickness of the electronic integrated circuit 200 shortens the heat conduction path from its interior to its back surface, providing favorable conditions for the subsequent design of heat dissipation structures.

[0067] In some embodiments, after step S106 described above, the preparation method of this application further includes: Step S107, as follows Figure 15 As shown, a first redistribution layer 500 is formed on the side where the second end 112 of the first conductive post 110 of the encapsulation layer 400 is located; the first redistribution layer 500 is electrically connected to the second end 112 of the first conductive post 110 and the electronic integrated circuit 200. In step S108, a heat dissipation structure (e.g., including multiple thermally conductive protrusions 510) is formed on the surface of the first wiring layer 500 away from the encapsulation layer 400, and the heat dissipation structure is electrically connected to the first wiring layer 500.

[0068] Specifically, the method for forming the first interconnect layer 500 includes, for example, coating a second dielectric layer 520 (e.g., polyimide PI) on the surface where the second end 112 of the first conductive post 110 of the packaging layer 400 is located, and forming an opening in the second dielectric layer 520 by photolithography to expose the second end 112 of the first conductive post 110 and the corresponding area on the back side of the electronic integrated circuit 200. Then, a metal interconnect layer 530 is formed by processes such as sputtering a seed layer (e.g., Ti / Cu) and electroplating copper, which is electrically connected to the second end 112 of the first conductive post 110 and the electronic integrated circuit 200. In some embodiments, the metal interconnect layer 530 can be one or more layers, and a second dielectric layer 520 is disposed between adjacent metal interconnect layers 530 for insulation isolation.

[0069] Compared to related technologies that fabricate high-density RDL on the packaging layer, this application integrates the second wiring layer (described in detail below) onto the surface of the photonic integrated circuit, undertaking the high-density interconnection of high-speed signals between the electronic integrated circuit and the photonic integrated circuit. The first wiring layer 500 can, for example, realize the signal fan-out of the first conductive pillar 110 and the heat conduction of the heat dissipation structure (e.g., including multiple thermally conductive bumps 510), without the need for high-density wiring to carry high-speed signals. Based on the above functional division, compared with related technologies, the first wiring layer 500 of this application can use fewer layers and a smaller thickness. The reduction in the number of layers and thickness shortens the heat conduction path from the back of the electronic integrated circuit 200 through the first wiring layer 500 and the thermally conductive bumps 510, reducing thermal resistance and improving heat dissipation efficiency; on the other hand, it reduces the warping stress caused by the second dielectric layer 520, which is more conducive to suppressing module warping. Meanwhile, since the second wiring layer is fabricated on the photonic integrated circuit wafer, it is not affected by the warping of the packaging layer, and can achieve smaller line widths and spacings and more layers. Thus, while ensuring the integrity of high-speed signals, the first wiring layer 500 can be simplified and thinned, taking into account the comprehensive needs of signal interconnection, heat dissipation and warping suppression.

[0070] Based on this, the heat dissipation structure includes multiple thermally conductive bumps 510, such as C4 Bumps (Controlled Collapse Chip Connection Bumps). A C4 Bump is a solder bump structure formed using flip-chip bonding technology, where the solder is controllably collapsed into a spherical shape under surface tension during reflow soldering. In this application, the C4 Bump serves two purposes: firstly, as an external solder ball, it bonds the optical engine to an external substrate (such as a PCB) to achieve electrical signal input / output; secondly, as a thermally conductive bump 510, it conducts the heat generated by the electronic integrated circuit 200 to the external substrate via the first multi-level wiring layer 500, achieving efficient downward heat dissipation. Before forming the thermally conductive bumps 510, an underbump metallization (UBM) layer can be formed on the surface of the first multi-level wiring layer 500 facing away from the package layer 400 to ensure the reliability of the connection between the thermally conductive bumps 510 and the first multi-level wiring layer 500.

[0071] Through steps S107 and S108, the heat generated by the electronic integrated circuit 200 is sequentially conducted to the external substrate via the first wiring layer 500 and the thermally conductive bump 510, achieving an efficient back-side heat dissipation path. The first wiring layer 500 serves not only as a signal wiring layer but also as a heat diffusion layer, dissipating heat from the back of the electronic integrated circuit 200 through the thermally conductive bump 510, further improving heat dissipation efficiency.

[0072] In some embodiments, after step S108 described above, the preparation method of this application further includes: Step S109: Provide a second temporary substrate 730; Step S110, as follows Figure 16 As shown, the second temporary substrate 730 is bonded to the surface of the first redistribution layer 500 opposite to the encapsulation layer 400. Step S111, as follows Figure 17 As shown, the first temporary substrate 710 is removed.

[0073] Step S112: Cut to obtain chip units.

[0074] Specifically, the second temporary substrate 730 is, for example, a rigid carrier plate, preferably made of glass, silicon, or ceramic. It provides mechanical support and protection for the packaging layer 400 and the internal electronic integrated circuit 200 and the first conductive pillar 110 in subsequent processes, preventing warping, deformation, or breakage. In step S110, the second temporary substrate 730 can be bonded to the surface of the first redistribution layer 500 away from the packaging layer 400 via a fourth interposer 740. This fourth interposer 740 is a bonding dielectric layer, preferably made of at least one of temporary bonding adhesive, photosensitive polymer, or thermoplastic material. It enables a strong bond between the second temporary substrate 730 and the first redistribution layer 500 under certain temperature, pressure, or light conditions, and allows the second temporary substrate 730 to be released after subsequent processes by heating, laser irradiation, or chemical dissolution.

[0075] In step S112 above, the entire wafer, after completing processes such as packaging, thinning, redistribution layer fabrication, and heat dissipation structure formation, is cut along the dicing lines to form multiple independent chip units. It is easy to understand that... Figures 6 to 12 as well as Figures 15 to 17 The diagram shows a partial structure of the entire wafer before dicing (corresponding to the area where a single chip unit is located). In reality, before dicing, multiple chip unit structures are fabricated simultaneously on the entire wafer, and these chip unit structures are isolated from each other by dicing channels. After dicing, each independent chip unit includes a packaging layer 400, a first conductive pillar 110 embedded in the packaging layer 400, an electronic integrated circuit 200, a first redistribution layer 500, and a heat dissipation structure disposed on the first redistribution layer 500. At this time, the photonic integrated circuit has not yet been mounted on the chip unit.

[0076] In some embodiments, such as Figure 18 As shown, the preparation method of this application further includes: A photonic integrated circuit 600 is provided, wherein a second wiring layer 610 is integrated on the surface of the photonic integrated circuit 600; The photonic integrated circuit 600 is mounted on the side where the first end 111 of the first conductive post 110 of the encapsulation layer 400 is located. The second wiring layer 610 is located between the photonic integrated circuit 600 and the packaging layer 400, and is electrically connected to the first end 111 of the first conductive post 110 and the second conductive post 210.

[0077] Specifically, the photonic integrated circuit 600 (PIC) is the chip responsible for optical signal processing in the optical engine. It includes a silicon photonic integrated circuit and integrates optical devices such as an optical modulator, photodetector, optical waveguide, grating coupler, or end-face coupler to realize the mutual conversion between electrical signals and optical signals. In this application, the photonic integrated circuit 600 works in conjunction with the electronic integrated circuit 200. That is, in the transmitting direction, the driver of the electronic integrated circuit 200 amplifies the electrical signal and drives the optical modulator in the photonic integrated circuit 600 to convert the electrical signal into an optical signal for output; in the receiving direction, the photodetector in the photonic integrated circuit 600 converts the received optical signal into an electrical signal, which is then amplified and processed by the transimpedance amplifier (TIA) in the electronic integrated circuit 200.

[0078] A second wiring layer 610 is integrated on the surface of the photonic integrated circuit 600. The second wiring layer 610 is a multilayer metal wiring layer pre-fabricated on the photonic integrated circuit 600, which includes alternating metal line layers and dielectric layers (such as PI). The second wiring layer 610 is used to redistribute the I / O pads of the photonic integrated circuit 600 and electrically connect them to the first end 111 of the first conductive post 110 and the end face of the second conductive post 210, thereby realizing the interconnection between the photonic integrated circuit 600 and the chip cell.

[0079] Since the second wiring layer 610 is fabricated on the photonic integrated circuit 600 using semiconductor technology, it is not affected by packaging layer warpage. Therefore, it can achieve smaller linewidths and spacings (e.g., below 2μm / 2μm) and more wiring layers (e.g., more than 3 layers), thereby providing higher wiring density and better signal integrity. At the same time, compared with the serial process in related technologies that requires sequential fabrication of RDL on the plastic-encapsulated wafer and then mounting of the photonic integrated circuit, the fabrication of the second wiring layer 610 can be at least partially paralleled with the chip unit side (i.e., the mounting, plastic encapsulation, thinning, and other processes of the virtual substrate 100 and the electronic integrated circuit 200). That is, while the chip unit side process is being carried out, the RDL fabrication on the photonic integrated circuit 600 side is also being carried out simultaneously. Finally, the two are bonded together, thereby shortening the overall packaging cycle of the optical engine.

[0080] In some embodiments, such as Figure 18 As shown, one end of the under-bump metal layer 440 is electrically connected to the first end 111 of the first conductive post 110 or the end face of the second conductive post 210, and the other end is electrically connected to the second redistribution layer 610 via solder balls 450. The solder balls 450 are disposed between the under-bump metal layer 440 and the second redistribution layer 610, and the electrical and mechanical connection between the under-bump metal layer 440 and the second redistribution layer 610 is achieved by a reflow soldering process.

[0081] Furthermore, the gap between the second redistribution layer 610 and the first dielectric layer 430 can be filled with underfill 620. Underfill 620 is a liquid epoxy resin adhesive that fills the gap between the second redistribution layer 610 and the first dielectric layer 430 through capillary action, and then cures upon heating. Underfill 620 at least covers the bonding interface between the solder balls 450 and the under-bump metal layer 440 and the second redistribution layer 610, thus alleviating thermal stress, enhancing mechanical strength, and protecting the solder balls and bonding interface from moisture and contaminant corrosion, thereby improving the long-term reliability of the optical engine.

[0082] After completing the mounting between the photonic integrated circuit 600 and the chip unit, as follows Figure 19 As shown, the preparation method of this application further includes: Remove the second temporary substrate 730.

[0083] Specifically, the method for removing the second temporary substrate 730 depends on the bonding method between the second temporary substrate 730 and the chip cell. Specifically, the second temporary substrate 730 is bonded to the first redistribution layer 500 via a fourth interposer 740 (e.g., temporary bonding adhesive or photosensitive adhesive). During removal, the bonding layer can be softened or decomposed by heating, lose its adhesiveness by laser irradiation, or dissolved by chemical dissolution, thereby peeling the second temporary substrate 730 off the chip cell. After peeling, residual material from the fourth interposer 740 is cleaned to expose the surfaces of the first redistribution layer 500 and the heat dissipation structure. After removing the second temporary substrate 730, the final optical engine product is obtained.

[0084] As another technical solution, embodiments of this application provide an optical engine, the structure of which is as follows: Figure 19 As shown, it is prepared based on the above-described preparation method of the embodiments of this application.

[0085] As another technical solution, this application embodiment also provides an optical engine, such as... Figure 19 As shown, it includes a chip unit and a photonic integrated circuit 600. The chip unit includes an encapsulation layer 400, a first conductive post 110 disposed within the encapsulation layer 400, and an electronic integrated circuit 200. The two ends of the first conductive post 110 are exposed from a first side and a second side of the encapsulation layer 400, respectively. A second conductive post 210 is disposed on the electronic integrated circuit 200, and the second conductive post 210 is exposed from the first side of the encapsulation layer 400. The thickness of the chip unit is less than or equal to 100 μm, meaning the overall thickness of the encapsulation layer 400, the first conductive post 110, and the electronic integrated circuit 200 is less than or equal to 100 μm.

[0086] In this application, the first side of the encapsulation layer 400 refers to the side where the photonic integrated circuit 600 is located (i.e., the upper side of the encapsulation layer 400), and the second side refers to the side opposite to the first side (i.e., the lower side of the encapsulation layer 400, the side facing the external substrate). The first end 111 of the first conductive post 110 protrudes from the first side of the encapsulation layer 400 for electrical connection with the second redistribution layer 610 on the side of the photonic integrated circuit 600; the second end 112 of the first conductive post 110 protrudes from the second side of the encapsulation layer 400 for electrical connection with the first redistribution layer 500. The second conductive post 210 on the electronic integrated circuit 200 protrudes from the first side of the encapsulation layer 400 for electrical connection with the second redistribution layer 610, thereby enabling signal transmission between the electronic integrated circuit 200 and the photonic integrated circuit 600.

[0087] The thickness of the chip unit (i.e., the distance from the first side surface to the second side surface of the packaging layer 400, including the overall structure of the packaging layer 400, the first conductive pillar 110, and the electronic integrated circuit 200) is less than or equal to 100 μm. The principle and effect of this setting have been described in detail in the above embodiments and will not be repeated here.

[0088] In some embodiments, the chip unit further includes a first redistribution layer 500. The first redistribution layer 500 is disposed on the second side of the packaging layer 400 and electrically connected to the second end 112 of the first conductive post 110 and the electronic integrated circuit 200. A heat dissipation structure is formed on the surface of the first redistribution layer 500 facing away from the packaging layer 400, and this heat dissipation structure is electrically connected to the first redistribution layer 500. The heat dissipation structure is used to conduct heat generated by the electronic integrated circuit 200 to the external environment or an external substrate. Further, in some embodiments, the heat dissipation structure includes a plurality of thermally conductive bumps 510 (e.g., C4 bumps), which are formed on the surface of the first redistribution layer 500 facing away from the packaging layer 400 and electrically connected to the first redistribution layer 500. The thermally conductive bump 510 is preferably made of solder alloy (such as SAC305 tin-silver-copper alloy). It is formed on the outer surface pad of the first rewiring layer 500 by balling and reflow soldering processes. On the one hand, it serves as an external solder ball for soldering to the external substrate, and on the other hand, it serves as a heat conduction path to conduct heat to the external substrate.

[0089] In some embodiments, the chip unit further includes a first dielectric layer 430, a under-bump metal layer 440, and an underfill 620. The first dielectric layer 430 is disposed on the surface of a first side of the package layer 400 to provide insulating protection for the first side of the package layer 400. The under-bump metal layer 440 is at least partially disposed through the first dielectric layer 430, and one end of the under-bump metal layer 440 is directly in contact with and electrically connected to the first end 111 of the first conductive post 110 or the end face of the second conductive post 210 through an opening in the first dielectric layer 430, and the other end is used to be electrically connected to the RDL (i.e., the second redistribution layer) of the subsequently mounted photonic integrated circuit side through solder balls. In some embodiments, the under-bump metal layer 440 is, for example, a multi-layer metal stack structure, such as Ti / Ni / Au or Ti / Cu / Ni / Au, used to provide a solderable surface, enabling the solder balls to spread evenly and be firmly soldered. At the same time, it can also prevent the tin (Sn) in the solder from interdiffusion with the copper in the first conductive pillar 110 or the second conductive pillar 210 to form brittle intermetallic compounds (IMC), ensuring the long-term reliability of the solder joint.

[0090] In some embodiments, in the optical engine described above, the critical dimension of the first conductive post 110 is less than or equal to 10 μm, and the center-to-center distance between two adjacent first conductive posts 110 is less than or equal to 30 μm. This allows the optical engine to realize more vertical interconnect channels per unit area, significantly improving I / O density while reducing module size. Furthermore, due to the small height and short path of the first conductive post 110, high-speed signal transmission loss is correspondingly reduced.

[0091] The optical engine provided in this embodiment can be fabricated using any of the methods described in the above embodiments of this application. It features excellent heat dissipation, low warpage, small size, high I / O density, and low signal loss. It can be directly used for coupling with external fiber optic connectors and system substrate mounting, making it suitable for high-speed optical communication scenarios such as CPO (co-packaged optoelectronics) or optical I / O. Of course, in practical applications, this optical engine can also be fabricated using other methods.

[0092] In the description of this specification, the references to terms such as "this embodiment," "an embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, those skilled in the art can combine and integrate the different embodiments or examples described in this specification and the features of different embodiments or examples without contradiction. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0093] The above description is merely a preferred embodiment and the technical principles employed in this application. Those skilled in the art will understand that this application is not limited to the specific embodiments described above, and various obvious changes, readjustments, combinations, and substitutions can be made without departing from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of this application. The scope of protection of this application is determined by the scope of the appended claims.

Claims

1. A method of manufacturing an optical engine, characterized by, include: A virtual substrate and an electronic integrated circuit are provided. A first conductive pillar is disposed on the virtual substrate, and a second conductive pillar is disposed on the electronic integrated circuit. The thickness of the virtual substrate is less than the thickness of the electronic integrated circuit. The virtual substrate and the electronic integrated circuit are mounted on the same side of the substrate; Encapsulation is performed to form an encapsulation layer covering the electronic integrated circuit, the virtual substrate, the first conductive pillar, and the second conductive pillar; The encapsulation layer has a first surface and a second surface that are opposite to each other, and the first surface is opposite to the substrate; The encapsulation layer is thinned from the first surface to expose the first end of the first conductive post and the second conductive post from the encapsulation layer; Remove the substrate; The encapsulation layer is thinned from the second surface to remove the dummy substrate, exposing the second end of the first conductive pillar from the encapsulation layer, and the electronic integrated circuit is thinned.

2. The production method according to claim 1, characterized by, After the encapsulation layer is thinned from the second surface, the encapsulation layer, the electronic integrated circuit, and the first conductive pillar together constitute the bottom chip, and the overall thickness of the bottom chip is less than or equal to 100 μm.

3. The preparation method according to claim 1, characterized in that, After thinning the encapsulation layer from the second surface, the method further includes: A first redistribution layer is formed on the side where the second end of the first conductive pillar is located in the encapsulation layer; the first redistribution layer is electrically connected to the second end of the first conductive pillar and the electronic integrated circuit; A heat dissipation structure is formed on the surface of the first rewiring layer away from the encapsulation layer, and the heat dissipation structure is electrically connected to the first rewiring layer.

4. The preparation method according to claim 3, characterized in that, The heat dissipation structure includes at least two thermally conductive protrusions; the thermally conductive protrusions are formed on the surface of the first rewiring layer away from the encapsulation layer and are electrically connected to the first rewiring layer.

5. The preparation method according to claim 1, characterized in that, Also includes: A photonic integrated circuit is provided, wherein a second wiring layer is integrated on the surface of the photonic integrated circuit; The photonic integrated circuit is mounted on the side where the first end of the first conductive pillar of the encapsulation layer is located. The second redistribution layer is located between the photonic integrated circuit and the packaging layer, and is electrically connected to the first end of the first conductive pillar and the second conductive pillar.

6. The preparation method according to claim 1, characterized in that, The method for preparing the virtual substrate includes: Provides a full-circle virtual base; The first conductive pillar is formed on the surface of the circular virtual substrate; The entire circular virtual base is thinned; The thinned circular virtual base is cut to obtain the virtual base; The thickness of the thinned circular virtual substrate is less than the thickness of the electronic integrated circuit.

7. The preparation method according to claim 3, characterized in that, After thinning the encapsulation layer from the first surface and before removing the substrate, the process further includes: Provide a first temporary substrate; The first temporary substrate is bonded to the surface of the encapsulation layer opposite to the substrate; After forming a heat dissipation structure on the surface of the first redistribution layer away from the encapsulation layer, the method further includes: Provide a second temporary substrate; The second temporary substrate is bonded to the surface of the first redistribution layer opposite to the encapsulation layer; Remove the first temporary substrate; Chip units are obtained by cutting.

8. An optical engine, characterized in that, include: A chip unit includes an encapsulation layer, a first conductive post disposed in the encapsulation layer, and an electronic integrated circuit; the two ends of the first conductive post are exposed from a first side and a second side of the encapsulation layer, respectively; the electronic integrated circuit has a second conductive post disposed thereon and is exposed from the first side of the encapsulation layer. A photonic integrated circuit is mounted on the first side of the packaging layer and has a second wiring layer integrated on its surface. The second wiring layer is located between the photonic integrated circuit and the packaging layer and is electrically connected to the first conductive post and the second conductive post. The thickness of the chip unit is less than or equal to 100 μm.

9. The optical engine according to claim 8, characterized in that, The chip unit also includes: The first wiring layer is disposed on the second side of the encapsulation layer and is electrically connected to the first conductive pillar and the electronic integrated circuit. A heat dissipation structure is formed on the surface of the first rewiring layer away from the encapsulation layer, and the heat dissipation structure is electrically connected to the first rewiring layer.

10. The optical engine according to claim 9, characterized in that, The heat dissipation structure includes at least two thermally conductive protrusions; the thermally conductive protrusions are formed on the surface of the first rewiring layer away from the encapsulation layer and are electrically connected to the first rewiring layer.

11. The optical engine according to claim 8, characterized in that, The chip unit also includes: A dielectric layer is disposed on the surface of the encapsulation layer on the first side; The under-bump metal layer is at least partially disposed through the dielectric layer, and one end of the under-bump metal layer is electrically connected to the first conductive post or the second conductive post, and the other end is electrically connected to the second redistribution layer through solder balls. The underfill adhesive fills the gap between the second redistribution layer and the dielectric layer.

12. An optical engine, characterized in that, It is prepared according to the preparation method described in any one of claims 1 to 7.