A heterogeneous integrated photonic platform supporting waveguides with large effective mode cross-sections and thermally optimized active components.

Abstract

The optical device has five elements fabricated on a single substrate. The first element has an active waveguide structure that supports the first mode. The second element has a passive waveguide structure that supports the second mode. The third element has a passive waveguide structure that supports the third mode. The tapered waveguide structure of the second or fifth element facilitates efficient adiabatic conversion between the second mode and the intermediate mode. The tapered waveguide structure in at least one of the second and third elements facilitates efficient adiabatic conversion between the second mode and the third mode. No adiabatic conversion occurs between the intermediate mode and the first mode.

Description

Technical Field

[0001] The present invention relates to an optical integrated circuit. Specifically, certain embodiments of the present invention relate to improving the performance of a heterogeneous integrated active component that uses optically coupled heterogeneous materials and includes a waveguide with a large effective mode cross-sectional area.

Background Art

[0002] A photonic integrated circuit (PIC), or an integrated optical circuit, is a device that integrates multiple optical functions and is similar to an electronic integrated circuit. The main difference between the two is that an optical integrated circuit provides functions corresponding to information signals carried on an optical carrier wave.

[0003] The most commercially used material platform for photonic integrated circuits is indium phosphide (InP), which allows all optically active and passive functions to be integrated on the same chip. In recent years, many PICs have been realized on the InP platform, but important research has been conducted in the past decade to use silicon (Si) or silicon nitride (SiN) instead of InP for the realization of PICs. This is because the former materials have several excellent characteristics and excellent processing capabilities, and the existing investment in electronic integrated circuits can be utilized.

[0004] The biggest challenge in using Si in PICs is that, as an indirect bandgap material, it is difficult to provide an electrically excited light source. This challenge is typically solved by assembling two or more chips made of different materials in separate processes. Such an approach is difficult because it requires extremely fine alignment, increases packaging costs, and limits scaling. Another way to solve the bandgap problem is to bond two different materials and process them together. This eliminates the need for precise alignment when bonding large or complete wafers of different materials, enabling mass production. In this disclosure, the term "hybrid" is used when describing the first approach, which involves the precise assembly of separately processed components, and the term "heterogeneous" is used when describing the latter approach, which involves bonding two materials and processing the bonded material to define waveguides and other target elements.

[0005] Conventional heterogeneous approaches to transfer optical signals between dissimilar materials have employed tapered structures whose dimensions gradually decrease until the effective mode refractive indices of the dissimilar materials match, enabling efficient power transfer. Such approaches are generally useful when the refractive index difference between materials is small, such as between Si and InP. However, when the effective refractive index difference is large, for example between SiN and gallium arsenide (GaAs) or InP, the dimensional requirements of the tapered tip become stringent, limiting efficient power transfer. Specifically, achieving good coupling may require extremely small tapered tip widths (to a few tens of nanometers). Achieving such dimensions is complex and potentially excessively costly.

[0006] While InP-based and Si-based PICs meet all current needs, they have several limitations. For example, the operating wavelength band is limited by material absorption, leading to increased losses; there are limitations on the maximum optical intensity and power that the PIC can handle; propagation losses are generally high (compared to Si waveguides and SiN waveguides); and coupling losses with optical fibers are generally high due to small mode sizes (high refractive index contrast between core and cladding).

[0007] To address the aforementioned limitations, alternative waveguide materials such as SiN, silicon oxynitride (SiONx), lithium niobate (LiNbO3), titanium dioxide (TiO2), tantalum pentoxide (Ta2O5), and aluminum nitride (AlN) have been investigated. Generally, such dielectric waveguides have high bandgap energies, enabling better high-power processing and transparency at shorter wavelengths, but the aforementioned materials also have low refractive indices. For example, the refractive index of SiN with a bandgap of approximately 5 eV is approximately 2, the refractive index of AlN with a bandgap of approximately 6 eV is approximately 2, and the refractive index of SiO2 with a bandgap of approximately 8.9 eV is approximately 1.44. Waveguides made of such materials support larger optical mode diameters due to the low refractive index contrast, and typically have low propagation loss and excellent coupling efficiency with optical fibers. On the other hand, the large refractive index difference between the waveguide material and InP / GaAs makes tapered approaches for coupling light on the chip between the waveguide and InP / GaAs difficult. In comparison, the refractive indices of both InP and GaAs are 3 or greater. Alternative hybrid approaches face the challenges already mentioned, namely the need for precise alignment and the resulting complex packaging and scaling constraints.

[0008] A recent approach to the aforementioned challenges, as presented in Patent Document 1, employs butt coupling in combination with a mode converter, enabling the use of heterogeneous processes without requiring extremely small taper widths. While butt coupling significantly relaxes the taper width requirement, it can present challenges when combined with waveguides that support large effective mode cross-sections. Supporting large optical mode sizes on a PIC typically requires thicker dielectric cladding, which can lead to reduced heat dissipation efficiency and negatively impact the performance of high-power active devices. [Prior art documents] [Patent Documents]

[0009] [Patent Document 1] U.S. Patent No. 10859764 [Overview of the project] [Means for solving the problem]

[0010] The present invention relates to heterogeneous components and PICs in which the thermal performance of active components is optimized while employing butt coupling and efficiently optically coupled to waveguides with large effective mode cross-sections. In particular, the embodiments described below relate to the detailed design of a photonic platform necessary for fabricating high-performance heterogeneous integrated lasers and active components coupled to waveguides with large effective mode cross-sections. [Brief explanation of the drawing]

[0011] [Figure 1] This is a cross-sectional view showing a device according to one embodiment of the present invention, cut in the vertical direction. [Figure 2] This is a cross-sectional view of a device according to several embodiments of the present invention. [Figure 3] This is a cross-sectional view of a device according to several embodiments of the present invention. [Figure 4]This is a cross-sectional view of the end face of a device according to several embodiments of the present invention. [Figure 5] This graph shows the simulation results of devices according to several embodiments of the present invention. [Figure 6] This graph shows the simulation results of devices according to several embodiments of the present invention. [Figure 7] This graph shows the simulation results of devices according to several embodiments of the present invention. [Figure 8] This graph shows the simulation results of devices according to several embodiments of the present invention. [Figure 9] This is a cross-sectional view of a device according to one embodiment of the present invention, as seen in the vertical direction. [Modes for carrying out the invention]

[0012] This specification describes embodiments of a platform for realizing photonic integrated circuits (PICs) using wafer bonding and heterogeneous material deposition, with improved optical coupling through the use of mode conversion and butt coupling schemes. More specifically, certain embodiments of the present invention relate to improved thermal performance of heterogeneous integrated lasers and active components when integrated with waveguides that support large effective mode cross-sections.

[0013] The following detailed description refers to the accompanying drawings, which constitute part of this specification. In the drawings, the same numbers indicate the same parts or components throughout, and the drawings illustrate examples of embodiments that can carry out the subject matter of this disclosure. It should be understood that other embodiments may be used, and structural or logical modifications may be made without departing from the scope of this disclosure. Therefore, the following detailed description should not be constrained to mean limitingly, and the scope of the embodiments is defined by the accompanying claims and their equivalents.

[0014] Descriptions may include references based on perspectives such as up / down, inside / outside, and above / below. Such descriptions are used solely to facilitate discussion and are not intended to limit the application of the embodiments described herein to any particular purpose. The descriptions may include phrases such as "in one embodiment" or "in one embodiment," which may refer to one or more of the same or different embodiments. Furthermore, terms such as "equip," "include," and "have" used in reference to embodiments of this disclosure are synonymous.

[0015] In this disclosure, the term "A and / or B" means (A), (B), or (A and B). In this disclosure, the term "A, B, and / or C" means (A), (B), (C), (A and B), (A and C), (B and C), or (A, B and C).

[0016] In this specification, the term “coupled” and its derivatives may be used. “Coupled” may mean that two or more elements or components are in direct contact in one or more physical, electrical, and optical manner. However, “coupled” may also mean that two or more elements or components cooperate or interact with each other while indirectly in contact, or that one or more other elements are coupled or connected between elements considered to be coupled. The term “directly coupled” means that two or more elements are in direct contact on at least some surfaces. The term “butt coupled” here means “end-face coupled” or axial coupled in the usual sense, meaning that the axial offset between the elements in question is minimal or zero. For example, if some thin intervening layer is formed between the elements, such as a thin coating layer typically used to provide high reflectivity or anti-reflective properties, the axial offset may be slightly greater than zero. It should be noted that, in order to accurately describe a butt coupled state, the axes of the two waveguide structures or elements do not need to be collinear. In other words, for example, in the case of an angled interface, the interface between elements does not need to be perpendicular to the axis of either element. For example, angled interfaces can be used to control back reflection. No adiabatic conversion occurs between butted structures.

[0017] In this specification, terms such as "active device", "active structure", or other "active" elements, parts, or components may be used. A device or part of a device called active can perform light generation, amplification, modulation, and / or detection using electrical contacts. This is in contrast to "passive devices" that have the main function of confining and guiding light and / or providing other functions commonly associated with passive devices, such as splitting, combining, filtering, and / or other functions. Some passive devices can provide functions that overlap with those of active devices, such as phase adjustment realized using thermal effects, etc. Based solely on the material composition or device structure, "active" and "passive" should not be absolutely distinguished. For example, a silicon device is considered active for specific modulation conditions or detection of low-wavelength radiation, but passive in most other situations.

[0018] FIG. 1 shows a top view of a photonic integrated circuit (PIC) device (100) according to an embodiment of the present invention, in which there are three different active elements 105, 110, 115. In other embodiments, the number of active elements may be less than or more than three. In any case, the PIC is composed of three different regions as shown in the figure, namely, an active region, a transition region, and a large effective mode cross-sectional area.

[0019] In the illustrated example, the active element 105 is a laser or a photodetector, and the active elements 110, 115 are semiconductor optical amplifiers or modulators, but many variations and combinations are possible. In the illustrated embodiment, as described with reference to FIGS. 2 to 4, all active elements are disposed on the heat sink 130 regardless of the presence or absence of the spacer layer. In other cases not shown, the heat sink is thermally coupled to at least one active element rather than all active elements on the PIC. In some embodiments, the thermal conductivity of the heat sink exceeds 10 W / mK. In some embodiments, the materials used to fabricate the heat sink include, but are not limited to, one or more of silicon, silicon nitride, copper, gold, aluminum, tungsten, silver, zinc, diamond, graphite, silicon carbide, and aluminum nitride. The main purpose of the heat sink 130 is to efficiently conduct the heat generated by the active element away from the element because it has a higher thermal conductivity than the thermal conductivity of the common dielectrics used to form the optical cladding. For example, the thermal conductivity of silicon dioxide (SiO2) is only about 1.3 W / mK.

[0020] The active element is efficiently and optically coupled to the large mode region 151 using the transition region 176, as described in more detail with reference to FIGS. 2 and 3.

[0021] In some embodiments, a laser (such as the active element 105 in Figure 1) is injection-locked to a resonator (such as element 160 in Figure 1) with a high quality factor that can be achieved in a large mode region. By using a waveguide with a large mode cross-section (effective mode cross-section), a quality factor exceeding 100 million can be achieved, and such a resonator can reduce the laser linewidth by more than an order of magnitude. Furthermore, if such a resonator is implemented in a so-called add / drop configuration (as also shown in Figure 1), the impact of reflections on laser performance can be significantly reduced. The output of such an injection-locked laser can be further amplified by a semiconductor optical amplifier (e.g., the active element 110). As described below, the thermal performance of both the laser and the amplifier is improved by using a heatsink 130. In other embodiments, an active device (e.g., the active element 115) may simply be efficiently coupled to a waveguide with a large mode cross-section, guided to facet 155, and an output with a relatively high-quality beam profile and a large effective mode cross-section may be supplied to an optical fiber or another large mode cross-section device.

[0022] In either of the above cases, the size of the heatsink 130 can be significantly larger than the size of the active element. The advantages of this arrangement are shown in Figure 8.

[0023] The dashed line labeled A in Figure 1 corresponds to one characteristic cross-sectional location, which will be explained in more detail with reference to Figures 2 and 3. The dashed line labeled B in Figure 1 corresponds to yet another characteristic cross-sectional location, which will be explained in more detail with reference to Figure 4.

[0024] Figure 2 is a schematic cross-sectional view showing one embodiment of an integrated optical device 200 that uses butt coupling for efficient bonding between dissimilar materials. The cross-section shown in Figure 2 corresponds to one embodiment of the device at the position indicated by the dashed line A in Figure 1. Three different regions are shown: the active region, the transition region, and the large mode region.

[0025] The illustrated cross-section includes a substrate 205, which may be any substrate suitable for processing semiconductors and dielectrics, consisting of Si, InP, GaAs, quartz, sapphire, glass, gallium nitride (GaN), silicon-on-insulator (SOI), or other materials known in the art relating to the present invention. In the illustrated embodiment, a material layer (layer) 204-1 is mounted on the upper surface of the substrate 205 by deposition, growth, transfer, bonding, or other means using techniques known in the art relating to the present invention. The primary purpose of layer 204-1 is to provide an optical cladding to material 202-1 (described later) if necessary to form an optical waveguide. Optical waveguides are generally realized by placing a high refractive index core between two low refractive index layers that act as cladding for confining light waves. In some embodiments, layer 204-1 is silicon dioxide (SiO2). Furthermore, in other embodiments, layer 204-1 is omitted, and the substrate 205 itself functions as the cladding.

[0026] Layer 202-1 is mounted by deposition, growth, transfer, bonding, or other means using techniques known in the aforementioned art, on top of layer 204-1 if present, and / or on top of substrate 205 if layer 204-1 is not present. The refractive index of layer 202-1 is higher than that of layer 204-1 if present, and higher than that of substrate 205 if layer 204-1 is not present, and in either case, layer 202-1 functions as a waveguide core for optical mode 254. In one embodiment, the material of layer 202-1 includes, but is not limited to, one or more of SiN, silicon oxynitride (SiONx), titanium dioxide (TiO2), tantalum pentoxide (Ta2O5), (doped)SiO2, lithium niobate (LiNbO3), alumina (Al2O3), and aluminum nitride (AlN). Either or both of layers 204-1, 202-1, can be patterned, etched, or redeposited (by defining waveguides, splitters, couplers, grids, and other passive components) to adjust their functionality, as is common in the aforementioned technical fields.

[0027] Material layer 204-2 is deposited, grown, transferred, bonded, or otherwise implemented on the upper surface of layer 202-1 using techniques known in the aforementioned field. The primary purpose of layer 204-2 is to provide an optical cladding to both material 202-1 and material 202-2 (described later) and to form an optical waveguide.

[0028] Layer 202-2 is deposited, grown, transferred, bonded, or otherwise mounted on top of layer 204-2 using techniques known in the art according to the present invention. The refractive index of layer 202-2 is higher than that of layer 204-2, providing a waveguide core for optical mode 253.

[0029] In one embodiment, the material for layer 202-2 may be one or more of SiN, SiONx, TiO2, Ta2O5, (doped)SiO2, LiNbO3, Al2O3, and AlN, but is not limited thereto. Either or both of layers 204-2 and 202-2 may have their functionality modified (by defining splitters, couplers, gratings, and other passive waveguide components) by patterning, etching, or revolving, as is commonly done in the art according to the present invention.

[0030] The heat sink layer 230, corresponding to the heat sink 130 in Figure 1, is deposited on a well-prepared and etched recess in layer 204-2 (as shown in Figure 2). In some cases, although not shown, a portion of layer 202-2 may be located beneath layer 201 (as described below), and the recess may extend to that portion. In yet another embodiment, the etched recess may be deep enough to penetrate layer 202-1 and extend to layer 204-1, and even reach the substrate 205. Numerical calculations have shown that the effect of the thickness of layer 230 (corresponding to the depth of the etched recess) saturates, as described with respect to Figure 6, and that even with a relatively thin thickness of about 100 nm, the thermal performance of the active device can be significantly improved.

[0031] The refractive index of the optional layer (layer) 208 is lower than that of layer 202-2, and the optional layer 208 covers layer 202-2 and / or layer 230, with layer 201 and (described later) layer 203 positioned below the optional layer 208, and the optional layer 208 functions as a spacer layer. When layer 208 is present above layer 202-2 and / or layer 230, the target thickness of layer 208 is typically in the range of 10 nm to several hundred nm, but the actual thickness of layer 208 may be zero to several hundred nm greater than the target thickness due to process non-uniformity. In other embodiments, the thickness of layer 208 may be greater, approaching 1000 nm, or layer 208 may not be present at all. In some embodiments, layer 208 is SiO2.

[0032] If layer 208 is not present (not shown), layer 201 is bonded to at least a portion of the total upper surface of one or more of layers 204-2, 230, and 202-2. In the illustrated embodiment, layer 201 is bonded to the upper surface of layer 208. The bond may be a linear molecular bond or may be facilitated by the use of additional materials, such as a metal layer or polymer film, as is known in the art of the present invention. Layer 201 may be patterned to form what is commonly called an active device (e.g., active elements 105, 110, 115 in Figure 1) and may consist of materials including, but not limited to, InP and InP-based ternary and quaternary materials, GaP, InAs, InSb, and their deformation and dielectrics, or other suitable materials for providing linear light emission, amplification, modulation, and / or detection. In some embodiments, layer 201 has a multilayer structure and consists of sublayers that provide both optical and electrical confinement and electrical junctions, as is known in the field of active devices. In some embodiments, the sublayer of layer 201 provides vertical confinement (up / down the z-axis as shown in Figure 2), while lateral confinement (surface perpendicular to the cross-section shown in Figure 2, the y-axis which is not visible in this cross-section) is provided by at least one etching, as is known in the field of active devices, and is described with reference to Figure 4.

[0033] In some embodiments, layer 201 can be efficiently and electrically excited to generate light emission and gain. In other embodiments, layer 201 can provide modulation and / or detection. The present invention enables efficient optical coupling between the waveguide formed in layer 201 and layers 202-2, 202-1. The aforementioned materials (layers) 202-1, 202-2 may provide additional functions such as broadband transmittance, large effective mode cross-section (layer 202-1), optical propagation loss, high-intensity processing, phase shift, coupling, splitting, filtering, nonlinear generation, and / or other functions known in the art according to the present invention by temperature, strain, or other adjustment mechanisms.

[0034] Efficient coupling is facilitated by layer 203. Layer 203 acts as an intermediate waveguide, and in some embodiments, layer 201 accepts the optical mode profile (indicated linearly as 251) supported by the waveguide providing the core, efficiently captures it as a mode profile (optical mode) 252, gradually transmits it to the mode profile 253 provided by layer 202-2, and finally to the mode profile 254 provided by layer 202-1. Mode profile 254 has a larger effective mode region (effective mode cross-section) than mode profile 253 (commonly defined in relation to nonlinear effects). In some embodiments, the condition for such a large effective mode cross-section is M × λ 2 It is sometimes defined as corresponding to an effective mode cross-section larger than λ². Here, λ is the operating wavelength and M is a numerical coefficient, the value of which is selected and set as appropriate depending on the application, but is always greater than or equal to 1. In some embodiments, for example, M is 10, which means that the condition for a large effective mode cross-section is an effective mode cross-section larger than 10 × λ².

[0035] The refractive index and dimensions characterizing layer 203 may be designed to support an optimized mode 252 that facilitates efficient butt coupling of mode profile 251, and may also be designed to leverage the tapered structure created in layers 202-1, 202-2, and / or layer 203 to efficiently convert its modes to modes having mode profile 254. Transitions between modes 252, 253, and / or mode 254 utilize techniques well known in the art according to the present invention, employing evanescent coupling and / or adiabatic coupling. In some embodiments, the refractive index of layer 203 is lower than that of layer 202-2. In some embodiments, the refractive index of layer 203 is 1.44 to 1.9. The thickness of layer 203 is an optimization parameter, and in some embodiments, it is 400 nm to 4000 nm. The thickness depends largely on the details of layer 201 and the size of the optical mode 251.

[0036] The upper cladding layer (layer) 204-3 of the waveguide realized in layer 203 and / or layer 202-2 may be ambient air (meaning no cladding material is actually deposited), or it may be any suitable material that is intentionally deposited, including but not limited to polymers, SiO2, SiN, SiONx, etc., as shown in Figure 2. In some embodiments, the same material is used for layer 204-3 and layer 208. In some (not shown) embodiments, the cladding function of layer 204-3 may be provided by multiple depositions, for example, one material providing cladding for optical mode 253 induced by a core formed in layer 202-2, and another material providing cladding for optical mode 252 induced by a core formed in layer 203. In any case, the refractive index of the cladding material is lower than the refractive index of the material forming the core of the mode waveguide. In yet another (not shown) embodiment, since the refractive index of layer 203 is lower than that of layer 202-2, layer 203 can provide cladding functionality to layer 202-2 and optical mode 253, at least partially.

[0037] In the embodiment shown in Figure 2, the support of thick dielectric cladding (layers) 204-1, 204-2, and / or cladding 204-3 supports a large effective mode cross-section in the large mode region represented by optical mode 254. It also provides an efficient heat sink for layer 201 of the active device and low coupling losses between optical modes 251 and 254. In some embodiments, the thickness (z-direction) of cladding 204-1 exceeds 2 μm. In some embodiments, a relatively thick upper cladding on the waveguide core 202-1 in the large mode region may further provide low propagation losses and a useful degree of mechanical protection for that core.

[0038] Figure 3 is a schematic cross-sectional view of another embodiment of the photonic integrated device 300 using butt joints for efficient bonding between different materials. This cross-section corresponds to another embodiment of the device located at the position indicated by dashed line A in Figure 1, and illustrates three different regions: the active region, the transition region, and the large mode region. Functional layers 301-330 correspond to functional layers 201-230 described in relation to Figure 2 (unless otherwise explicitly defined).

[0039] In contrast to the embodiment shown in Figure 2, the embodiment shown in Figure 3 does not have two distinct waveguide core layers similar to layers 202-1 and 202-2, vertically separated by a cladding layer similar to layer 204-2. Instead of two elements corresponding to layers 202-1 and 202-2, there is a single waveguide core element (core, layer) 302 with a partially etched taper to adjust the thickness along the x-axis, which affects the transition between mode 353 supported by the thicker portion of the waveguide core element 302 and optical mode 354 supported by the thinner portion. An optional etching stop layer can improve process control of the etching depth, thereby reducing the thickness (z-direction) of element (layer) 302 in the larger mode region. The cladding layer 304-2 is stacked on top of layer 302 in the larger mode region. The transitions between optical modes 351, 352, and 353 follow the same flow as described above for optical modes 251, 252, and 253 in Figure 2. The transition between optical modes 353 and 354 utilizes a taper created by shallow etching of layer 302.

[0040] Figure 4 and section 400 show cross-sectional end views corresponding to the characteristic locations marked B in Figure 1. Functional layers 401-430 correspond to functional layers 201-230 described in relation to Figure 2 (unless explicitly defined otherwise). Section 400 shows an example of a cut region containing the active layer 401 and the heat sink layer 430. Layer 401 consists of multiple sublayers that provide the functionality necessary for realizing the active device. Such layers can be realized using any binary, ternary, or quaternary composition and any doping level, as is known in the design techniques of III / V semiconductor optical devices, to realize contact layers, active regions (quantum wells, quantum dots, pn junctions, pin junctions), cladding, gradient layers, isolated / confined heterostructure layers, bandgap smoothing layers, etch stop layers, etc. Optical mode 450 is restricted in the lateral direction (y-axis) by at least one etching (two are shown in Figure 4), while in the vertical direction, the refractive index difference between sublayers of layer 401 is used and, in some embodiments, restricted to cladding realized by at least one of layers 430, 408, 404-1, and 404-2. In some embodiments, additional material (not shown, but extensions of layers 202-2, 302, 204-2, and 304-2) is embedded within layer 430 and functions not only as cladding but also as an "anti-disping" element, as described below with reference to Figure 9.

[0041] The vertical position and height of the modes are defined by the sublayers of layer 401 and also by the cladding, as described above. Because the centers of the optical modes can be located away from the bottom surface (z-direction) of layer 401, and because there is a large difference in refractive index between layer 401 and the lower cladding layers (layers 430, 408, 404-1, 404-2, etc.), the optical modes 450 hardly overlap with layer 430. This allows for the placement of materials within layer 430 that have a high optical loss but minimize the impact on the mode loss of optical modes 450. The thickness (z-direction) and composition of the optional spacer layer (layer) 408 can further contribute to reducing mode loss due to material loss in layer 430. In some embodiments, the target thickness is typically in the range of 10 nm to several hundred nm, but due to process non-uniformity, the actual thickness is zero to several hundred nm greater than the target thickness. In other embodiments, the thickness of layer 408 is greater, approaching 1000 nm. Because the heat sink made of layer 430 is close to the active region where most of the heat is generated, thermal performance is significantly improved, as shown in the simulation results in Figures 5 and 6. One or more contacts 409 enable electrical control of the active device.

[0042] Figure 5 shows exemplary simulation results of temperature and heat flow through the laser cross-section at position "B" described in Figure 4, with and without a heat sink (Graphs 500, 510) and with a heat sink (Graphs 520, 530). In the latter case, the heat sink is fabricated from a 100 nm thick copper layer, and the stage temperature (corresponding to the bottom of each plot) is set to the same 25°C. As shown in the graphs in the left column, the temperature inside the laser cavity (junction temperature) decreased significantly from approximately 140°C (Graph 500) to approximately 60°C (Graph 520). The right column shows the heat flow, and Graph 530 clearly demonstrates the efficient heat sink function leading to the heat sink compared to Graph 510, which is without the intentionally introduced heat sink element.

[0043] Figure 6 shows an exemplary simulation of the average temperature of the laser cavity (junction temperature) as a function of the heatsink thickness for the thickness of the two lower cladding layers (layers 204-1 and 204-2, as described in Figure 2). Graph 600 shows an embodiment where the bottom cladding thickness is 1.5 μm, and Graph 620 shows an embodiment where the bottom cladding thickness is 15 μm. With a thickness of just 100 pm, the temperature for the 1.5 pm lower cladding drops from approximately 140°C to 70°C, and for the 15 pm lower cladding, it drops from 200°C to 100°C, demonstrating that even relatively thin heatsink layers have a very significant effect. Thermal performance improves further as the heatsink thickness increases, but the effect is not as pronounced. (In this particular simulation,) as the heatsink thickness approaches 400nm, the junction temperature can be reduced to approximately 70°C if the bottom cladding is thick, and to 40°C if the bottom cladding is thin. It is clear that even with a moderately thin heatsink, the thermal conditions at the junction can be significantly improved, and lower temperatures lead to improved wall-plug efficiency, output power, and the threshold for catastrophic optical damage in active devices.

[0044] Figure 7 and Graph 700 show an exemplary simulation of the average temperature of the laser cavity (junction temperature) as a function of the thickness of the spacer layer (e.g., corresponding to layer 208 in Figure 2). The thickness of the spacer layer has an almost exponential effect on the mode loss (because the evanescent field decays almost exponentially). However, the effect of the spacer layer thickness on thermal performance is much smaller, and for thicknesses less than 1000 nm, the effect is almost linear, which can lead to a large optimization space where excellent thermal performance can be achieved.

[0045] Figures 8 and Graph 800 show illustrative simulations of the average temperature (junction temperature) of the laser cavity as a function of the extension of the heatsink below the laser and the heatsink extension that extends beyond its lateral footprint. Such an extension, for example, in the case of PIC device 100 in Figure 1, is to extend the heatsink in the y-direction to be wider than the active element. Increasing the area and volume of the heatsink significantly reduces the temperature of the laser cavity.

[0046] Figure 9 shows a cross-sectional view of the heat sink 900 in the vertical direction of one embodiment of the heat sink 930. In this embodiment, there is a region within the heat sink that includes a layer 902, which in some embodiments corresponds to additional material from layer 202-2 described above in the description of Figure 2, and / or layer 302 described in relation to Figure 3. In other embodiments, layer 902 corresponds to additional material from layers 204-2, 304-2, as previously described in the description of Figure 4. The main purpose of newly providing layer 902 in the recesses prepared and etched for the heat sink 930 is to improve the uniformity of the planarization process of the upper surface of the heat sink 930, for example using chemical mechanical polishing (CMP), and to avoid the aforementioned "dishing". Although a simple arrangement of rectangular regions is shown, various patterns can be designed to improve such uniformity, as is known in the field of CMP or other planarization techniques.

[0047] It should be noted that the exemplary embodiments described above merely teach an example of a heterogeneous integrated laser and active component utilizing the present invention, and many other similar configurations are conceivable. Furthermore, such lasers and active components can be combined with several other components, such as various filtering elements, amplifiers, monitor photodiodes, modulators, and / or other photonic components, to provide additional functionality and improved performance.

[0048] Embodiments of the present invention offer numerous advantages. The integrated platform enables scalable manufacturing of PICs made from multiple materials, providing higher performance and / or the ability to operate over a wide wavelength range. Furthermore, this platform can handle higher optical power compared to typical Si waveguide-based or InP waveguide-based PICs.

[0049] As shown in Figure 2, the present invention typically utilizes a process flow in which a compound semiconductor material piece is wafer-bonded onto a carrier wafer equipped with a dielectric waveguide, and then a semiconductor manufacturing process known in the art according to the present invention is applied. This allows for precise definition of the optical alignment between the active waveguide and the passive waveguide, typically by a lithography process, eliminating the need for precise physical alignment. The aforementioned lithography-based alignment enables scalable manufacturing of devices using wafer-scale technology.

[0050] Since the optical coupling between the modes of the active and passive layers is reciprocal, taking Figure 2 as an example, the device structure can be configured to facilitate optical transmission from region (layer) 201 to region 202-1, but it should be noted that it can also facilitate transmission in the reverse direction from region 202-1 to region (layer) 201. Furthermore, it can be understood that in a properly configured PIC, there are no limitations on the number or direction, and multiple such transitions can be realized.

[0051] Other approaches have relied on die-attaching pre-fabricated optical active devices to passive waveguides. This approach requires extremely strict alignment precision, which typically exceeds the precision that typical die bonders can provide. This aspect limits the throughput and optical coupling performance of the aforementioned processes.

[0052] Embodiments of optical devices described herein may be incorporated into a variety of other devices and systems, including, but not limited to, various computing and / or civilian electronic devices / equipment, communication systems, medical devices, and various sensors and sensing systems.

Claims

1. It comprises a first element, a second element, a third element, a fourth element, and a fifth element fabricated on a common substrate, The first element comprises an active waveguide structure that supports a first optical mode, The second element comprises a passive waveguide structure having a second element waveguide core that supports a second optical mode, The third element comprises a passive waveguide structure having a third element waveguide core that supports a third optical mode, The fourth element comprises a heat sink thermally coupled to the first element, The fifth element is at least partially abutted against the first element and comprises an intermediate waveguide structure that supports intermediate optical modes. The effective mode cross-section characterizing the third optical mode is larger than the effective mode cross-section characterizing the second optical mode. The tapered waveguide structure in at least one of the second element and the fifth element facilitates efficient adiabatic conversion between the second optical mode and the intermediate optical mode. The tapered waveguide structure in at least one of the second and third elements facilitates efficient adiabatic conversion between the second and third optical modes. A device in which no adiabatic conversion occurs between the intermediate light mode and the first light mode.

2. The device according to claim 1, wherein the heat sink has at least one of silicon, silicon nitride, copper, gold, aluminum, tungsten, silver, zinc, diamond, graphite, silicon carbide, and aluminum nitride.

3. The device according to claim 1, wherein the first element and the fourth element are separated by a spacer having a thickness of more than 10 nm.

4. The device according to claim 1, wherein each of the cores of the waveguide structure of the second element and the core of the waveguide structure of the third element has at least one of silicon nitride, silicon oxynitride, titanium dioxide, tantalum pentoxide, doped silicon dioxide, lithium niobate, and alumina.

5. The device according to claim 2, wherein the thickness of the heat sink is 25 nm to 400 nm.

6. The device according to claim 2, wherein the heat sink further comprises at least one of silicon nitride, silicon oxynitride, titanium dioxide, tantalum pentoxide, doped silicon dioxide, lithium niobate, and alumina for more uniform flattening of the upper surface.

7. The device according to claim 1, wherein the core thickness of the waveguide structure of the third element is 20 nm to 200 nm, and the core thickness of the waveguide structure of the second element is 150 nm to 600 nm.

8. Designed to operate with light of wavelength L, The effective mode cross-sectional area characterizing the third optical mode is L ２ The device according to claim 1, which is larger than the device described in claim 1.

9. It comprises a first element, a second element, a third element, and a fourth element fabricated on a common substrate, The first element comprises an active waveguide structure that supports a first optical mode, The second element comprises a first passive waveguide structure having a first passive waveguide core that supports a second optical mode, and a second passive waveguide structure having a second passive waveguide core that supports a third optical mode. The third element comprises a heat sink thermally coupled to the first element. The fourth element is at least partially abutted against the first element and comprises an intermediate waveguide structure that supports intermediate optical modes. The tapered waveguide structure in at least one of the second element and the fourth element facilitates efficient adiabatic conversion between the second optical mode and the intermediate optical mode. The shallowly etched tapered structure in at least a portion of the second element facilitates efficient adiabatic conversion between the second optical mode and the third optical mode. The effective mode cross-section characterizing the third optical mode is greater than the tangent of the effective mode cross-section characterizing the second optical mode. A device in which no adiabatic conversion occurs between the intermediate light mode and the first light mode.

10. The device according to claim 9, wherein the heat sink has at least one of silicon, silicon nitride, copper, gold, aluminum, tungsten, silver, zinc, diamond, graphite, silicon carbide, and aluminum nitride.

11. The device according to claim 1, wherein the first element and the third element are separated by a spacer having a thickness of more than 10 nm.

12. The device according to claim 9, wherein each of the cores of the first passive waveguide structure and the core of the second passive waveguide structure has at least one of silicon nitride, silicon oxynitride, titanium dioxide, tantalum pentoxide, doped silicon dioxide, lithium niobate, and alumina.

13. The device according to claim 9, wherein the thickness of the heat sink is 25 nm to 400 nm.

14. The device according to claim 10, wherein the heat sink further comprises at least one of silicon nitride, silicon oxynitride, titanium dioxide, tantalum pentoxide, doped silicon dioxide, lithium niobate, and alumina for more uniform flattening of the upper surface.

15. The device according to claim 9, wherein the thickness characterizing the core of the second passive waveguide structure is 20 nm to 200 nm, and the thickness characterizing the core of the first passive waveguide structure is 150 nm to 600 nm.

16. Designed to operate with light of wavelength L, The effective mode cross-sectional area characterizing the third optical mode is L ２ The device according to claim 9, which is larger than the device described in claim 9.

17. The device according to claim 9, wherein the second element has an etching stop layer.

Images