CN117858653A

CN117858653A - Optical fiber to chip interconnection

Info

Publication number: CN117858653A
Application number: CN202280052040.3A
Authority: CN
Inventors: C·R·贾尔斯; P·J·文策尔; 张龙; P·J·普帕拉基斯; L·埃尔辛格
Original assignee: Nubis Communications Inc
Current assignee: Nubis Communications Inc
Priority date: 2021-05-25
Filing date: 2022-05-25
Publication date: 2024-04-09
Also published as: EP4346547A1; WO2022251855A1; CA3221063A1

Abstract

A method of assembling an optical device, the method comprising: providing a photonic integrated circuit comprising a plurality of vertical coupling elements disposed along a major surface of the photonic integrated circuit; attaching an optical subassembly to the photonic integrated circuit; removably connecting a fiber optic connector to the ferrule frame, wherein the fiber optic connector is attached to the fiber array; aligning the ferrule frame with the optical subassembly using an active alignment process; and securely connecting the ferrule frame to the optical subassembly after the active alignment process.

Description

Optical fiber to chip interconnection

Cross Reference to Related Applications

The present application claims priority from U.S. provisional patent application 63/192,852, U.S. provisional patent application 63/208,759, U.S. provisional patent application 63/210,437, U.S. provisional patent application 63/212,013, U.S. provisional patent application 63/223,685, U.S. provisional patent application 63/225,779, U.S. provisional patent application 63/245,005, U.S. provisional patent application 63/245,011, U.S. provisional patent application 025, U.S. provisional patent application 17/693,040, U.S. provisional patent application 63/316,551, and U.S. provisional patent application 17/693,040. The entire contents of the above application are incorporated by reference.

Background

Technical Field

Various example embodiments relate to optical communications devices and, more particularly, but not exclusively, to methods and devices for interconnecting fiber arrays with planar photonic integrated circuits.

Background

This section introduces aspects that may facilitate a better understanding of the disclosure. Accordingly, the statements of this section are to be read in this light, and not as admissions of prior art or nothing in the prior art.

With the increase in input/output (I/O) capacity of electronic processing chips, electrical signals may not provide adequate I/O capacity over the limited size of a practical electronic chip package. A viable alternative may be to interconnect electronic chip packages using optical signals, which may typically be delivered at a much higher I/O capacity per unit area than electrical I/O. In some examples, the optical fiber is optically coupled to the photonic integrated circuit, wherein the input optical signal is transmitted from the external device to the photonic integrated circuit through the optical fiber, and the output optical signal is transmitted from the photonic integrated circuit to the external device through the optical fiber.

Disclosure of Invention

Various embodiments of a connector assembly for optically connecting one or more optical fibers and an array of vertical coupling elements of a Photonic Integrated Circuit (PIC) are disclosed herein. In a general aspect, a method includes: providing a photonic integrated circuit comprising a plurality of vertical coupling elements disposed along a major surface of the photonic integrated circuit; attaching an optical subassembly to the photonic integrated circuit; removably connecting a fiber optic connector to the ferrule frame, wherein the fiber optic connector is attached to the fiber array; aligning the ferrule frame with the optical subassembly using an alignment process; and securely attaching the ferrule frame to the optical subassembly after an active alignment process.

Implementations can include one or more of the following features. The alignment process may include an active alignment process that includes transmitting light between at least one optical fiber and the photonic integrated circuit.

The active alignment process may include:

transferring light between at least one optical fiber of the array of optical fibers and the photonic integrated circuit through the optical subassembly and at least one vertical coupling element of the plurality of vertical coupling elements, and

the position of the ferrule frame relative to the optical subassembly is adjusted based on at least one characteristic of the light transmitted between the at least one optical fiber and the photonic integrated circuit.

The method may include removing the fiber optic connector from the ferrule frame.

The ferrule frame may include an opening to allow light from the fiber array to be transmitted to the optical subassembly.

The method may include passing a portion of the optical subassembly through an opening of the ferrule frame and positioning an end of the fiber optic connector adjacent the optical subassembly.

The method may include passing a portion of the fiber optic connector through an opening of the ferrule frame and positioning an end of the fiber optic connector adjacent the optical subassembly.

Removably connecting the fiber array to the ferrule frame may include at least one of: (i) Aligning the fiber array with the ferrule frame using one or more alignment pins; (ii) Using one or more clamps to secure the fiber array to the ferrule frame; (iii) Using one or more magnets to connect the fiber array to the ferrule frame; or (iv) using a removable adhesive to connect the fiber array to the ferrule frame.

The array of optical fibers may comprise a two-dimensional array of optical fibers.

The two-dimensional array of optical fibers may include at least two rows of optical fibers.

In some examples, the fiber array may include at least 10 cores.

In some examples, the fiber array may include at least 50 cores.

In some examples, the fiber array may include at least 100 cores.

The optical subassembly may include a first lens array, and the active alignment process may include projecting light from the array of optical fibers through the first lens array to a corresponding vertical coupling element, including passing light from at least one of the optical fibers through a corresponding lens to the corresponding vertical coupling element.

The optical subassembly may include a second lens array, and the active alignment process may include projecting light from the fiber array through the first and second lens arrays to the at least one vertical coupling element.

The optical subassembly may include a beam shifter, and the active alignment process may include projecting light from the fiber array through the first lens array, the beam shifter, and the second lens array to the at least one vertical coupling element.

The optical subassembly may include a half-wave plate, and the active alignment process may include projecting light from the fiber array through the first lens array, the beam shifter, the half-wave plate, and the second lens array to the at least one vertical coupling element.

The optical subassembly may include a spacer block disposed between the first lens array and the second lens array along an optical path, and the active alignment process may include projecting light from the fiber array through the first lens array, the spacer block, and the second lens array to the at least one vertical coupling element.

The optical subassembly may include a half-wave plate, and the active alignment process may include projecting light from the fiber array through the first lens array, the spacer block, the half-wave plate, and the second lens array to the at least one vertical coupling element.

The optical fiber connector may include a first lens array, and the active alignment process may include projecting light from the optical fiber array through the first lens array to a corresponding vertical coupling element, including passing light from at least one of the optical fibers through a corresponding lens to the corresponding vertical coupling element.

The active alignment process may include adjusting the position of the ferrule frame relative to the optical subassembly to maximize the overall efficiency of light transfer between the fiber array and the photonic integrated circuit.

At least half of the ferrule frame may be made of at least one of glass, metal or plastic by weight.

The ferrule frame may comprise a material that is transparent or translucent to Ultraviolet (UV) light, and fixedly connecting the ferrule frame to the optical subassembly may comprise attaching the ferrule frame to the optical subassembly using a UV-curable adhesive.

Adjusting the position of the ferrule frame relative to the optical subassembly may include adjusting the position of the ferrule frame along a plane substantially parallel to a major surface of the photonic integrated circuit.

Adjusting the position of the ferrule frame along the plane substantially parallel to a major surface of the photonic integrated circuit may include at least one of: (i) Adjusting the position of the ferrule frame along an x-axis relative to a major surface of the photonic integrated circuit; (ii) Adjusting the position of the ferrule frame along a y-axis relative to a major surface of the photonic integrated circuit; or (iii) rotating the ferrule frame about a z-axis relative to a major surface of the photonic integrated circuit. The x-axis and the y-axis may be substantially parallel to a major surface of the photonic integrated circuit, and the z-axis may be substantially perpendicular to a major surface of the photonic integrated circuit.

Adjusting the position of the ferrule frame relative to the optical subassembly may include adjusting a distance of an end of the fiber optic connector relative to the optical subassembly.

Adjusting the position of the ferrule frame relative to the optical subassembly may include adjusting an angle of inclination of an end face of the fiber optic connector relative to the optical subassembly.

In some examples, aligning the ferrule frame with the optical subassembly may include aligning the ferrule frame with the optical subassembly with an accuracy of at least 10 μm accuracy.

In some examples, aligning the ferrule frame with the optical subassembly may include aligning the ferrule frame with the optical subassembly with an accuracy of at least 1 μm accuracy.

In some examples, aligning the ferrule frame with the optical subassembly may include aligning the ferrule frame with the optical subassembly with an accuracy of at least 0.1 μm accuracy.

Each of the vertical coupling elements may comprise at least one of: single polarization vertical grating couplers, turning mirrors, polarization diversity vertical grating couplers, vertical cavity surface emitting lasers, surface normal line modulators, or photodiodes.

The beam shifter may include polarization dependent optics (polarization-dependent optical element).

The optical subassembly may include a turning mirror that turns a first optical path between the optical fiber and a corresponding vertical coupling element. The first optical path may include a first optical path section between the vertical coupling element and a reflective surface of the turning mirror, and a second optical path section between the reflective surface of the turning mirror and the optical fiber, the second optical path section being at an angle θ1 with respect to the first optical path section, and θ1 being in a range of 20 ° to 160 °.

In some examples, θ1 is in the range of 45 ° to 110 °.

In some examples, θ1 is in the range of 80 ° to 100 °.

After the ferrule frame is securely connected to the optical subassembly, the ferrule frame may be oriented such that, when the fiber optic connector is removably connected to the ferrule frame, at least some of the optical fibers in the array output a light beam in a direction substantially parallel to a major surface of the photonic integrated circuit, the output light beam travels at an angle θ2 with respect to a main surface of the photonic integrated circuit, and θ2 is 0 ° or more and 10 ° or less.

The optical subassembly may include a beam shifting element disposed between the steering mirror and the ferrule frame.

The optical subassembly may include a lens array disposed between the turning mirror and the photonic integrated circuit.

The method may include controlling a machine using a computer to align the ferrule frame with the optical subassembly using the active alignment process.

In another general aspect, an apparatus includes: a photonic integrated circuit comprising a plurality of vertical coupling elements disposed along a major surface of the photonic integrated circuit; an optical subassembly attached to the photonic integrated circuit; and a ferrule frame configured to enable a fiber optic connector to be removably connected to the ferrule frame and aligned with the optical subassembly. The fiber optic connector is connected to an array of optical fibers, and the optical subassembly is configured to transfer light between the array of optical fibers and the vertical coupling element on the photonic integrated circuit. The ferrule frame is aligned with the optical subassembly using an active alignment process in which light is transferred between at least one optical fiber in the array of optical fibers and the photonic integrated circuit through the optical subassembly and at least one of the plurality of vertical coupling elements. The position of the ferrule frame relative to the optical subassembly is adjusted based on at least one characteristic of the light transmitted between the at least one optical fiber and the photonic integrated circuit. After the active alignment process, the ferrule frame is securely connected to the optical subassembly.

Implementations can include one or more of the following features. In some examples, the ferrule frame enables the fiber array to be aligned with the optical subassembly with an accuracy of at least 10 μm.

In some examples, the ferrule frame enables the fiber array to be aligned with the optical subassembly with an accuracy of at least 1 μm.

In some examples, the ferrule frame enables the fiber array to be aligned with the optical subassembly with an accuracy of at least 0.1 μm.

The optical subassembly may include a first lens array, and a ferrule module may be configured to align the fiber array with the lens array.

The optical subassembly may include a second lens array, and the second lens array may be positioned along a beam path between the first lens array and the vertical coupling element.

The optical subassembly may include a beam shifter.

The optical subassembly may include a half-wave plate positioned along the beam path between the beam shifter and the second lens array.

The optical subassembly may include a birefringent plate having an aperture, and the birefringent plate may be positioned along the beam path between the beam shifter and the second lens array.

The optical subassembly may include a spacer block disposed between the first lens array and the second lens array along the beam path.

The optical subassembly may include a half-wave plate positioned along the beam path between the spacer block and the second lens array.

The fiber optic connector may include a first lens array, and the ferrule module may be configured to align the first lens array with the optical subassembly.

The optical subassembly may include a second lens array positioned along a beam path between the first lens array and the vertical coupling element.

The optical subassembly may include a beam shifter.

Each optical fiber may include one or more cores and the optical subassembly may include at least one lens configured to communicate light with a single one of the cores and a single one of the vertical coupling elements.

Each optical fiber may include one or more cores and the optical subassembly may include a plurality of optical waveguides, each optical waveguide optically connecting a respective one of the cores and a respective one of the vertical coupling elements.

At least some of the optical waveguides may be tapered.

The optical subassembly may include one or more polarizing beamsplitters.

The optical subassembly may include one or more polarization rotating elements.

Each optical fiber includes one or more cores, the optical subassembly may be configured to transfer light between a first number of the cores and a second number of the vertical coupling elements, and the second number is greater than the first number.

The ferrule frame may comprise a material that is transparent or translucent to Ultraviolet (UV) light, and a UV curable adhesive is used to securely attach the ferrule frame to the optical subassembly.

The beam shifter may include a polarization dependent optical element.

The optical subassembly may include a turning mirror that turns a first optical path between the optical fiber and a corresponding vertical coupling element. The first optical path may include a first optical path section between the vertical coupling element and a reflective surface of the steering mirror, and a second optical path section between the reflective surface of the steering mirror and the optical fiber, the second optical path section being at an angle θ1 with respect to the first optical path section, and θ1 being in a range of 20 ° to 160 °.

In some examples, θ1 is in the range of 45 ° to 110 °.

In some examples, θ1 is in the range of 80 ° to 100 °.

The ferrule frame may be oriented such that when the fiber optic connector is removably connected to the ferrule frame and the device is in operation, at least some of the optical fibers in the array output a light beam in a direction substantially parallel to a major surface of the photonic integrated circuit, the output light beam travels at an angle θ2 with respect to a main surface of the photonic integrated circuit, and θ2 is 0 ° or more and 10 ° or less.

In another general aspect, an apparatus includes a plurality of photonic integrated circuits, each photonic integrated circuit including a plurality of coupling elements. The apparatus includes a plurality of optical subassemblies, each optical subassembly attached to a corresponding photonic integrated circuit of the plurality of photonic integrated circuits. The apparatus includes a plurality of ferrule frames, each ferrule frame configured to enable a corresponding fiber optic connector to be removably connected to the ferrule frame and aligned with a corresponding one of the optical subassemblies. Each fiber optic connector is connected to an array of optical fibers, and the corresponding optical subassembly is configured to transfer light between the array of optical fibers and the corresponding coupling element on the corresponding photonic integrated circuit. Each ferrule frame enables the fiber array to be aligned with the corresponding optical subassembly with an accuracy of at least 10 μm accuracy.

Implementations can include one or more of the following features. In some examples, each ferrule frame enables the fiber array to be aligned with the corresponding optical subassembly with an accuracy of at least 1 μm accuracy.

In some examples, each ferrule frame enables the fiber array to be aligned with the corresponding optical subassembly with an accuracy of at least 0.1 μm accuracy.

Each optical subassembly may include a turning mirror that turns a first optical path between an optical fiber and a corresponding coupling element. The first optical path may include a first optical path section between the coupling element and a reflective surface of the steering mirror, and a second optical path section between the reflective surface of the steering mirror and the optical fiber, the second optical path section having an angle θ1 with respect to the first optical path section, and θ1 being in a range of 80 ° to 100 °.

In another general aspect, an apparatus includes: a storage device that stores instructions; and at least one data processor configured to execute the instructions and implement a process comprising controlling a machine to align the ferrule frame with the optical subassembly using an active alignment process. The optical subassembly is optically coupled to a photonic integrated circuit. The ferrule frame is configured to enable a fiber optic connector to be removably connected to the ferrule frame, the fiber optic connector attached to an array of optical fibers, and the ferrule frame is configured to align the optical fibers with the optical subassembly to transmit light between the optical fibers and the photonic integrated circuit.

In another general aspect, a data center includes any of the devices described above.

In another general aspect, a method includes operating the data center described above.

In another general aspect, a method includes operating any of the devices described above.

In another general aspect, a method includes assembling any of the devices described above.

In another general aspect, a method includes processing data using any of the devices described above.

The systems, devices, and methods described in this document may have one or more of the following advantages. By using the ferrule frame, the fiber array can be conveniently aligned with the optical subassembly with high precision to achieve efficient light transfer between the fiber core and the photonic integrated circuit. The process for aligning the ferrule frame with the optical subassembly is simple and can be performed in a relatively short amount of time, thereby making the process suitable for mass production operations. The ferrule frame can be manufactured simply and cost effectively. A data center having a large number of rack-mounted communication systems, each with a large number of pluggable modules, each configured to connect to an optical fiber array, may operate with increased efficiency and reduced maintenance costs.

Drawings

Other aspects, features, and benefits of the various disclosed embodiments will become more apparent from the following detailed description and drawings, by way of example, in which:

FIG. 1 illustrates a block diagram of an optical communication system in which at least some embodiments may be practiced;

FIG. 2 illustrates a schematic side view of an integrated optical device that may be used in the optical communication system of FIG. 1, in accordance with one embodiment;

fig. 3A-3G schematically illustrate various examples of one or more optical fibers that may be used in the optical communication system of fig. 1, in accordance with some embodiments;

FIG. 4 schematically illustrates an example fiber array that may be used in the optical communication system of FIG. 1, in accordance with an embodiment;

FIG. 5 illustrates a schematic cross-sectional side view of a fiber-to-PIC connector arrangement that may be used in the integrated optical device of FIG. 2, in accordance with an embodiment;

FIG. 6 shows a schematic cross-sectional side view of a fiber-to-PIC connector arrangement that may be used in the integrated optical device of FIG. 2, according to another embodiment;

FIG. 7 shows a schematic cross-sectional side view of a fiber-to-PIC connector arrangement that may be used in the integrated optical device of FIG. 2, according to yet another embodiment; and is also provided with

Fig. 8A and 8B illustrate schematic cross-sectional side views of a portion of a fiber-to-PIC connector arrangement that may be used in the integrated optical device of fig. 7, in accordance with some embodiments.

Fig. 8C to 8E are schematic diagrams of an example operation of walk-off crystals.

Fig. 9 is a schematic diagram of an example of a fiber-to-PIC connector.

Fig. 10A is a side view of an example of a polarization diversity assembly.

Fig. 10B is a side view of an example of a fiber-to-PIC connector.

Fig. 10C and 10D are top views of examples of polarization diversity assemblies.

Fig. 11A is a top view of an example of a fiber-to-PIC connector.

Fig. 11B and 11C are schematic diagrams showing examples of the direction of beam displacement from the crystal.

Fig. 12 is a schematic diagram of an example of a fiber-to-PIC connector.

Fig. 13A is a side view of an example of a polarization diversity assembly.

Fig. 13B is a schematic view of a fiber optic connector.

Fig. 13C is a top view of an example of a birefringent aperture plate.

Fig. 13D is a top view of an example of an array of grating couplers.

Fig. 14A to 18B show schematic diagrams of examples of arrangements of grating couplers and corresponding birefringent aperture plates.

Fig. 19A to 20E show schematic diagrams of examples of arrangements of fiber ports, corresponding birefringent aperture plates, and corresponding arrangements of grating couplers.

Fig. 21A to 21D are schematic diagrams of examples of birefringent aperture plates.

Fig. 22 is a schematic diagram of an array of grating couplers that achieve active alignment during assembly.

Fig. 23 is a schematic diagram of an example of a fiber-to-PIC connector.

Fig. 24A and 24B are side views of examples of fiber-to-PIC connectors.

Fig. 24C is a schematic diagram of an example of a walk-off direction of an optical fiber into a PIC connector.

Fig. 25 is a schematic diagram of an example of an optical energy source providing optical power through a single optical fiber.

Fig. 26 and 27 are schematic diagrams of examples of optical energy sources that provide optical power through a plurality of optical fibers.

Fig. 28 is a schematic diagram of an example of a fiber-to-PIC connector.

Fig. 29 and 30 are schematic diagrams of wavelength division multiplexers.

Fig. 31A to 32 are tables showing the allocation of wavelength division multiplexing channels.

FIG. 33A is a top view of an example of an optoelectronic device.

Fig. 33B and 33C are side views of example configurations of optoelectronic devices.

Fig. 34A is a side view of an example of a fiber-to-PIC connector.

Fig. 34B is a top view of an example of a fiber-to-PIC connector.

Fig. 35 is a schematic diagram of an example of a fiber-to-PIC connector that may process wavelength division multiplexed optical signals.

Fig. 36A to 36C show schematic diagrams of examples of arrangements of fiber ports, corresponding birefringent aperture plates and corresponding arrays of grating couplers.

Fig. 37A and 37B are diagrams illustrating examples of waveguide routing from a grating coupler to a modulator on a PIC.

Fig. 38 is a schematic diagram of an example of a fiber-to-PIC connector that may process wavelength division multiplexed optical signals from multiple rows of fibers.

Fig. 39 is a schematic diagram of an example of a fiber-to-PIC connector including a filter-based wavelength-division demultiplexer and a multiplexer.

Fig. 40 is a schematic diagram of an example of a fiber-to-PIC connector including an isolator.

Fig. 41 is a side view of an example of a fiber-to-PIC connector.

Fig. 42 is a top view and a side view of an example of a circularly asymmetric (or rotationally asymmetric) optical lens.

Fig. 43 is a schematic diagram showing a top view and a side view of a circular asymmetric optical lens array.

Fig. 44 is a schematic diagram showing a top view and a side view of a circularly symmetric optical lens array.

Fig. 45A is a schematic diagram illustrating an example of a fiber-to-PIC connector using a circular asymmetric optical lens array.

Fig. 45B is a schematic diagram showing an example of a fiber-to-PIC connector using a circularly symmetric optical lens array.

Fig. 46A and 46B are schematic diagrams illustrating examples of fiber-to-PIC connectors each coupling an array of grating couplers to an array of optical fibers having end faces polished at an angle.

Fig. 47 is a schematic diagram of an example of a fiber-to-PIC connector including an array of circularly asymmetric lenses and an array of circularly symmetric lenses.

Fig. 48 is a schematic diagram of an example of a fiber-to-PIC connector including an assembly of two different circular asymmetric lens arrays.

Fig. 49 is a schematic diagram of an example of a fiber-to-PIC connector including two circular asymmetric lens arrays, walk-off crystals, and a birefringent aperture plate.

Fig. 50 shows top and side views of an example of a patterned birefringent plate.

Fig. 51 shows top and side views of another example of a patterned birefringent plate.

Fig. 52A is a perspective view of an example of a patterned birefringent plate.

Fig. 52B to 52D are side views of examples of patterned birefringent plates.

Fig. 53A and 53B are schematic diagrams showing examples of a birefringent plate having a pattern generated in a birefringent element.

Fig. 53C is a schematic diagram showing incident light passing through the patterned birefringent plate.

FIGS. 54 through 60 are schematic diagrams of examples of optoelectronic data processing systems using circular polarization maintaining fibers.

Fig. 61 through 64 are schematic diagrams of examples of fiber-to-photonic integrated circuit connectors.

Fig. 65 is a schematic diagram of an example of a co-packaged optical module and the optical path of a light beam between a vertical coupling element and a fiber core.

Fig. 66 shows side and top views of an example of a co-packaged optical subassembly.

Fig. 67A and 67B are schematic diagrams illustrating that components in an optical subassembly may have several degrees of mechanical movement.

FIG. 68 is a schematic diagram of an example of a process for assembling an optical stack including a photonic integrated circuit, an optical subassembly, and a ferrule frame.

Fig. 69 is a top view of an example of a ferrule frame.

Fig. 70 is a schematic view of an example fiber array ferrule adapter coupled between an MPO connector and an optical assembly.

FIG. 71A is a top view of an example of an optoelectronic device.

FIG. 71B is a side view of an optoelectronic device.

Fig. 71C is a front view of an example of a ferrule frame.

FIG. 72 is a schematic diagram of an example of a process for assembling an optical stack including a photonic integrated circuit, an optical subassembly including turning mirrors, and a ferrule frame.

Fig. 73-75 are examples of optical stacks including a ferrule frame, turning mirrors, and spacer blocks.

Detailed Description

To accommodate the increasing demand for chip-to-chip interconnect bandwidth, the use of optical I/O may be beneficial.

Fig. 1 illustrates a block diagram of a communication system 100 in which at least some embodiments may be practiced. As shown, the system 100 includes a fiber 102 ₁ -102 ₁₁ Appropriately interconnected integrated optical communications device 101 ₁ -101 ₆ Optical fiber 102 ₁ -102 ₁₁ A communication path is established between the optical communication devices. The communication system 100 may also include one or more external optical energy modules 103, the external optical energy modules 103 generating Continuous Wave (CW) light or generating one or more periodic or non-periodic bursts of light for use in one or more integrated optical communication devices 101 ₁ -101 ₆ Used in the present invention. Some end-to-end communication paths may pass through the external optical energy module 103 (see, e.g., device 101 ₂ And 101 ₆ The communication path shown between). For example, device 101 ₂ And 101 ₆ The communication path between may be defined by fiber optic link 102 ₇ And 102 ₈ Together, whereby light from an external optical energy source 103 is multiplexed to the optical fiber link 102 ₇ And 102 ₈ And (3) upper part. Some end-to-end communication paths may pass through multiplexing unit 104 (see, e.g., apparatus 101 ₂ And 101 ₆ The communication path shown between). For example, device 101 ₂ And 101 ₆ The communication path between may be defined by fiber optic link 102 ₁₀ And 102 ₁₁ Together, whereby light from an external optical energy source 103 can be multiplexed to the optical fiber link 102 within the multiplexing unit 104 ₁₀ And 102 ₁₁ And (3) upper part.

The various elements of communication system 100 may benefit from the use of optical interconnects that may use photonic integrated circuits including optoelectronic devices that are co-packaged and/or co-integrated with electronic chips including integrated circuits.

As used herein, the term "photonic integrated circuit" (or PIC) should be construed to encompass Planar Lightwave Circuits (PLCs), integrated optoelectronic devices, wafer-level products on substrates, individual photonic chips and dies, and hybrid devices. Example material systems that may be used to fabricate various PICs may include, but are not limited to, III-V semiconductor materials, silicon photons, silicon-based silicon dioxide products (silicon-on-silicon products), silicon glass-based PLCs, polymer integration platforms, lithium niobate and derivatives, nonlinear optical materials, and the like. Both packaged devices (e.g., wired and/or encapsulated chips) and unpackaged devices (e.g., die) may be referred to as PICs.

PICs are used in a variety of applications in the fields of telecommunications, instrumentation, and signal processing. PICs typically use optical waveguides to implement and/or interconnect various circuit components, such as optical switches, couplers, routers, splitters, multiplexers/demultiplexers, filters, modulators, phase shifters, lasers, amplifiers, wavelength converters, optical-to-optical (O/E) and electro-optical (E/O) signal converters, and the like. The waveguides in PICs are typically solid-state-on-chip optical conductors that guide light due to the refractive index contrast between the core and cladding of the waveguide. PICs typically include planar substrates on which optoelectronic devices are grown by additive manufacturing processes and/or on which optoelectronic devices are etched by subtractive manufacturing processes, e.g., using a multi-step sequence of photolithography and chemical processing steps.

An "optoelectronic device" may operate under both light and current (voltage), and may include one or more of the following: (i) an electrically driven light source, such as a laser diode; (ii) an optical amplifier; (iii) a photoelectric converter, such as a photodiode; and (iv) an optoelectronic element that can control the propagation of light and/or certain properties of light, such as an optical modulator or switch. The corresponding optoelectronic circuit may additionally include one or more optical elements and/or one or more electronic components of an optoelectronic device capable of using the circuit in a manner consistent with the intended function of the circuit. Some optoelectronic devices may be implemented using one or more PICs.

As used herein, the term "integrated circuit" (IC) should be construed to encompass both non-packaged and packaged dies. In a typical IC fabrication process, wafers of silicon or other suitable materials are used to produce die (chips) in relatively large batches. The circuits and optical circuits may be developed on the wafer using a multi-step sequence of photolithography and chemical processing steps. Each wafer is then cut ("singulated") into a number of pieces (chips, dies), each piece including a respective copy of the circuit being fabricated. Each individual die may be packaged or remain unpackaged as appropriate before being incorporated into a larger circuit.

The term "hybrid circuit" may refer to a multi-component circuit made up of a plurality of monolithic ICs and possibly some discrete circuit components, all attached to each other so as to be mountable on and electrically connectable to a common base or carrier. A representative hybrid circuit may include (i) one or more packaged or unpackaged dies, wherein some or all of the dies include optical, optoelectronic, and/or semiconductor devices; and (ii) one or more optional discrete components, such as connectors, resistors, capacitors, and inductors. IC. Electrical connections between the die and the discrete components may be formed, for example, using patterned conductive (e.g., metal) layers, ball grid arrays, solder bumps, wire bonds, and the like. Individual ICs may include any combination of one or more respective substrates, one or more redistribution layers (RDLs), one or more interposer layers, one or more laminates, and the like.

In some embodiments, individual chips may be stacked. As used herein, the term "stacked" refers to an ordered arrangement of packaged or unpackaged dies, in which the major planes of the stacked dies are substantially parallel to each other. The stack may be mounted on the carrier in an orientation in which the major planes of the stacked dies are parallel to each other and/or to the major plane of the carrier.

A "principal plane" of an object, such as a die, PIC, substrate, or IC, is a plane parallel to a substantially planar surface of the principal plane that has a maximum dimension, e.g., length and width, in all exterior surfaces of the object. The substantially planar surface may be referred to as a major surface. The outer surface of an object having a relatively large dimension (e.g., length) and a relatively small dimension (e.g., height) is commonly referred to as an edge of the object.

Fig. 2 illustrates a schematic cross-sectional side view of an example integrated optical communications device 200 in accordance with an embodiment. Apparatus 200 may be used, for example, to implement apparatus 101 of fig. 1 ₁ -101 ₆ Is provided.

Device 200 includes a PIC 210, and PIC 210 is based on any suitable PIC technology/material platform, such as, but not limited to, silicon photons, indium phosphide, or lithium niobate. PIC 210 has supported on its substrate 201 appropriately connected passive optical elements and/or arrays thereof, such as waveguides 220, couplers, splitters, filters, delay lines, etc., and optoelectronic elements and/or arrays thereof, such as modulators, detectors, and tunable phase shifters. Some of these elements may be vertical coupling elements 231, with the vertical coupling elements 231 configured to couple light to/from the PIC. Herein, a "vertical" direction is a direction perpendicular to a major surface of the PIC. In the context of the present disclosure, the term "vertically coupled" means coupled at an angle substantially out of plane with respect to, but not necessarily perpendicular to, a major surface of substrate 201. Vertical coupling is typically implemented at an angle between 0 degrees (vertical) and 45 degrees measured from the surface normal of the major surface of the substrate. The vertical coupling may be performed from the top side (e.g., waveguide side) of the PIC (271 in fig. 2) or from the bottom side (e.g., substrate side) of the PIC (272 in fig. 2).

In some embodiments, the vertical coupling element 231 may be implemented as, for example, a turning mirror, a vertical grating coupler, an elephant coupler (eleuthar coupler), or as a 3D vertical coupling structure that is 3D printed onto the PIC, suitably connected to a passive optical element or optoelectronic element. In an example embodiment, the vertical coupling element 231 may be implemented, for example, using any of the vertical coupling elements disclosed in the following patent documents: US 2015/0037044, US 2015/0125115, US2015/0293305, US 9927575, US 2018/032959, US 2019/0258175 and US10025043. All of these U.S. patents and U.S. patent application publications are incorporated herein by reference in their entirety.

In some embodiments, the vertical coupling element 231 may be a surface normal optoelectronic element, such as a surface normal modulator, a surface normal detector, or a surface normal laser, for example, a Vertical Cavity Surface Emitting Laser (VCSEL). In an example embodiment, the vertical coupling element 231 may be implemented, for example, using any of the vertical coupling elements disclosed in U.S. patent and U.S. patent application publications US2019/0312642, US10025043, and US 8488921, all of which are incorporated herein by reference in their entirety.

The vertical coupling elements 231 may be geometrically differently arranged in an array 230 of such vertical coupling elements.

In some embodiments, some optical or optoelectronic elements may be spatially co-located or interspersed with some of the vertical coupling elements 231 of the array 230.

In some embodiments, some optical or optoelectronic elements may be located in areas of the PIC that do not intersect the vertical coupling array 230.

The optical and optoelectronic components of the PIC are suitably connected to an electronic integrated circuit 260, such as a driver amplifier, a transimpedance amplifier, an electronic control circuit, digital logic, a microcontroller, a microprocessor, and/or an electronic switch. Some of the electronic circuitry may be spatially co-located or interspersed with some of the vertical coupling elements of the array 230, and some of the electronic circuitry may be located in regions that are spatially disjoint from the array 230. Some electronic circuits may be monolithically integrated with the optical or optoelectronic components of the PIC. Some of the electronic circuitry may be located on a separate chip from the PIC and may be electrically connected to the PIC using suitable electrical interconnect techniques such as bond wires, balls, bumps, micro-bumps, pillars, and membranes (e.g., in a stacked form).

Of particular interest in the context of the present disclosure are connector structures 271 and 272, the connector structures 271 and 272 being capable of enabling (possibly pluggable and/or removable) connection between M spatial paths of one or more optical fibers 202 as part of the optical fiber link 102 and N vertical coupling elements of the array 230 of PICs. In some embodiments, the numbers N and M are different integers greater than one. In some other embodiments, n=m.

In the context of the present disclosure, the term "spatial path" refers to an optical path through one or more spatially coupled cores of a single-mode or multimode optical fiber, a core of a multi-core optical fiber, or a few-mode optical fiber, the spatial path being configured to transmit different signals in different spatial modes thereof. The spatial path may carry signals in one or more polarizations and/or at one or more wavelengths. In some embodiments, the spatial path may be polarization preserving. The one or more optical fibers 202 may include single mode fibers, multimode fibers, few mode fibers, multi-core fibers, and/or polarization maintaining fibers. The one or more optical fibers 202 may include dispersion shifted, dispersion compensated, non-zero dispersion shifted, standard single mode dispersion, and/or high dispersion optical fibers. The one or more optical fibers 202 may be fixedly attached (e.g., glued) to the connector element 250, for example, by positioning individual optical fibers in a single hole provided within the connector element 250, or by positioning individual optical fibers in a linear array of V-grooves and stacking a plurality of such linear arrays to form a 2D array. Thus, the M spatial paths of one or more optical fibers 202 may form an array having a particular geometric layout and having a particular spatial path spacing in the fiber end face plane 243. The fiber-optic endface plane 243 may be parallel to a major surface of the PIC (e.g., as shown in the illustrated detail of structure 271 of fig. 2), or may be at a non-zero angle relative to a major surface of the PIC (e.g., as shown in the illustrated detail of structure 272 of fig. 2). In various embodiments, the angle relative to the major surface of the PIC may be suitably selected between 0 degrees (in which case the corresponding fiber-optic endface plane is parallel to the major surface of the PIC) and 90 degrees (in which case the corresponding fiber-optic endface plane is perpendicular to the major surface of the PIC).

Connector element 240 may be fixedly attached (e.g., glued) to PIC 210, for example, by aligning and then attaching the connector element to PIC 210 during assembly. Connector element 240 may be attached to either of the two major surfaces of PIC 210. Connector element 240 may be fixedly or movably attached to connector element 250 in connector mating plane 241. Connector mating plane 241 may be parallel to a major surface of the PIC (e.g., as in structure 271 of fig. 2) or may be angled with respect to a major surface of the PIC (e.g., as in structure 272 of fig. 2). The angle relative to the major surface of the PIC may be selected between 0 degrees (in which case the corresponding connector mating plane is parallel to the major surface of the PIC) and 90 degrees (in which case the corresponding connector mating plane is perpendicular to the major surface of the PIC). In some embodiments, connector elements 240 and 250 may include mechanical structures that enable elements 240 and 250 to self-align. For example, such mechanical structures may be implemented using a cylindrical or conical column and hole arrangement (post-and-hole arrangement), a rod and slot arrangement (rod-and-groove arrangement), or a ball and hole arrangement (ball-and-hole arrangement). The connector elements 240 and 250 may further include mechanical structures, such as suitable snap-fit mechanisms, that can hold the elements 240 and 250 in place after mating.

Any of the connector elements 240 and 250 may include one or more of the following: (i) Reflective optical elements such as dielectric interfaces or metal interfaces; (ii) refractive optical elements such as lenses or prisms; (iii) a diffractive optical element, such as a grating; (iv) Birefringent optical elements such as calcite crystals, polarization gratings or waveplates; (v) A 3D waveguide or nanostructure written in a suitable host material such as glass; and/or (vi) 3D printed optical waveguides, microstructures or nanostructures. The combination of connector elements 240 and 250 is generally designed to properly map the M spatial paths of one or more optical fibers 202 in the fiber-optic endface plane 243 to the N vertical coupling elements of the array 230 in the coupling plane 242. A corresponding set of optical fibers 202, connector elements 240 and 250, and vertical coupling array 230 together form a connector assembly 271 or 272. Some embodiments disclosed herein are specifically directed to providing an optimized design of connector assemblies 271 and 272, for example, with respect to tolerances in manufacturing, assembly, and operation. Some of such embodiments may scale to a relatively large number of spatial paths, e.g., M >100.

Fig. 3A-3G illustrate a configuration of one or more optical fibers 202 according to some embodiments. More specifically, fig. 3A-3G schematically illustrate example cross-sectional views of one or more optical fibers 202 in a fiber coupling plane 243 in accordance with various embodiments.

Fig. 3A illustrates supporting m=6 spatial lanesA one-dimensional (1D) array of radial single-core single-mode fibers. Each of the six optical fibers shown includes a respective cladding 301 and a respective core 302, typically made of glass having different refractive indices, such that the refractive index of the cladding is lower than the refractive index of the core to create a dielectric optical waveguide. In some embodiments, more complex refractive index profiles, such as refractive index grooves, multi-refractive index profiles, or progressively changing refractive index profiles, may also be used. In some embodiments, more complex geometries such as non-circular cores or cladding layers, photonic crystal structures, photonic bandgap structures, or nested antiresonant non-nodal hollow core structures may also be used. For any of these structures, the geometry, structure, and material properties may be appropriately selected to allow a single guided (e.g., lateral) mode to propagate over the operating wavelength range of the system 100. Three feature sizes are of particular interest in the context of the present disclosure: (i) Effective core diameter D _core Generally defined as the decrease in light intensity of a propagating mode within the fiber to a value of 1/e of the light intensity at the center of the core ² The diameter at time (sometimes also referred to as the mode field diameter); (ii) Minimum core-to-core spacing S within an array _min The method comprises the steps of carrying out a first treatment on the surface of the And (iii) the maximum core-to-core spacing S within the array _max . FIG. 3A shows a feature size D corresponding to this particular embodiment _core 、S _min And S is _max 。

Fig. 3B shows a two-dimensional (2D) array of single-core single-mode optical fibers supporting m=12 spatial paths. FIG. 3B shows a feature size S corresponding to this particular embodiment _min And S is _max 。

Fig. 3C shows a two-dimensional (2D) array of single-core single-mode optical fibers supporting m=17 spatial paths. FIG. 3C shows a feature size S corresponding to this particular embodiment _min And S is _max 。

Although fig. 3A-3C illustrate only three example geometric array layouts and spacings, other geometric array layouts may be used in various alternative embodiments. Based on the description provided, one of ordinary skill in the art will be able to make and use such other geometric array layouts without any undue experimentation. Some embodiments may also be constructed using one or more arrays of optical fibers having different properties, such as a mixture of optical fibers having different refractive index profiles, different effective core diameters, and the like.

Fig. 3D shows a multi-core single mode fiber supporting m=7 spatial paths. The multi-core fiber includes a cladding 301, typically made of glass having different refractive indices, and seven cores 302, such that the refractive index of the cladding is lower than that of the cores. In some embodiments, more complex refractive index profiles, such as refractive index grooves, multi-refractive index profiles, or progressively changing refractive index profiles, may also be used. More complex geometries such as non-circular cores, non-circular cladding, photonic crystal structures, photonic bandgap structures, or nested antiresonant non-nodal hollow core structures may also be used. For any of these structures, the geometry, structure, and material properties may be selected to allow a single guided (e.g., transverse) mode of each core to propagate over the operating wavelength range of the system 100. Regardless of the complexity of these structures, the effective core diameter D of each core can be defined _core . Different cores within an optical fiber may have nominally the same or substantially different (e.g., differing by more than 10%) effective core diameters. FIG. 3D shows a feature size D corresponding to this particular embodiment _core 、S _min And S is _max 。

Fig. 3E shows a multi-core single mode fiber supporting m=4 spatial paths. FIG. 3E shows a feature size S corresponding to this particular embodiment _min And S is _max 。

Fig. 3F shows a multi-core single mode fiber supporting m=8 spatial paths. FIG. 3F shows a feature size S corresponding to this particular embodiment _min And S is _max 。

Fig. 3G shows a multi-core single mode fiber supporting m=4 spatial paths. FIG. 3G shows a feature size S corresponding to this particular embodiment _min And S is _max 。

Although fig. 3D-3G only show four example geometric core layouts and spacings, other geometric core layouts may be used in various alternative embodiments. Based on the description provided, one of ordinary skill in the art will be able to make and use such other geometric core layouts without any undue experimentation.

Fig. 4 illustrates a configuration of one or more optical fibers 202 in accordance with some embodiments. More specifically, fig. 4 schematically illustrates an example cross-sectional view of one or more optical fibers 202 in a fiber coupling plane 243 in accordance with various embodiments.

Fig. 4 illustrates an example two-dimensional (2D) array of multi-core single mode fibers supporting m=90 spatial paths. In some embodiments, different optical fibers within an array may have different respective core counts, different respective effective core diameters, and/or different respective rotational orientations. FIG. 4 shows a feature size S corresponding to this particular embodiment _min And S is _max 。

In some embodiments, some cores of some of the multi-core fibers shown in fig. 3D-3H and fig. 4 may be designed to be substantially uncoupled, e.g., exhibit core-to-core crosstalk of less than 20dB over a propagation distance of 1km, or may be designed to be relatively strongly coupled. Some cores of the single-core and/or multi-core optical fibers shown in fig. 3 and 4 may be designed to be few-mode or multi-mode, i.e., may be designed to propagate a relatively small number (e.g., < 10) or a relatively large number (e.g.,. Gtoreq.10) of transverse modes.

Important additional aspects of coupling a large number of spatial paths from optical fiber 202 to PIC 210 may include consideration of the relative sizes of the practically available optical fibers and optical elements, optoelectronic elements, and electronic elements, and their placement within a correspondingly large array. For example, the relatively close desired spacing in some areas of the PIC may indicate that forming a larger array may be difficult, which creates difficult scalability issues. In addition, in some cases, the relative alignment of a typical core and a typical vertical grating coupler may require placement accuracy on the order of 1 micron or better to achieve low coupling loss. However, such requirements may not be compatible with typical accuracy achieved using conventional passive alignment processes, which may disadvantageously require the use of slower and/or more expensive active alignment processes.

After investigating some of the disadvantages of the existing fiber-to-PIC coupling structures, they have been identified and examined by analysis, modeling and simulationVarious designs of removable fiber-to-PIC connected optical coupling structures suitable for mass production supporting arrays of large numbers of spatial paths. In particular, contemplated solutions may allow for efficient coupling between M spatial paths of one or more optical fibers 202 and an array 230 of N vertical coupling elements by implementing some or all of the following features: (i) Amplifying or narrowing the minimum core-to-core spacing of the optical fibers in the fiber-optic endface plane 243 by a first factor (denoted as a) to match the minimum spacing between the vertical coupling elements in the coupling plane 242; (ii) Amplifying or narrowing the maximum core-to-core spacing of the optical fibers in the fiber-optic endface plane 243 by a second factor (denoted B) to match the maximum spacing between the vertical coupling elements in the coupling plane 242; (iii) With a third factor (denoted as C ₁ ) Enlarging or reducing the effective core diameter of the optical fiber in the fiber-optic endface plane 243 to match the effective vertical grating coupler size in the coupling plane 242; (iv) By a fourth factor (denoted as C ₂ ) The effective core diameter of the optical fibers in the fiber-optic endface plane 243 is enlarged or reduced to achieve a substantially different (e.g., larger) effective beam diameter in the connector mating plane 241 as compared to the effective beam diameter in the fiber-optic endface plane 243; and/or (v) altering the effective cross-sectional geometry of the plurality of spatial paths in at least some regions between the fiber-optic endface plane 243, the connector mating plane 241, and the coupling plane 242. In an example embodiment, the factor A, B, C ₁ And C ₂ At least some or all of the factors may be different.

For examples of possible benefits that may be obtained, a=b=2 and C may be considered ₁ Example embodiment of =1.5. In this particular embodiment, C ₁ Allowing for relaxed alignment tolerances for attaching connector assembly 240 to PIC 210. Scaling of a and B allows for even wider optical waveguide spacing within PIC 210, thereby potentially reducing waveguide-to-waveguide crosstalk and/or enabling the use of relatively large arrays.

Fig. 5 illustrates a fiber-to-PIC connector arrangement 500 that may be used in apparatus 200 (fig. 2) in accordance with an embodiment. As shown, the connector arrangement 500 includes an array 501 of multicore fibers (MCFs) 202 connected to connector elements 250. The end faces of the MCF 202 are arranged substantially in the same plane, i.e., the fiber end face plane 243 (see also fig. 2). The connector element 250 is further connected to the connector element 240 and the interface between the two connector elements comprises a connector mating plane 241 (see also fig. 2).

The connector element 250 includes one collimating lens 551 per MCF 202. In an example embodiment, collimating lens 551 may be arranged to provide an enlarged beam spot size in connector mating plane 241. For example, an effective core diameter of 10 microns together with the focal length f of the collimating lens 551 ₁ Is f ₁ =500 micrometers may result in an effective beam diameter in the connector mating plane 241 of about 100 micrometers.

Connector element 240 includes one focusing lens 541 per MCF 202. The longitudinal dimensions of the connector elements 240 and 250 may be selected such that the connector mating plane 241 is located at any convenient position between the collimating lens 551 and the focusing lens 541. For example, such dimensions may be selected such that the beam diameter in connector mating plane 241 expands by C ₂ And approximately 10 times. Such an extension may be beneficial because it may significantly simplify connector alignment. In alternative embodiments, other longitudinal dimensions may be similarly selected to achieve a factor of C ₂ Other values of (2).

In the example embodiment shown in fig. 5, each focusing lens 541 has a focal length f ₂ ＝2f ₁ . This focal length ratio results in the entire core pattern of each MCF being magnified a=2 times in the coupling plane 242. For example, a minimum core-to-core spacing S _min (see, e.g., fig. 3D) is also twice as large in the coupling plane 242. This amplification applies to both the pitch of the MCF cores and the characteristic mode size corresponding to each individual core.

To independently select the effective magnification applied to the individual spatial paths between the fiber-optic endface plane 243 and the coupling plane 242, each spatial path is directed through a respective individual lens 542. For example, in the embodiment of FIG. 5, each individual lens 542 has a focal length f ₃ =70 microns, and thus is relatively reduced by 75%. Thus, the overall characteristic pattern size is achievedMagnification C ₁ =2x0.75=1.5. The larger effective mode size in the coupling plane 242 may advantageously help relax positional tolerances of the connector element 240 relative to the array of vertical coupling elements 230 in the coupling plane 242 as compared to the fiber end face plane 243.

In some embodiments, some or all of lenses 542 may be laterally offset from the center of the corresponding illumination beam. Such lateral offset causes the light beam 543 directed toward the vertical coupling element 231 of the array 230 to strike the coupling element at a desired coupling angle (e.g., not necessarily along a normal to the major surface of the corresponding PIC). Note that in this example, the maximum core-to-core spacing remains substantially unchanged because the applied magnification occurs on a per MCF basis, thereby achieving a B value of b≡1.

In the above example, the geometry scaling parameter set { A, B, C } ₁ ，C ₂ And is approximately 2,1,1.5, 10. However, the geometry scaling parameter set { A, B, C ₁ ，C ₂ Other combinations of values of the may also be achieved, for example, by appropriate selection of the relevant dimensions, positions and focal lengths. From the above description, one of ordinary skill in the art will be able to implement such other numerical combinations as desired without any undue experimentation.

Furthermore, the lens system shown in fig. 5 represents only one of many possible ways to perform independent array pattern scaling and mode size scaling using refractive optical elements. For example, a given array pattern scaling may be performed on any different subset of the spatial paths corresponding to the fiber-optic endface plane 243. The different subsets may have the same or different corresponding magnification factors. When the individual subsets are scaled differently, the overall array pattern geometry of the fiber-optic endface plane 243 may be transformed to produce geometrically different array patterns in the coupling plane 242.

In some embodiments, pattern scaling may also be performed over the entire set of spatial paths corresponding to the fiber-optic endface plane 243, such as by using a single lens element 551 that spans laterally across the entire array 501, thereby producing a geometrically similar scaled image of the array 501 in the coupling plane 242, as the latter is presentedThe lens system in the fiber-optic endface plane 243 is now presented. Parameter sets { A, B, C {2,2,1.5, 10} may be implemented using an example embodiment of this design ₁ ，C ₂ }。

In some embodiments, mode field diameter scaling may be performed on any subset of the spatial paths corresponding to the fiber-optic endface plane 243, and the same or different corresponding scaling (e.g., magnification) factors may be used for different spatial paths.

In some embodiments, aspheric lenses and arrays thereof may be used. Such lenses may be manufactured, for example, using wafer-level processing techniques.

In some embodiments, the functions of lenses 541 and 542 may be combined into a single aspheric refractive element that may be 3D printed using techniques such as those provided for sale by nanostrinbe corporation (Nanoscribe of Eggenstein-Leopoldshafen, germany) of taiyin-riobert harbor, germany.

As will be appreciated by those of ordinary skill in the art, it is also possible to set the angle of the optical fiber 202 relative to the major plane of the PIC and to select the angle of incidence of the individual light beams 543 on the vertical coupler array 230, for example, by mounting the optical fiber 202 in the connector element 250 in an angled or curved manner, tilting the connector mating plane 241 at an angle relative to the major plane of the PIC, and/or introducing metallic or dielectric reflective interfaces, refractive elements (e.g., prisms), and/or diffractive elements (e.g., gratings) at appropriate locations within the assembly 500.

Fig. 6 illustrates a fiber-to-PIC connector arrangement 600 that may be used in apparatus 200 (fig. 2) in accordance with another embodiment. As shown, the connector arrangement 600 includes an array 501 of MCFs 202 connected to the connector element 250. The end faces of the MCF 202 are arranged substantially in the same plane, i.e., the fiber end face plane 243 (see also fig. 2). The connector element 250 is further connected to the connector element 240 and the interface between the two connector elements comprises a connector mating plane 241 (see also fig. 2).

Connector element 250 includes an array of 3D waveguides 652 formed (e.g., optically written) in a suitable matrix material such as glass using a suitable technique, such as some of the products offered for sale by optoscript corporation of livinston, england (Optoscribe of Livingston).

In some embodiments, the 3D waveguides 652 written into the connector element 250 may expand or appropriately geometrically rearrange the array geometry of the spatial paths provided by the optical fibers 202 at the fiber-end-face plane 243. In the embodiment shown in FIG. 6, the 3D waveguide is positioned at C between the fiber endface plane 243 and the connector mating plane 241 ₂ The factor of =2 expands the mode field diameter.

In some embodiments, the 3D waveguides 652 of the connector element 250 may individually expand the mode field diameters of the respective waveguides to achieve expanded beam connection at the connector mating plane 241. This may be achieved by using a tapered or inverted tapered structure within the 3D waveguide arrangement 652 and/or by varying one or more 3D waveguide write parameters, such as the scan speed or repetition rate of the femtosecond laser pulses used to write the 3D waveguide 652, resulting in a larger 3D waveguide mode field diameter.

In some embodiments, the 3D waveguide 652 in the connector element 250 may also introduce bend angles, e.g., to accommodate different angles of incidence of light from the optical fiber 202 (e.g., from a fiber-optic endface plane that is not parallel to the major surface of the PIC). In some embodiments, the 3D waveguide bend may be combined with a change in reflection or refraction angle due to a properly placed dielectric or metal interface (not explicitly shown in fig. 6) or with a change in diffraction angle due to a properly placed grating (not explicitly shown in fig. 6).

The connector element 240 may use a 3D waveguide 644, some of the mode field diameters of the waveguide 644 expanding at the connector mating face 241 relative to typical fiber mode field diameters in the fiber end face plane 243 to substantially match the mode field diameters of the corresponding waveguides of the connector element 250 at the connector mating face 241.

In some embodiments, the 3D waveguide 644 of the connector element 240 may appropriately change the array size, array geometry, mode size, and angle of incidence in order to match the corresponding geometric parameters at the coupling plane 242.

In the example embodiment shown in fig. 6, each waveguide mode field diameter is from C at the connector mating plane 241 ₂ The magnification of =2 is reduced to 75% thereof, thereby producing a total mode field diameter magnification C from the optical fiber 202 in the fiber end face plane 243 to the vertical coupler array 230 in the coupling plane 242 ₁ =2x0.75=1.5. The larger effective mode size in the coupling plane 242 may advantageously help relax positional tolerances of the connector element 240 relative to the array 230 of vertical coupling elements 231 in the coupling plane 242 as compared to the fiber end face plane 243.

In some embodiments, some or all of waveguide bends 645 may establish a desired coupling angle with vertical coupling elements 231 of array 230.

The 3D waveguide system described above should be seen as only one of many possible embodiments that may be used to perform independent array mode scaling, array mode geometry transformation, spot size scaling, and angle of incidence adaptation. In some embodiments, a hybrid assembly is also possible and may be considered to be a functional equivalent of the embodiments described above. Some embodiments may use any suitable combination of diffractive, reflective or refractive surfaces, 3D waveguides, and 3D printed structures within either or both of connector elements 240 and 250.

Some embodiments may be configured to use polarization diversity optics within connector assemblies 271 and 272. For example, some cores of the one or more optical fibers 202 may transmit randomly polarized signals or may transmit polarization multiplexed signals. In addition, some vertical grating couplers may be polarization sensitive. Properly coupling dual polarized light from one or more optical fibers 202 to PIC 210 may therefore benefit from polarization diversity vertical coupling elements, such as two-dimensional polarization diversity vertical grating couplers. Some polarization diversity vertical coupling elements may have an inherently higher insertion loss than that of a single polarization vertical coupling element. Thus, it may be beneficial to replace one polarization diversity vertical coupling element with a pair of single polarization vertical coupling elements and perform polarization diversity outside the PIC, e.g., within connector assemblies 271 and 272.

Some embodiments may benefit from the use of polarization diversity optics such as disclosed in U.S. patent No. 9927575, which is incorporated herein by reference in its entirety.

Fig. 7 illustrates a fiber-to-PIC connector arrangement 700 that may be used in apparatus 200 (fig. 2) in accordance with yet another embodiment. As shown, the connector arrangement 700 includes an array 501 of MCFs 202 connected to the connector element 250. The end faces of the MCF 202 are arranged substantially in the same plane, i.e., the fiber end face plane 243 (see also fig. 2). The connector element 250 is further connected to the connector element 240 and the interface between the two connector elements comprises a connector mating plane 241 (see also fig. 2).

The embodiment shown in fig. 7 is configured to couple m=8 spatial paths of optical fibers 202 in fiber-end-face plane 243 to n=16 > M vertical coupling elements 231 of array 230.

The connector element 250 includes one collimating lens 551 per MCF 202. In an example embodiment, collimating lens 551 may be arranged to provide an enlarged beam spot size in connector mating plane 241. The connector element 250 further includes a polarization diversity component 757.

Fig. 8A shows a schematic side view 810 of a sub-element of a polarization diversity assembly 757 according to an embodiment. As shown, the assembly 757 includes a birefringent beam shifting element 753 (also referred to as a walk-off element). In some embodiments, element 753 may be made of, for example, properly oriented calcite, yttrium vanadate (YVO ₄ ) Or a-BBO, such as those offered for sale by Fuzhou Meiyang photoelectric limited (MT-Optics of Fuzhou, fujian, china) in Fuzhou, fujian, china. The birefringent beam shifting element 753 is used to split the incident beam 754 into a corresponding pair of outgoing beams 755a and 755b. Thus, beams 755a and 755b include respective light of two orthogonal polarization states of incident beam 754. To prepare light beams 755a and 755b for coupling to parallel aligned (opposite to orthogonal orientation) vertical grating couplers in array 230, light beam 755b passes through half wave plate 756 to rotate the polarization of the light in the half wave plate.The light beams 755a and 755b have the same polarization state through half wave plate 756 and are therefore properly tuned to use the parallel oriented vertical grating couplers in array 230.

In an alternative embodiment, light beam 755a (instead of light beam 755 b) may pass through half-wave plate 756 to rotate the polarization of the light in the half-wave plate. In various embodiments, half-wave plate 756 may be made of, for example, quartz crystals, polymer retardation films, or may be 3D printed. In some embodiments, the polarization diversity structure 757 may be fabricated using wafer-level optical processing and assembly.

In some embodiments, polarization diversity structure 757 may be inserted at other locations within connector element 240 or 250 in fiber-to-PIC array connector arrangement 700, for example, between lens 741 and lens 742 or between lens 551 and lens 541.

In some embodiments, some elements of polarization diversity structure 757 may be functionally separated and placed at different locations within connector elements 271 and 272. For example, birefringent beam shifting element 753 may be placed between fiber end face plane 243 and lens 551, and half wave plate 756 may be placed between lens 541 and lens 542.

Fig. 8B shows a schematic side view 820 of a sub-element of a polarization diversity assembly 757 according to an alternative embodiment. This particular embodiment uses a polarization-sensitive grating 853, such as the polarization-sensitive grating offered for sale by imagineooptix corporation of dallemm, north Carolina, USA, imagineOptix of Durham, north Carolina, USA, for splitting the incident light beam 754 into two circularly polarized light beams 855a and 855b, the polarization of which are mutually orthogonal. Light beams 855a and 855b are directed through optical layer 858, with optical layer 858 having a sufficient thickness to sufficiently laterally separate the light beams. The second polarization grating 859 is then used to diffract the laterally separated beams 855a and 855b such that the so diffracted beams become parallel to the original beam 754. The subsequent optical layer 860, including the quarter-wave polarization retarder element 861 and the three-quarter-wave polarization retarder element 862, then converts the polarization of the two light beams 855a and 855b into the same linear polarization state. In an example embodiment, this linear polarization state is the appropriate polarization state in a vertical grating coupler for achieving efficient optical coupling of linearly polarized light beams 755a and 755b to array 230.

The physical principle of the polarization diversity structure 757 can be explained as follows. Fig. 8C to 8E include schematic diagrams showing the splitting of an incident light beam 830 into two polarized light beams by a polarization diversity structure 757. For example, incident light beam 830 includes two orthogonal polarization components that are spatially split off crystal 832 into a first polarization component 834 and a second polarization component 836. Referring to fig. 8C, an incident light beam 830 may have a polarization component in any direction, and walk-off crystal 832 separates the polarization component into a first polarization component 834 and a second polarization component 836 whose polarizations are orthogonal to each other and separated by a distance d. As shown in fig. 8D and 8E, component 834, having a polarization direction perpendicular to the plane of the figure, passes straight off crystal 832, while component 836, having a polarization direction parallel to the plane of the figure, is shifted a distance D relative to component 834.

At least one of the two spatially separated polarization components 834, 836 is then rotated by half wave plate 838 such that the resulting polarization of the two spatial paths is the same. In the example of fig. 8D, the polarization of the second polarization component 836 is rotated 90 ° to generate the polarization component 840 such that the polarization components 834 and 840 have the same polarization. In the example of fig. 8E, the polarization of first polarization component 834 is rotated 90 ° to generate polarization component 842 such that polarization components 836 and 842 have the same polarization. The two spatially demultiplexed polarization components (834 and 840) or (836 and 842) are then incident on a polarization sensitive vertical grating coupler such as 231 in fig. 2 for coupling into a photonic integrated circuit (e.g., 210). In some examples, the grating coupler is responsive only to TE (transverse electric) polarization or TM (transverse magnetic) polarization. TE polarized light is characterized by its electric field perpendicular to the plane of incidence. For TE light, the magnetic field perpendicular to the electric field in the isotropic material lies in the plane of incidence.

In some embodiments, the fiber-to-PIC arrangement provides polarization management of two orthogonal polarizations between the fiber and polarization dependent PIC coupling elements, which may be vertical coupling elements or edge coupling elements responsive to, for example, circular polarization, linear polarization, or any other polarization state. The examples described below use photonic integrated circuit edge coupling elements, such as vertical grating couplers, that respond to a given linear polarization.

In the example of fig. 8A, birefringent beam shifting element 753 and half-wave plate 756 are both positioned between fiber end face plane 243 and connector mating plane 241. In some embodiments, birefringent beam shifting element 753 may be positioned between fiber end face plane 243 and connector mating plane 241, and half-wave plate 756 may be positioned between connector mating plane 241 and coupling plane 242.

Referring to fig. 9, in some embodiments, the fiber-to-PIC connector 900 includes a polarization diversity structure 902, the polarization diversity structure 902 including a walk-off element 904 and a spatially varying birefringent element 906. The walk-off element 904 is positioned between the collimating lens 551 and the connector mating plane 241. Spatially varying birefringent elements 906 are positioned between the individual lenses 542 and the coupling plane 242. For example, an incident beam 908 from core 302 passes through collimating lens 551 and is split into a pair of beams 755a and 755b by walk-off element 904. The polarization of light beam 755a is orthogonal to the polarization of light beam 755b. The light beam 755a passes through the spatially varying birefringent element 906, and the spatially varying birefringent element 906 rotates the polarization direction of the light beam 755a to be the same as the polarization direction of the light beam 755b. The two beams 755a and 755b are then coupled to photonic integrated circuit 210 by vertical coupling element 231.

In some embodiments, the spatially varying birefringent element 906 or beam shifting element 753 of fig. 7 and 8A may be replaced with a birefringent aperture plate (BHP), wherein the plate of birefringent material comprises an opening or aperture such that the light beam may pass through the aperture without any change in polarization direction while rotating the polarization direction of the light beam passing through the birefringent material by, for example, 90 °. The birefringent aperture plate may be used in combination with the walk-off element such that the light beam is split by the walk-off element into a first polarization component and a second polarization component, the first polarization component being directed through an aperture in the birefringent aperture plate and the second polarization component being directed through the birefringent material, thereby producing two polarized light beams having the same polarization direction.

Fig. 10A-10D illustrate various views of a fiber-to-PIC connector 1000 or portion thereof. Referring to fig. 10A, in some embodiments, a fiber-to-PIC connector 1000 is configured to optically couple a row of incident light beams 1008 from a row of input cores 302. The fiber-to-PIC connector 1000 includes a polarization diversity assembly 1002, the polarization diversity assembly 1002 including a walk-off crystal 1004 and a birefringent aperture plate 1006. Birefringent aperture plate 1006 has a birefringent material at location 1012 that acts as half-wave plate 756 of fig. 7 and 8A. The incident light beam 1008 is split by the walk-off crystal 1004 into two light beams 755a and 755b that initially have different polarization states (e.g., orthogonal polarization states). One of the two beams 755a and 755b is rotated by the birefringent material in the birefringent aperture plate 1006, after which the two beams 755a and 755b have the same polarization state.

Fig. 10B is a side view of the fiber-to-PIC connector 1000 viewed in direction a. An incident light beam 1008 from the core 302 passes through a collimating lens 551. Walk-off crystal 1004 splits an incident light beam 1008 into two beam components and birefringent aperture plate 1006 rotates the polarization direction of one of the beam components, thereby producing two light beams having the same polarization state. The light beam passes through the second lens 541 and the third lens 542 and is directed to the vertical coupling element 231 on the photonic integrated circuit 210.

Fig. 10C is a top view of an example of a polarization diversity assembly 1002, the polarization diversity assembly 1002 including a walk-off crystal 1004 and a birefringent aperture plate 1006 having an aperture 1020. Walk-off crystal 1004 causes polarization splitting, causing beam shifting in direction a (direction a is shown in fig. 10A), which in this example is parallel to the row direction. The incident light beam from the core 302 has a polarization component that is shifted from the initial position 1014 to the second position 1016 by the walk-off crystal 1004, wherein the shift direction is parallel to the row direction. The dashed line 1018a represents the positions of a set of the second lens 541 and the third lens 542. The broken line 1018b represents another set of the second lens 541 and the third lens 542. The centers of the second lens 541 and the third lens 542 are offset from the center of the polarized beam component, which causes the polarized beam component to be refracted by the lens and directed to the perpendicular grating coupler at an incident angle θ in the range between 0 and 90 °.

Fig. 10D is a top view of another example of a polarization diversity assembly 1002 in which polarization splitting causes a beam shift perpendicular to direction a (direction a is shown in fig. 10A). In this example, the beam shift direction is perpendicular to the row direction. The incident light beam from core 302 has a polarization component that is shifted from initial position 1014 to second position 1022 by walk-off crystal 1004, where the shift direction is perpendicular to the row direction. Each of the broken lines 1018a and 1018c represents a set of the second lens 541 and the third lens 542. The centers of the second lens 541 and the third lens 542 are offset from the center of the polarized beam component, which causes the polarized beam component to be refracted by the lens and directed to the perpendicular grating coupler at an incident angle θ in the range of 0 to 90 °.

Fig. 11A is a top view of an example of a fiber-to-PIC connector 1100 in which a row of four pairs of cores 302 are aligned with four pairs of apertures 1020 of a birefringent aperture plate 1006. Fig. 11B is a schematic diagram showing an example in which the walk-off crystal 1004 shifts the polarization component of each light beam in the direction 1022 parallel to the row direction. Fig. 11C is a schematic diagram showing an example in which the walk-off crystal 1004 shifts the polarization component of each light beam in a direction 1024 perpendicular to the row direction. The walk-off crystal may be designed to shift the polarization component of each beam in any predetermined direction.

In some embodiments, the birefringent plate may have regions of different thickness such that when the two polarization components provided by the walk-off element pass through two regions of different thickness, the resulting light beams have the same polarization direction. For example, if the half-wave plate has a thickness d1, the difference in thickness of the two regions may be equal to d1. The birefringent plate may have a plurality of pairs of regions, each pair of regions comprising a first region of thickness d1+d2 and a second region of thickness d 2. The two orthogonally polarized beam components from the walk-off element are directed to a pair of regions in which the polarization of one beam component is rotated 90 ° relative to the other beam component, thereby producing two polarized beams of the same polarization direction.

Fig. 12 is a schematic diagram of an example of a fiber-to-PIC connector 1000 in which a half-wave plate (e.g., 756) is implemented as a birefringent plate 1006 made of a birefringent material 1200, wherein the birefringent plate 1006 includes a region without the birefringent material (i.e., with holes 1020). The incident light beam 1008 from the input core 302 is split by the walk-off element 1004 into a first light beam component 1026 having an "x" polarization and a second light beam component 1028 having a "y" polarization. The figure shows an example where the "x" polarization is TE polarization and the "y" polarization is TM polarization. The first beam component 1026 passes through an aperture 1020 in the birefringent plate 1006 and retains its polarization. The second beam component 1028 passes through the birefringent material 1200 and its polarization changes from "y" polarization to "x" polarization. Then, the first beam component 1026 and the second beam component 1028 having the same or parallel polarization are incident on the grating couplers 231a and 231b, respectively. Grating couplers 231a and 231b couple first beam component 1026 and second beam component 1028 to photonic integrated circuit 210. Typically, the "aperture" plate 1006 has spatially varying birefringence configured to convert incident polarized light into a grating coupler polarization state, e.g., having a polarization state that maximizes the coupling efficiency of the grating coupler.

The fiber-to-PIC connector 1000 may combine two output optical signals from the photonic integrated circuit 210 into an output optical beam that is transmitted on an output fiber core. For example, photonic integrated circuit 210 outputs two optical signals that are converted into optical beams 1030 and 1032 by grating couplers 231c and 231d, respectively, wherein optical beams 1030 and 1032 have the same polarization state. Beam 1030 passes through third lens 542 and second lens 541 and then through aperture 1020 in birefringent aperture plate 1006. Beam 1030 passes straight through walk-off element 1004 without changing direction. The light beam 1032 passes through the third lens 542 and the second lens 541, and then passes through the birefringent material 1200 in the birefringent aperture plate 1006, the birefringent aperture plate 1006 rotates the polarization direction of the light beam 1032 by 90 °. Beam 1032 is displaced a distance from element 1004 and combined with beam 1030. The combined beam passes through a collimating lens 551 and is directed to an output core 1034.

Fig. 13A to 13D are schematic diagrams showing the relationship among the optical fiber connector, the birefringent aperture plate, and the grating coupler. Fig. 13A is a schematic diagram of a fiber-to-PIC connector 1000 that is identical to the fiber-to-PIC connector shown in fig. 10A. Fig. 13B is a schematic diagram of a fiber optic connector 1300, the fiber optic connector 1300 including a transmitter fiber port (e.g., 1302), a receiver fiber port (e.g., 1304), and an optical energy fiber port (e.g., 1306). The circles show the input fiber positions (e.g., 302). This example includes 3 rows of 12 fibers. For example, the rows may be spaced 500 μm apart, and the fibers within a row may be spaced 250 μm apart. In this example, an orange circle (e.g., 1302) represents a Transmitter (TX) fiber port, a brown circle (e.g., 1304) represents a Receiver (RX) fiber port, and a red circle (e.g., 1306) represents an optical energy fiber port. Additional information regarding fiber optic connectors is provided in U.S. provisional patent application 63/145,368 filed 2/3 2021, the entire contents of which are incorporated by reference.

Fig. 13C is a top view of an example of a birefringent aperture plate 1006 having apertures 1020. Aperture 1020 is located at the position of beam 755 a. The walk-off direction is indicated by arrow 1308.

Fig. 13D is a top view of an example of an array of six rows of twelve grating couplers 1310 (231 in fig. 2, 5-7) mounted on top of a photonic integrated circuit. Each grating coupler 1310 is a TE coupler, where the direction of the electric field is represented by arrow 1312. Walk-off crystal 1004 splits each input beam into two beam components having orthogonal polarization states, and birefringent aperture plate 1006 rotates the polarization direction of one of the two beam components so that the two beam components have the same polarization state upon reaching grating coupler 1310. Orange triangles (e.g., 1316) represent grating couplers for Transmit (TX) signals output through Transmitter (TX) fiber port 1302. If the transmit signal has a single polarization, only one grating coupler is required for the corresponding Transmitter (TX) fiber port 1302. Brown triangles (e.g., 1318) represent a grating coupler for Receiving (RX) signals. Due to the random polarization of the input signal, two grating couplers 1310 are required for each corresponding Receiver (RX) fiber port 1304. The red triangle (e.g., 1320) represents a grating coupler corresponding to the optical energy fiber port 1306. White triangle 1314 shows: (i) grating couplers are not present at those locations; (ii) Grating couplers exist at those locations but are not coupled to receive or transmit optical signals; or (iii) a grating coupler is connected to the alignment waveguide to assist in alignment calibration.

Fig. 14A-18B illustrate examples of grating coupler orientations that may be used with the same fiber optic array shown in fig. 13B. In each of these examples, the grating couplers are of the same type, e.g., all TE grating couplers or all TM grating couplers, and the grating couplers are aligned in the same direction. Fig. 14A to 18B show examples of TE grating couplers. The same principle can be applied to a TM grating coupler with a suitably tuned birefringent aperture plate.

Fig. 14A and 14B show an example of an arrangement of a grating coupler 1400 and a corresponding birefringent aperture plate 1402, the birefringent aperture plate 1402 comprising an aperture 1404 at a predetermined position. This example assumes that the walk-off element outputs a beam with an electric field in direction 1406. Each input beam is split by the walk-off element into a first beam component and a second beam component. The first beam component passes straight through the walk-off element and the second beam component is displaced a distance from the first beam component in the walk-off direction 1408. The first beam component has an electric field in a direction 1406 aligned with the grating coupler. The aperture 1404 is positioned to allow the first beam component to pass through without affecting the polarization state. In this example, grating coupler 1400 is oriented to maximize coupling of the light beam with the electric field in direction 1406.

Fig. 15A and 15B illustrate an example of an arrangement of a grating coupler 1500 and a corresponding birefringent aperture plate 1502, the birefringent aperture plate 1502 comprising an aperture 1504 at a predetermined position. This example assumes that the walk-off element outputs a beam with an electric field perpendicular to direction 1506. Each input beam is split by the walk-off element into a first beam component and a second beam component. The first beam component passes straight through the walk-off element and the second beam component is displaced a distance from the first beam component along walk-off direction 1508. The second beam component has an electric field in a direction 1506 aligned with the grating coupler. The aperture 1504 is positioned to allow the second beam component to pass through without affecting the polarization state. In this example, grating coupler 1500 is oriented to maximize coupling of the light beam with the electric field in direction 1506.

Fig. 16A and 16B illustrate an example of an arrangement of a grating coupler 1600 and a corresponding birefringent aperture plate 1602, the birefringent aperture plate 1602 including an aperture 1604 at a predetermined position. This example assumes that the walk-off element outputs a beam with an electric field perpendicular to direction 1606. Each input beam is split by the walk-off element into a first beam component and a second beam component. The first beam component passes straight through the walk-off element and the second beam component is shifted a distance from the first beam component in the walk-off direction 1608. The second beam component has an electric field in a direction 1606 aligned with the grating coupler. The aperture 1604 is positioned to allow the second beam component to pass through without affecting the polarization state. In this example, grating coupler 1600 is oriented to maximize coupling of the light beam with the electric field in direction 1606.

Fig. 17A and 17B illustrate an example of an arrangement of a grating coupler 1700 and a corresponding birefringent aperture plate 1702, the birefringent aperture plate 1702 comprising an aperture 1704 at a predetermined position. This example assumes that the walk-off element outputs a beam with an electric field parallel to direction 1706. Each input beam is split by the walk-off element into a first beam component and a second beam component. The first beam component passes straight through the walk-off element and the second beam component is shifted a distance from the first beam component along walk-off direction 1708. The first beam component has an electric field in a direction 1706 aligned with the grating coupler. The aperture 1704 is positioned to allow the first beam component to pass through without affecting the polarization state. In this example, grating coupler 1700 is oriented to maximize coupling of the light beam with an electric field in direction 1706.

Fig. 18A and 18B illustrate an example of an arrangement of a grating coupler 1800 and a corresponding birefringent aperture plate 1802, the birefringent aperture plate 1802 including an aperture 1804 at a predetermined position. This example assumes that the walk-off element outputs a beam with an electric field perpendicular to direction 1806. Each input beam is split by the walk-off element into a first beam component and a second beam component. The first beam component passes straight through the walk-off element and the second beam component is shifted a distance from the first beam component along walk-off direction 1808. The second beam component has an electric field in a direction 1806 aligned with the grating coupler. The aperture 1804 is positioned to allow the second beam component to pass through without affecting the polarization state. In this example, grating coupler 1800 is oriented to maximize coupling of the beam with the electric field in direction 1806.

Fig. 13A to 13D, 14A, 14B, 15A, 15B, 16A, 16B, 17A, 17B, 18A, and 18B illustrate examples of various orientations of grating couplers and corresponding birefringent aperture plates. The orientation of the grating coupler may depend on, for example, the optical waveguide routing layout. The grating coupler may have an orientation different from the examples described above. For example, the orientation of the grating coupler may be selected based on a desired optical waveguide layout, and then the walk-off element oriented such that the beam component output from the walk-off element is parallel or orthogonal to the direction of the electric field, where the grating coupler has the greatest coupling efficiency. The birefringent aperture plate is designed such that the aperture is positioned where the beam component does not need to be rotated in the polarization direction to achieve maximum coupling efficiency of the grating coupler.

Fig. 19A-19C and 20A-20C illustrate examples of grating coupler orientations that may be used with the same fiber array shown in fig. 13B. Fig. 19D is an enlarged view of fig. 19A, and fig. 20D is an enlarged view of fig. 20A. In fig. 19D and 20D, reference numerals "P", "R", and "T" denote an optical energy fiber port, a receiver fiber port, and a transmitter fiber port, respectively. Fig. 19E is an enlarged view of fig. 19C, and fig. 20E is an enlarged view of fig. 20C. In fig. 19E and 20E, reference numerals "P", "R", and "T" denote an optical energy grating coupler for coupling optical energy light, a receiver grating coupler for coupling in or receiving an optical signal, and an emitter grating coupler for coupling out or emitting an optical signal, respectively. In these examples, some grating couplers are positioned at locations aligned between the core locations to achieve higher densities. The grating couplers are of the same type, e.g., TE grating couplers or TM grating couplers, and the grating couplers are aligned in the same direction. Fig. 19A to 20C show examples of TE grating couplers. The same principle can be applied to a TM grating coupler with a suitably tuned birefringent aperture plate.

Fig. 19A to 19C are schematic diagrams of examples of an arrangement of the optical fiber port 1900, a birefringent aperture plate 1902 having an aperture 1910 at a predetermined position, and an arrangement of the grating coupler 1904. This example assumes that the walk-off element outputs a beam with an electric field perpendicular to direction 1906. Each input beam is split by the walk-off element into a first beam component and a second beam component. The first beam component passes straight through the walk-off element and the second beam component is shifted a distance from the first beam component along walk-off direction 1908. The second beam component has an electric field in direction 1906. In this example, grating coupler 1904 is oriented to maximize coupling of the light beam with the electric field in direction 1906. Because the polarization of the second beam component is already aligned with the grating coupler, aperture 1910 is positioned to allow the second beam component to pass through without changing the polarization state. Fig. 19A shows an arrangement of three rows of 12 fiber ports 1900. Walk-off direction 1908 is parallel to the row direction. Some of the grating couplers are positioned at locations aligned between the locations of the fiber ports 1900 to achieve higher densities. For example, the distance between two adjacent grating couplers 1904 may be about half the distance between two adjacent cores in a row.

Fig. 20A to 20C are schematic diagrams of examples of an arrangement of the optical fiber ports 2000, a birefringent aperture plate 2002 having an aperture 2010 at a predetermined position, and an arrangement of the grating couplers 2004. This example assumes that the walk-off element outputs a beam with an electric field parallel to direction 2006. Each input beam is split by the walk-off element into a first beam component and a second beam component. The first beam component passes straight through the walk-off element and the second beam component is displaced a distance from the first beam component in the walk-off direction 2008. The first beam component has an electric field in direction 2006. In this example, grating coupler 2004 is oriented to maximize coupling of the light beam with the electric field in direction 2006. Because the polarization of the first beam component is already aligned with the grating coupler, aperture 2010 is positioned to allow the first beam component to pass through without changing the polarization state. Fig. 20A shows an arrangement of three rows of 12 fiber ports. The walk-off direction 2008 is at an angle (e.g., 45 °) relative to the row direction. Some of the grating couplers are positioned at locations aligned between the locations of the fiber ports 2000 to achieve higher densities. In the example of fig. 20A, the distance between adjacent cores in a row is the same as the distance between adjacent cores in a column. For example, the distance between two adjacent grating couplers 2004 may be about 70% of the distance between two adjacent cores in a row.

Referring to fig. 21A to 21D, the birefringent aperture plate may have apertures of one or more of various shapes. Fig. 21A is a schematic diagram of an example of a birefringent aperture plate 2100 having an aperture 2102 in a circular shape. Fig. 21B is a schematic diagram of an example of a birefringent aperture plate 2104 having an aperture 2106 having a square shape. Fig. 21C is a schematic diagram of an example of a birefringent aperture plate 2108 with a strip-shaped aperture 2110. In some examples, the aperture may have a rectangular shape. Fig. 21D is a schematic diagram of an example of a birefringent aperture plate 2116 comprising a plurality of individual strip-like sheets 2112 spaced apart, wherein the spaces 2114 between the strip-like sheets 2112 form the "holes" of the birefringent aperture plate 2116. In some embodiments, the apertures of the birefringent aperture plate may have a combination of different shapes, and these shapes may have any geometry.

In general, a micro-optical system may include a plurality of optical elements, including an optical birefringent element for changing polarization of light. In some applications, the patterned birefringent element may have cross-regional non-uniform birefringence to alter polarization modifications induced by the element. One example is a Half Wave Plate (HWP) in which an array of through holes is drilled through the elements such that the polarization of light passing through the through holes is unchanged, while light passing through the half wave plate material may undergo a polarization change.

In some embodiments, the patterned birefringent plate may be generated by incorporating the birefringent element into a second element in a micro-optical system, wherein the pattern is generated in the birefringent element after incorporation by a process such as etching, mechanical removal, laser cutting, laser ablation, or the like. The patterned birefringent plate has a non-uniform birefringence in a plane through which the light passes such that different portions of the light pass through different portions of the patterned birefringent plate having different birefringence. The patterned birefringent plate may affect the optical properties of different parts of the light differently. Thus, for example, the patterned birefringent plate may rotate the polarization of the first partial light by a first amount and the polarization of the second partial light by a second amount.

Another method of creating patterned birefringent plates with patterns in the birefringent material is to change the birefringence of the birefringent element without removing material or with minimal material. One example is the use of localized laser heating to create a pattern of altered birefringence in a material such as crystalline quartz. Heating crystalline quartz close to its melting point may damage the crystalline structure, resulting in fused silica of amorphous structure without birefringence. Patterned localized heating to create fused silica regions within the crystalline quartz can have an effect similar to the formation of holes or strips in the birefringent aperture plates shown in fig. 21A-21D. In some examples, one or more particle beams, such as electron beams, may excite the crystalline quartz to change its birefringence.

Fig. 50 shows a top view 5000 and a side view 5002 of an example of a patterned birefringent plate 5004 in which patterned birefringent elements are attached to second elements. Patterned birefringent plate 5004 has non-uniform birefringence across a plane parallel to the top surface of patterned birefringent plate 5004 and exhibits non-uniform birefringence properties with respect to light passing through patterned birefringent plate 5004 by entering top side 5006 (or bottom side 5008) and exiting bottom side 5008 (or top side 5006) of patterned birefringent plate 5004.

In some implementations, the optical birefringent element 5010 is bonded to the second optical element 5012 and a pattern is formed in the optical birefringent element 5010. In some implementations, the pattern is first formed in the optical birefringent element 5010, and then the patterned optical birefringent element 5010 is bonded to the second optical element 5012. In the example shown in fig. 50, the patterned birefringent element 5010 can be generated by removing the birefringent material. Such removal may be accomplished by a variety of methods including, for example, mechanical removal, laser ablation, and etching. The etching may be, for example, liquid phase etching (also referred to as wet chemical etching) or plasma phase etching (also referred to as dry etching). The removal process may be controlled, for example, by using an etch stop layer at the interface between the birefringent material 5010 and the second element 5012 to prevent removal of material from the second optical element 5012. For example, the etch stop layer may be an anti-reflective coating that enhances light transmission between the birefringent material 5010 and the second optical element 5012, wherein the anti-reflective coating is also etch resistant.

In some embodiments, using a chemical etching process to remove the birefringent material instead of using mechanical drilling to create openings in the birefringent material has the following advantages: the resulting patterned birefringent element 5010 is relatively clean without the large amounts of debris that can occur when mechanical drilling is used to create the openings. This beneficial feature is significant when the pattern in the birefringent element 5010 has a small size. In the example shown in fig. 50, the patterned optical birefringent element 5010 includes a plurality of strips 5014 of birefringent material extending parallel to each other. In the areas between the strips 5014 of birefringent material, the birefringent material has been etched away.

In some implementations, the width d1 of each birefringent material strip 5014 is substantially equal to the width d2 of the etched areas 5016 between the birefringent material strips 5014. The widths d1 and d2 are selected to be slightly larger than the diameter of the beam passing through the birefringent material strip 5014 and the etched area 5016. In some examples, width d1 may also be different than width d2.

For example, a light beam from a fiber core may be split into a first light beam component 5018 (e.g., similar to 834 or 2308) and a second light beam component 5020 (e.g., similar to 836 or 2310) by a birefringent light beam shifting element, walk-off element, polarization sensitive grating (e.g., 753 of fig. 7 and 8A, 853 of fig. 8B, 832 of fig. 8C to 8E, 904 of fig. 9, 1004 of fig. 10A to 10C, 12 and 13A, 2306 of fig. 23, 2408 of fig. 24A, 2906 of fig. 29, 3008 of fig. 30, 3400 of fig. 34A and 34B, 3512 of fig. 35, 3908 of fig. 39, 4002 of fig. 40), wherein the first light beam component 5018 and the second light beam component 5020 are separated in the walk-off direction 5022. The walk-off element may be designed such that after exiting the walk-off element, the first beam component 5018 has a polarization orthogonal to the polarization of the second beam component 5020. The walk-off element directs the first beam component 5018 to the etched area 5016 and the second beam component 5020 to the birefringent material strip 5014. The birefringent material strip 5014 changes the polarization of the second beam component 5020, while the first beam component 5018 passes through the etched area 5016 without any polarization change. The thickness of the patterned optical birefringent element 5010 may be selected such that the patterned optical birefringent element 5010 is functionally equivalent to a half wave plate with openings at the etched areas 5016. In this example, the polarization of the second beam component 5020 is rotated 90 +n 180, with 0 n being an integer. After passing through the patterned optical birefringent element 5010, the polarization of the first beam component 5018 becomes parallel to the polarization of the second beam component 5020.

The birefringent beam shifting element, walk-off element or polarization sensitive grating may also be used as a beam combiner to combine the beam components emitted from the vertical coupling element and passing through the patterned optical birefringent element 5010 into one beam that is transmitted to the corresponding fiber core.

In the example of fig. 50, the birefringent material is completely etched away in the etched region 5016. In some implementations, the birefringent material is partially etched such that the polarization of the first beam component 5018 is also rotated by the remaining birefringent material in the etched region 5016. The unetched birefringent material strip 5014 has a greater thickness and the polarization of the second beam component 5020 is rotated a greater amount. The depth of the etch may be selected such that the polarization of the second beam component 5020 is rotated 90 deg. + n x 180 deg., 0 n.ltoreq.n, with respect to the polarization of the first beam component 5018, n being an integer.

In some embodiments, the diameter of the light beam from the fiber optic core upon reaching the patterned optical birefringent element 5010 may be in the range of 49 μm to 999 μm, and the width d1 of each birefringent material strip 5014 and the width d2 of the etched area 5016 may be in the range of 50 μm to 1000 μm. In some examples, the diameter of the light beam from the fiber core upon reaching the patterned optical birefringent element 5010 may be in the range of 99 μm to 599 μm, and the width d1 of each birefringent material strip 5014 and the width d2 of the etched area 5016 may be in the range of 100 μm to 600 μm. In some examples, the diameter of the light beam from the fiber core upon reaching the patterned optical birefringent element 5010 may be in the range 199 μm to 399 μm, and the width d1 of each birefringent material strip 5014 and the width d2 of the etched area 5016 may be in the range 200 μm to 400 μm.

The arrangement of the parallel strips of birefringent material 5014 and etched areas 5016 corresponds to the arrangement of the optical fibers coupled to the fiber optic connector. For example, the optical fibers may be arranged in a two-dimensional array having at least 2 rows and at least 8 columns. The etched region may comprise at least 2 parallel strips of etched region. The patterned optical birefringent element 5010 shown in fig. 50 comprises 3 parallel strips 5014 of birefringent material and can be used in fiber optic connectors that couple an array of 3 rows of optical fibers to a corresponding array of vertical coupling elements on a photonic integrated circuit.

In some embodiments, the etched region may have a shape substantially similar to a circle, an ellipse, a triangle, a square, a rectangle, or a polygon having n sides, n being an integer greater than 4, and the shape measured along a plane parallel to the top surface of the second optical element 5012. In this example, the size of the etched area is selected to be slightly larger than the size of the beam. For example, each of the etched regions may have a size in a range of 50 μm to 1000 μm measured in a direction parallel to the walk-off direction. For example, each of the etched regions may have a size in a range of 100 μm to 600 μm measured in a direction parallel to the walk-off direction. For example, each of the etched regions may have a size in a range of 20 μm to 400 μm measured in a direction parallel to the walk-off direction.

The arrangement of etched areas corresponds to the arrangement of optical fibers coupled to the fiber optic connector. For example, the optical fibers may be arranged in a two-dimensional array having at least 2 rows and at least 8 columns. In this example, the etched areas may also be arranged in a two-dimensional array having at least 2 rows and at least 8 columns.

The dimensions of the etched areas or strips of birefringent material are provided as examples only. It should be appreciated that the etched areas and strips of birefringent material may have larger dimensions, for example, when larger fiber cores are used, when the fiber cores are spaced farther apart from each other, and/or when the vertical coupling elements are spaced farther apart from each other.

The half-wave plates or birefringent aperture plates of fig. 7, 9, 10A, 11B, 11C, 12, 13A, 13C, 14B, 15B, 16B, 17B, 18B, 19B, 20B, 21A-21D, 23, 28, 29, 30, 35, 36, 38-40, and 49 may be replaced with patterned birefringent plates, wherein some of the birefringent material is removed by etching, similar to patterned birefringent plate 5004 of fig. 50. For example, manufacturing the fiber-to-PIC connector 700 includes manufacturing the connector element 250. Manufacturing the connector element 250 includes manufacturing a patterned birefringent plate and attaching the birefringent plate to another optical element, for example, an optical element including a collimating lens array 551. Manufacturing the fiber-to-PIC connector 900 includes manufacturing the connector element 240. Manufacturing the connector element 240 includes manufacturing a patterned birefringent plate and attaching the birefringent plate to another optical element, for example, an optical element including the second lens array 541 and the third lens array 542.

The fiber-to-PIC connectors (e.g., 700, 900) will also be referred to as fiber connectors, and the connector elements (e.g., 240, 250) will also be referred to as fiber connector components. Thus, in some embodiments, fabricating the fiber optic connector includes generating a patterned birefringent plate by attaching a birefringent element to a second optical element, and applying a removal process (e.g., an etching process) to remove portions of the optical birefringent material at the plurality of first regions such that after the removal process, the plurality of first regions have no optical birefringent material or have reduced thickness of the optical birefringent material. Manufacturing the fiber optic connector further includes attaching the patterned birefringent plate to another connector component (e.g., a connector component including the collimating lens array 551 or the second and third lens arrays 541, 542).

Fig. 51 shows a top view 5100 and a side view 5102 of an example of a patterned birefringent plate 5104 in which a patterned optical birefringent element 5110 is attached to a second optical element 5112. The birefringent plate 5104 has non-uniform birefringence across a plane parallel to the top surface of the birefringent plate 5104, and exhibits non-uniform birefringence properties with respect to light passing through the birefringent plate 5104 by entering the top side 5106 (or bottom side 5108) and exiting the bottom side 5108 (or top side 5106) of the birefringent plate 5104.

In some embodiments, the optical birefringent element 5110 without a pattern is bonded to the second optical element 5112 and after bonding to the second optical element 5112, a pattern is generated in the optical birefringent element 5110. In some embodiments, a pattern is formed in the optical birefringent element 5110 to create a patterned optical birefringent element 5110 and the patterned optical birefringent element 5110 is bonded to the second optical element 5112. The pattern is selected for a particular application in which the pattern may include the degree and orientation of optical birefringence that varies across the patterned optical birefringent element 5110.

In some embodiments, the birefringent pattern may be generated by locally heating the birefringent material to change the material properties, resulting in a changed birefringence, reduced birefringence, or substantially no birefringence. No material is removed during the localized heating, or only a small amount of material is removed during the localized heating. For example, localized laser heating may be applied to crystalline quartz (which has birefringent properties) to produce amorphous fused silica regions with altered birefringence, low birefringence, or substantially no birefringence. In some examples, laser light is applied to each region in turn to create regions of lower birefringence in turn. In some examples, multiple laser beams are applied in parallel to respective regions (e.g., by using multiple laser sources or splitting each of one or more laser beams into multiple beams) to produce regions of lower birefringence in parallel.

A material is considered to have substantially no birefringence if it rotates the polarization of light, for example, by less than 5 ° or, in some examples, by less than 1 °. For example, an inverse process may also be used in which localized heating of the non-birefringent material may generate birefringent regions.

In the example of fig. 51, localized heating is applied to the first region 5112 (in the shape of parallel strips) of the optical birefringent element 5110 such that the first region 5112 has reduced or substantially no birefringence. After localized heating, the optical birefringent element 5110 has a stripe pattern comprising first regions 5112 alternating with second regions 5114, wherein the first regions 5112 comprise parallel strips of a birefringent material with low or no birefringence and the second regions 5114 comprise parallel strips of a birefringent material that retains its birefringence.

In some embodiments, using localized heating to reduce birefringence of a material rather than using mechanical drilling to create openings in a birefringent material has the following advantages: the patterned birefringent element thus produced is relatively clean without the large amounts of debris that can occur when mechanical drilling is used to create the openings. This beneficial feature is significant when the pattern in the birefringent element 5110 has a small size. In the example shown in fig. 51, the patterned optical birefringent element 5110 includes a plurality of strips 5114 of birefringent material extending parallel to each other. In the areas 5112 between the strips 5114 of birefringent material, localized heating has been applied to reduce the birefringence, e.g., substantially no birefringence.

In some embodiments, the width d3 of each strip of material with reduced or no birefringence in the first region 5112 is substantially equal to the width d4 of the strip of birefringent material in the second region 5114. The widths d3 and d4 are selected to be slightly larger than the diameter of the light beam passing through the strip 5112 in the first region and the strip 5114 in the second region. In some examples, width d3 is different than width d4.

For example, a light beam from a fiber core may be split into a first light beam component 5118 (e.g., similar to 834 or 2308), and a second light beam component 5120 (e.g., similar to 836 or 2310) by a birefringent light beam shifting element, walk-off element, or polarization sensitive grating (e.g., 753 of fig. 7 and 8A, 853 of fig. 8B, 832 of fig. 8C to 8E, 904 of fig. 9, 1004 of fig. 10A to 10C, 12 and 13A, 2306 of fig. 23, 2408 of fig. 24A, 2906 of fig. 29, 3008 of fig. 30, 3400 of fig. 34A and 34B, 3512 of fig. 35, 3908 of fig. 39, 4002 of fig. 40), wherein the first light beam component 5118 and the second light beam component 5120 are separated in a walk-off direction 5122. The walk-off element may be designed such that after exiting the walk-off element, the first beam component 5118 has a polarization orthogonal to the polarization of the second beam component 5120. The walk-off element directs the first beam component 5118 toward a first region 5112 having reduced or substantially no birefringence and directs the second beam component 5120 toward a second region 5114 which retains its birefringent properties. The birefringent material in the second region 5114 causes the polarization of the second beam component 5120 to change by a first amount, while the first beam component 5018 passes through the first region 5112 without any polarization change (if the first region 5112 has substantially no birefringence) or with a smaller amount of polarization change (if the first region 5112 has reduced birefringence). The thickness of the patterned optical birefringent element 5110 and the thickness of the first region 5112 subjected to localized heating may be selected such that the patterned optical birefringent element 5110 is functionally equivalent to a half-wave plate having an opening at the first region 5110. In this example, the polarization of second beam component 5120 is rotated 90 +n×180° relative to the polarization of first beam component 5118, 0 n being an integer. After passing through the patterned optical birefringent element 5110, the polarization of the first beam component 5118 becomes parallel to the polarization of the second beam component 5120.

The birefringent beam shifting element, walk-off element or polarization sensitive grating may also be used as a beam combiner to combine the beam components emitted from the vertical coupling element and passing through the patterned optical birefringent element 5110 into one beam that is transmitted to the corresponding fiber core.

In some embodiments, the light beam from the fiber core may be in the range of 49 μm to 999 μm upon reaching the patterned optical birefringent element 5110, and the width d4 of each birefringent material strip in the second zone 5114 and the width d3 of the material strip with reduced or substantially no birefringence in the first zone 5112 may be in the range of 50 μm to 1000 μm. In some examples, the light beam from the fiber core may be in the range of 99 μm to 599 μm upon reaching the patterned optical birefringent element 5110, and the width d4 of each birefringent material strip in the second zone 5114 and the width d3 of the material strip with reduced or substantially no birefringence in the first zone 5112 may be in the range of 100 μm to 600 μm. In some examples, the light beam from the fiber core may be in the range of 199 μm to 399 μm upon reaching the patterned optical birefringent element 5010, and the width d4 of each birefringent material strip in the second zone 5114 and the width d3 of the material strip with reduced or substantially no birefringence in the first zone 5112 may be in the range of 200 μm to 400 μm.

The arrangement of parallel strips of birefringent material in the second zone 5114 and parallel strips of material having reduced or substantially no birefringence in the first zone 5112 corresponds to the arrangement of optical fibers coupled to the optical fiber connector. For example, the optical fibers may be arranged in a two-dimensional array having at least 2 rows and at least 8 columns. In this example, the locally heated region (i.e., the first region 5112) can include at least 2 parallel strips of material having reduced or substantially no birefringence. The patterned optical birefringent element 5110 shown in fig. 51 comprises 3 parallel strips of material with reduced or substantially no birefringence and may be used in an optical fiber connector that couples an array of 3 rows of optical fibers to a corresponding array of vertical coupling elements on a photonic integrated circuit.

In some embodiments, each of the plurality of localized heating regions may have a shape substantially similar to a circle, an ellipse, a triangle, a square, a rectangle, or a polygon having n sides, n being an integer greater than 4, and the shape being measured along a plane parallel to the top surface of the second optical element 5112. In this example, the size of the local heating region is selected to be slightly larger than the size of the beam. For example, each of the local heating regions may have a size in the range of 50 μm to 1000 μm measured in a direction parallel to the walk-off direction. For example, each of the local heating regions may have a size in the range of 100 μm to 600 μm measured in a direction parallel to the walk-off direction. For example, each of the local heating regions may have a size in the range of 200 μm to 400 μm measured in a direction parallel to the walk-off direction.

The arrangement of the localized heating regions corresponds to the arrangement of optical fibers coupled to the fiber optic connector. For example, the optical fibers may be arranged in a two-dimensional array having at least 2 rows and at least 8 columns. In this example, the localized heating regions may also be arranged in a two-dimensional array having at least 2 rows and at least 8 columns.

The dimensions of the locally heated region or stripe that produces reduced or substantially no birefringence are provided as examples only. It should be appreciated that the etched areas and strips of birefringent material may have larger dimensions, for example, when larger fiber cores are used, when the fiber cores are spaced farther apart from each other, and/or when the vertical coupling elements are spaced farther apart from each other.

The half-wave plates or birefringent aperture plates of fig. 7, 9, 10A, 11B, 11C, 12, 13A, 13C, 14B, 15B, 16B, 17B, 18B, 19B, 20B, 21A-21D, 23, 28, 29, 30, 35, 36, 38-40 and 49 may be replaced by patterned birefringent plates, wherein localized heating is applied to some areas of the birefringent material to reduce birefringence, similar to patterned birefringent plate 5104 of fig. 51. For example, manufacturing the fiber-to-PIC connector 700 includes manufacturing the connector element 250. Manufacturing the connector element 250 includes manufacturing a patterned birefringent plate and attaching the birefringent plate to another optical element, for example, an optical element including a collimating lens array 551. For example, manufacturing the fiber-to-PIC connector 900 includes manufacturing the connector element 240. Manufacturing the connector element 240 includes manufacturing a patterned birefringent plate and attaching the birefringent plate to another optical element, for example, an optical element including the second lens array 541 and the third lens array 542.

Thus, in some embodiments, fabricating the fiber optic connector includes applying localized heating to the birefringent plate to create a patterned birefringent plate, wherein the localized heating alters the birefringence of a plurality of first regions in the birefringent plate such that the first regions have a different birefringence than the second regions that do not receive the localized heating. Manufacturing the fiber optic connector further includes attaching the patterned birefringent plate to another connector component (e.g., a connector component including the collimating lens array 551 or the second and third lens arrays 541, 542).

Referring to fig. 52A-52D, in some embodiments, a pattern of altered birefringence, reduced birefringence, or substantially zero birefringence may be generated within the volume of the birefringent element. The opposite is possible, wherein a birefringent pattern is generated within the volume of the non-birefringent element.

Fig. 52A is a perspective view of an example of a patterned birefringent plate 5200 that includes a stripe pattern of altered birefringence. A strip of material 5202 with reduced or no birefringence is created within the volume of the patterned birefringent plate 5200.

In some embodiments, a varying birefringence pattern may be generated throughout the thickness of the birefringent element. Fig. 52B is a side view of an example of a patterned birefringent plate 5204 in which a strip of material 5206 with reduced or no birefringence extends from a top surface 5208 to a bottom surface 5210 of the patterned birefringent plate 5204.

In some embodiments, a modified birefringence pattern may be generated within the volume of the birefringent element. Fig. 52C is a side view of an example of patterned birefringent plate 5212 in which a strip of material 5214 with reduced or no birefringence is positioned within the volume of patterned birefringent plate 5212 at a first distance from the top surface 5208 and a second distance from the bottom surface 5210 of patterned birefringent plate 5212.

In some embodiments, a modified birefringence pattern may be generated near the surface of the birefringent element. Fig. 52D is a side view of an example of patterned birefringent plate 5216 in which a strip of material 5218 with reduced or no birefringence extends from a top surface 5208 to a position a distance from a bottom surface 5210 of patterned birefringent plate 5216.

The different types of patterns shown in fig. 52B to 52D can be generated by focusing laser beams at different depths in the volume of the birefringent plate to locally heat the birefringent material at different depths. In these examples, the localized heating regions (e.g., 5206, 5214, 5218) are integral parts of the patterned birefringent plate, and no glue or adhesive is used to bond the localized heating regions to other parts of the patterned birefringent plate.

Fig. 53A and 53B are schematic diagrams showing examples of a birefringent plate having a pattern generated in a birefringent element. Fig. 53A shows an example of a birefringent plate 5300 with parallel strips 5302 patterned in the birefringent element. Fig. 53B shows an example of a birefringent plate 5304 with a circular array 5306 patterned in the birefringent element.

Fig. 53C is a schematic diagram showing incident light 5308 passing through patterned birefringent plate 5310 including first region 5312 comprising a pattern of material having zero birefringence. For example, the incident light 5308 may be a single wavefront or beamlet pattern. The polarization of the light 5314 remains unchanged when passing through the non-birefringent pattern 5312. When passing through the birefringent material of the birefringent plate 5310, the polarization of the light 5316 changes.

In some embodiments, the process of manufacturing a fiber optic connector component (e.g., connector element 250 of fig. 7 or connector element 240 of fig. 9) includes applying localized heating or localized excitation to a birefringent plate (e.g., 5104) to alter the birefringence of a two-dimensional pattern of first regions (e.g., 5112) in the birefringent plate such that the first regions have a birefringence that is different from the birefringence of second regions (e.g., 5114) that do not receive localized heating or localized excitation. The process includes coupling a birefringent plate to a second optical component (e.g., 5112) to form a fiber optic connector component configured to be coupled to at least one of a plurality of optical fibers or a plurality of vertical coupling elements on a photonic integrated circuit, and the birefringent plate includes non-uniform birefringent properties with respect to light passing through the birefringent plate.

For example, the birefringent plate may be configured to change the polarization state of the first set of light beams (e.g., 5118) passing through the first region of the birefringent plate relative to the polarization state of the second set of light beams (e.g., 5120) passing through the second region of the birefringent plate.

For example, localized heating or localized excitation may be applied to the birefringent plate to alter the birefringence of a two-dimensional array of first regions in the birefringent plate such that the array of first regions has a different birefringence than the birefringence of second regions that do not receive localized heating, and the two-dimensional array comprises at least 2 rows and at least 8 columns.

For example, the spacing between two adjacent rows in a two-dimensional array may be the same as the spacing between two adjacent columns in a two-dimensional array.

For example, the fiber optic connector components may be configured to optically couple to a two-dimensional array of optical fibers.

For example, the fiber optic connector component may be configured to optically couple to a two-dimensional array of vertical coupling elements on a photonic integrated circuit.

For example, the fiber optic connector assembly may be configured to enable transmission of the first set of light beams and the second set of light beams between a two-dimensional array of optical fibers and a two-dimensional array of vertical coupling elements. The birefringent plate may be configured to change the polarization state of the first set of light beams passing through the first region of the birefringent plate relative to the polarization state of the second set of light beams passing through the second region of the birefringent plate.

For example, each of at least some of the first regions may have a substantially circular shape, an elliptical shape, a triangular shape, a square shape, or a rectangular shape.

For example, the two-dimensional pattern of first regions may comprise at least 2 parallel strips of first regions. Applying localized heating to the birefringent plate may include applying localized heating to alter the birefringence of at least 2 parallel first zone strips in the birefringent plate such that the at least 2 parallel first zone strips have a birefringence that is different from the birefringence of the second zone that does not receive localized heating.

In some examples, the width of each first region stripe is in a range of 50 μm to 1000 μm. In some examples, the width of each first region stripe is in a range of 100 μm to 600 μm. In some examples, the width of each first region stripe is in the range of 200 μm to 400 μm.

For example, the fiber optic connector component may be configured to optically couple to a plurality of vertical coupling elements on the photonic integrated circuit, and the fiber optic connector is configured to enable transmission of the first set of light beams and the second set of light beams between the plurality of optical fibers and the plurality of vertical coupling elements.

For example, applying localized heating or localized excitation to the birefringent plate may include applying localized heating or localized excitation to the birefringent plate to reduce birefringence at the first region.

For example, applying localized heating or localized excitation to the birefringent plate may include applying localized heating or localized excitation to the birefringent plate to reduce the birefringence at the first region to substantially zero birefringence.

For example, applying localized heating or localized excitation to the birefringent plate may include applying localized heating or localized excitation to the birefringent plate to reduce birefringence at the first region such that the second set of light beams passing through the second region have a polarization rotated by about 90 ° +n×180° relative to the polarization of the first set of light beams passing through the first region, 0 n being an integer.

For example, the fiber optic connector component may include a walk-off element configured to: receiving a plurality of light beams from the plurality of optical fibers, splitting the light beams into a first light beam component and a second light beam component, the second light beam component having a polarization orthogonal to the polarization of the first light beam component, directing the first light beam component to a first region having a lower birefringence, and directing the second light beam component to a second region having a higher birefringence.

For example, applying localized heating or localized excitation to the birefringent plate may include applying localized heating or localized excitation to the birefringent plate to reduce birefringence at the first region such that the second set of light beams passing through the second region have a polarization rotated by about 90 ° +n×180° relative to the polarization of the first set of light beams passing through the first region, 0 n being an integer. The walk-off element may be configured such that upon exiting the walk-off element, the first beam component has a first polarization and the second beam component has a second polarization substantially orthogonal to the first polarization. The first and second regions of the birefringent plate may be configured such that, after passing through the first and second regions, the first beam component has a polarization substantially parallel to the polarization of the second beam component.

For example, the walk-off element may separate the first beam component and the second beam component along a walk-off direction, and each of at least some of the first regions may have a dimension in the range of 50 μm to 1000 μm measured in a direction parallel to the walk-off direction.

For example, each of at least some of the first regions may have a dimension in the range of 100 μm to 600 μm measured in a direction parallel to the walk-off direction.

For example, each of at least some of the first regions may have a dimension in the range of 200 μm to 400 μm measured in a direction parallel to the walk-off direction.

For example, applying localized heating or localized excitation to the birefringent plate may include applying localized heating or localized excitation to the birefringent plate to change the birefringence of the two-dimensional pattern of the first region in the birefringent plate such that the birefringent plate is configured to change the polarization of light passing through the birefringent plate in a manner equivalent to changing the polarization of light passing through a half-wave plate having openings at the two-dimensional pattern of the first region.

For example, the birefringent plate includes a first surface and a second surface. Applying localized heating to the birefringent plate may include applying localized heating to a first region extending from the first surface to the second surface. See the example of the birefringent plate 5204 in fig. 52B.

For example, the birefringent plate includes a first surface and a second surface. Applying localized heating or localized excitation to the birefringent plate may include applying localized heating or localized excitation to a first region positioned within the birefringent plate and spaced a first distance from the first surface and a second distance from the second surface. See the example of birefringent plate 5212 in fig. 52C.

For example, the birefringent plate includes a first surface and a second surface. Applying localized heating or localized excitation to the birefringent plate may include applying localized heating or localized excitation to a first region extending from the first surface to a location inside the birefringent plate, the first region being spaced a distance from the second surface. See the example of birefringent plate 5216 in fig. 52D.

For example, one or more laser beams may be used to apply localized heating.

For example, one or more particle beams may be used to apply local excitation.

During the manufacture of the fiber optic connector components, the various optical components in the fiber optic connector need to be properly aligned to ensure that the light beam from the optical fiber can be properly transmitted to the vertical coupling element on the photonic integrated circuit. For example, to fabricate connector element 250, collimating lens array 551 needs to be aligned with polarization diversity assembly 757. Time is required to align the optical assembly, so it is preferable to manufacture the fiber optic connector component in a manner that reduces the time taken to align the optical assembly.

In some embodiments, a process for manufacturing a plurality of fiber optic connector components includes aligning and bonding a plurality of optical assemblies in parallel to form an assembly, and then cutting the assembly to singulate (singulate) individual fiber optic connector components. For example, a first substrate having a plurality of non-singulated first optical assemblies is aligned with and bonded to a second substrate having a plurality of non-singulated second optical assemblies to form an assembly comprising the first substrate and the second substrate. The assembly is then cut to singulate individual fiber optic connector components including the first optical assembly and the second optical assembly.

In some embodiments, a process for manufacturing a fiber optic connector component includes: providing a first module having a plurality of non-singulated lens arrays; and providing a second module having a plurality of non-singulated patterned birefringent plates. Each patterned birefringent plate comprises a birefringent material, the patterned birefringent plate comprising a plurality of first regions having reduced or no birefringence compared to a plurality of second regions. The process includes aligning the plurality of non-singulated lens arrays in a first module with the plurality of non-singulated patterned birefringent plates in a second module; bonding the first module to the second module to form a first assembly; and cutting the first assembly to singulate the first and second modules to produce a plurality of fiber optic connector components. Each fiber optic connector assembly includes a singulated birefringent plate and a singulated lens array.

In some embodiments, the patterned birefringent plate may be configured to be equivalent to a quarter wave plate with apertures, using similar principles as described above.

In some embodiments, the array of grating couplers may include a first subset of grating couplers for coupling optical signals between the core and the photonic integrated circuit, and a second subset of grating couplers not for coupling optical signals between the core and the photonic integrated circuit. The second subset of grating couplers may be used for alignment purposes.

Referring to fig. 22, an array of grating couplers 2200 includes an emitter-grating coupler 2202 (denoted by reference "T") for coupling out or emitting an optical signal, a receiver-grating coupler 2204 (denoted by reference "R") for coupling in or receiving an optical signal, and an optical energy source-grating coupler 2206 (denoted by reference "P") for coupling in optical energy source light. This example assumes walk-off direction 2208. The array of grating couplers 2200 includes unused transmit-grating couplers, e.g., 2210a and 2210b, which may be connected by waveguides (e.g., 2212) to achieve active alignment during assembly. The term "unused emitter-grating coupler" refers to a grating coupler that is not used to couple optical signals between an optical fiber and a photonic integrated circuit. For example, an "unused transmit-grating coupler" is positioned adjacent to another transmit-grating coupler, wherein the unused transmit-grating coupler is displaced from the second transmit-grating coupler in the walk-off direction.

The photonic integrated circuit may be designed such that an optical signal is output from the photonic integrated circuit to the grating coupler 2210a, and a photodetector detects the light received from the grating coupler 2210 b. During assembly of the photonic integrated circuit and the fiber to PIC connector, the light received from the grating coupler 2210b is monitored to optimize the fiber to PIC connector to fiber alignment with the photonic integrated circuit, for example, by finding the alignment between the photonic integrated circuit and the fiber to PIC connector that achieves the highest light transmission efficiency from the light output port of the photonic integrated circuit to the grating coupler 2210a and from the grating coupler 2210b to the photodetector.

By using unused transmit-grating couplers within the grating coupler array for alignment purposes, there is no need to increase the overall footprint of the grating coupler array. The geometry of the array of grating couplers can be maintained. In the example of fig. 22, the array of grating couplers occupies a total rectangular footprint. In order to provide a grating coupler for alignment purposes, the geometry of the array need not extend beyond the rectangular footprint.

Silicon photonic integrated circuits may have limitations (e.g., soft limitations) in the optical power that they can handle. Excessive optical power may result in nonlinear excessive waveguide loss. The optical power on the photonic integrated circuit is kept below a certain value to avoid excessive nonlinear waveguide losses. The fiber core may transmit optical energy light having a power greater than that which may be properly handled by the photonic integrated circuit. The optical energy beam splitter may split the optical energy beam from the core into two or more optical energy beams such that each optical energy beam has a power level suitable for the photonic integrated circuit.

Referring to fig. 23, in some embodiments, the fiber-to-PIC connector 2300 includes an optical energy fiber port configured to receive optical energy light 2302 from a Polarization Maintaining Fiber (PMF) 2304 having an axis aligned at 45 ° relative to an exit axis of the exit element 2306. The walk-off element 2306 causes half of the light from the optical fiber 2304 to travel to each of the plurality of walk-off paths, thereby producing a first optical energy beam 2308 and a second optical energy beam 2310. The first optical energy beam 2308 is coupled to the photonic integrated circuit 2316 through a first grating coupler 2312 and the second optical energy beam 2310 is coupled to the photonic integrated circuit 2316 through a second grating coupler 2314. Each grating coupler stably receives half of the power transmitted by polarization maintaining fiber 2304. In this way, polarization maintaining fiber 2304 may transmit optical energy light at twice the amount of power that may be properly handled by photonic integrated circuit 2316.

24A-24C, in some embodiments, the optical energy beam from polarization maintaining fiber 2304 may be split into four beams to allow polarization maintaining fiber 2304 to transmit optical energy light having a power four times the amount of power that may be properly handled by photonic integrated circuit 2316.

Fig. 24A is a side view of a fiber-to-PIC connector 2400 that includes an optical energy fiber port configured to receive an optical energy beam 2402 from a Polarization Maintaining Fiber (PMF) 2304, the axis of which is aligned at 45 ° relative to the walk-off axis of the first walk-off element 2408. The first walk-off element 2408 causes half of the light from the fiber 2304 to travel to each of the plurality of walk-off paths, thereby producing beam 1 (2404) and beam 2 (2406). The quarter waveplate 2426 changes the linearly polarized light beam 1 and the light beam 2 into circularly polarized light beams. The second walk-off element 2410 performs a second polarization splitting of the light beams 1 and 2. Thus, the light beam 1 is split into the light beam 1a and the light beam 1b, and the light beam 2 is split into the light beam 2a and the light beam 2b. This results in a 1:4 power split. The second walk-off element 2410 rotates at an angle θ (e.g., 90 °) relative to the first walk-off element 2408 such that the walk-off direction of the second walk-off element 2410 is at an angle θ relative to the walk-off direction of the first walk-off element 2408. The subsequent birefringent aperture plate rotates the polarization of some of the beams to ensure that the polarization directions of all beams are properly aligned with the grating coupler.

Fig. 24B is a second side view of the fiber-to-PIC connector 2400. The second side view is the first side view of fig. 24A from the right side. As shown in fig. 24B, the second walk-off element 2410 causes half of the light from the light beam 1 (2404) to travel to each of the plurality of walk-off paths, thereby generating the first light beam 1a and the second light beam 1B. The second walk-off element 2410 causes half of the light from the light beam 2 (2406) to travel to each of the plurality of walk-off paths, thereby generating a third light beam 2a and a fourth light beam 2b. The half-wave plate 2412 rotates the second light beam 1b and the fourth light beam 2b such that the first light beam 1a, the second light beam 1b, the third light beam 2a and the fourth light beam 2b have the same polarization direction.

Fig. 24C is a schematic diagram illustrating the location 2414 of the optical energy light 2402 received from the Polarization Maintaining Fiber (PMF) 2304. The first walk-off element 2408 causes the light beam 2 (2406) to be shifted to a position 2418 in the first walk-off. Beam 1 remains at the same location 2414 as the optical energy source light 2402. The second walk-off element 2410 causes the beam 1b to be shifted to position 2422 in the second walk-off 2420. Beam 1a remains at the same location 2414 as beam 1 (2402). The second walk-off element 2410 causes the beam 2b to be shifted to position 2424 in the second walk-off 2420. Beam 2a remains at the same location 2418 as beam 2 (2406).

The light beams 1a, 1b, 2a, 2b are coupled to the photonic integrated circuit by four grating couplers. Each grating coupler stably receives one quarter of the power transmitted by polarization maintaining fiber 2304. In this way, polarization maintaining fiber 2304 can transmit optical energy light having a power four times the amount of power that can be properly handled by photonic integrated circuits.

In some examples, instead of using the quarter wave plate 2426, the second walk-off element 2410 may be aligned at 45 ° relative to the first walk-off element 2408 to achieve the same splitting effect.

In some embodiments, a third walk-off element is used to split the four light beams 1a, 1b, 2a, 2b into eight light beams. This allows polarization maintaining fiber 2304 to transmit optical energy light having a power eight times the amount of power that can be properly handled by the photonic integrated circuit.

Additional walk-off elements may be used to further split the beam to allow polarization maintaining fiber 2304 to transmit optical energy light having a greater power, such as 16, 32, 64, 128 or more times the amount of power that can be properly handled by the photonic integrated circuit.

Referring to fig. 25, coupling an optical energy source to a photonic integrated circuit may require careful polarization alignment because modulators on the photonic integrated circuit may be polarization sensitive, i.e., only one fixed linear polarization that effectively modulates light. The laser 2500 may emit linearly polarized light and a Linear Polarization Maintaining Fiber (LPMF) 2502 may be used to connect the external optical energy source 2500 to the photonic integrated circuit 2504.

Referring to fig. 26, if more than one energy input is required, the LPMF 2502 may be oriented 45 degrees relative to the Polarizing Beam Splitter (PBS) 2600, thereby achieving 1:2 optical power splitting of the optical energy source at the PBS 2600. In this example, equal power splitting may require accurate angular alignment of the LPMF 2502 at the laser 2500 and PBS2600, which may increase the cost of the package assembly.

In some embodiments, circularly polarized light fibers may be used as distribution fibers for optical energy sources. Referring to fig. 27, a quarter wave plate 2700 is provided at the laser 2500 to change the polarization state of the laser output from linear to circular. A Circular Polarization Maintaining Fiber (CPMF) 2702 transmits circularly polarized light from the quarter wave plate 2700 to the polarizing beam splitter 2600. Since circular polarization is a superposition of two linear polarizations, PBS2600 still performs 1:2 power splitting. An advantage of this design is that CPMF 2702 can be mounted to laser 2500 and PBS2600 at any angle of rotation, thereby reducing alignment or packaging costs. This general architecture is also applicable to other polarization splitting interfaces, including 2D grating couplers.

For example, circular polarization maintaining fiber 2702 may be made from a rotating birefringent fiber that is generated by rotating a polarization maintaining preform during drawing. The fiber is designed to maintain circular polarization: if the input light is right-hand circularly polarized, the output light will also be right-hand circularly polarized; and if the input light is left-hand circularly polarized, the output light will also be left-hand circularly polarized. Circular polarization may be considered as a superposition of two linear polarizations with a defined phase angle (e.g., 90 °) between them. The circular polarization maintaining fiber 2702 maintains a phase angle (e.g., 90 °) difference between two linear polarizations as they travel along the fiber.

Referring to fig. 28, a fiber-to-PIC connector 2800 (identical to connector 2300 of fig. 23) receives optical energy light from a circular polarization maintaining fiber 2802. The CPMF fiber 2802 may be attached at any random angle relative to the walk-off crystal to perform the indicated power splitting.

The techniques described above for the examples of fig. 27 and 28 may also be used to perform polarization splitting at the input of the edge-coupled interface.

Fig. 29 shows a schematic diagram of WDM multiplexer 2900. In this example, WDM multiplexer 2900 can multiplex 4 wavelengths. The same principle can be used to design a WDM multiplexer that can multiplex N wavelengths, where N is an integer greater than 4. For example, WDM multiplexer 2900 can operate as follows. Four grating couplers 2902 emit 4 signals at different light Wavelengths (WL) and all at the same polarization. Birefringent aperture plate 2904 rotates 2 out of 4 WLs, e.g., WL2 and WL 4. Fig. 2912 shows the polarization directions of WL1, WL2, WL3, and WL4 after passing through the birefringent aperture plate 2904. The first walk-off element 2906 combines the two polarizations (from the inverse operation of polarization separation in the above embodiments). Waveplate 2908, shown in red, is a higher order (relatively thicker) waveplate, the thickness of which is designed such that:

● The polarization of WL1 is not changed ("full-wave chip at WL 1");

● Rotating the polarization of WL2 by 90 degrees ("half-wave plate at WL 2");

● Rotating the polarization of WL3 by 90 degrees ("half-wave plate at WL 2"); and

● The polarization of WL4 is not changed ("full-wave chip at WL 4").

FIG. 2914 shows the polarization directions of WL1, WL2, WL3, and WL4 after passing through waveplate 2908. The second walk-off element 2910 combines (a) the light beam at wl1+wl2 of one polarization and (b) the light beam at wl3+wl4 of orthogonal polarization. The thickness of second walk-off element 2910 is about twice the thickness of first walk-off element 2906 because the shift between the two beams combined by second walk-off element 2910 is about twice the shift between the two beams combined by first walk-off element 2904.

Fig. 30 is a crystal plate including a quartz half wave plate 3002 and yttrium vanadate (YVO ₄ ) Schematic of an example of WDM multiplexer 3000 with waveplate 3006.

Referring to FIG. 31A, table 3100 shows 20And 0GBASE-FR4 wavelength division multiplexing channel allocation. Referring to FIG. 31B, table 3102 shows a 200GBASE-LR4 WDM channel allocation. These channels provide 800GHz spacing and 368GHz windows. Channel L in tables 3100 and 3102 ₀ 、L ₁ 、L ₂ And L ₃ The center wavelengths of (a) may correspond to wavelengths WL1, WL2, WL3, and WL4 in fig. 29 and 30.

Referring to fig. 32, a table 3200 shows 400GBASE-FR8 wavelength division multiplexed channel allocation. The center wavelengths of the eight channels in table 3200 may be used in a WDM multiplexer having three walk-off elements multiplexing eight different wavelengths.

The fiber-to-PIC connectors described above may be oriented such that the optical axis is parallel or perpendicular (or at any other angle) to the top surface of the photonic integrated circuit. The optical axis of the fiber-to-PIC connector refers to the optical axis of the walk-off element and the birefringent aperture plate.

FIG. 33A is a top view of an example of an optoelectronic device 3300. FIG. 33B is a side view of a first configuration of an optoelectronic device 3300 in which the fiber-to-PIC connector has an optical axis parallel to the top surface of photonic integrated circuit 3304. The lens array 3308 couples the light beam propagating in a direction perpendicular to the top surface of the PIC 3304, and the steering mirror 3306 changes the propagation direction of the light beam. The birefringent aperture plate 3310 is attached to the steering mirror 3306. Walk-off crystal 3312 is attached to birefringent aperture plate 3310. Lens array 3314 is attached to walk-off crystal 3312. The fiber array 3316 is optically coupled to a lens array 3314.

Fig. 33C is a side view of a second configuration of optoelectronic device 3300 in which fiber-to-PIC connector 3310 has an optical axis perpendicular to the top surface of photonic integrated circuit 3304. The walk-off element outputs a light beam that propagates in a direction perpendicular to the top surface of PIC 3304. The turning mirror 3312 changes the propagation direction of the light beam. In the example of fig. 33A-33C, the turning mirror (e.g., 3306, 3312) changes the direction of the light beam to achieve horizontal fiber attachment, i.e., where the fiber extends parallel to the top surface of the photonic integrated circuit.

Fig. 34A is a side view of a fiber-to-PIC connector 3400 optically coupling an optical fiber 3402 to a PIC 3404. The fiber-to-PIC connector 3400 enables edge coupling of optical signals.

Fig. 34B is a top view of a fiber-to-PIC connector 3400 optically coupling an optical fiber 3402 to a PIC 3404. The fiber-to-PIC connector 3400 enables an optical signal to be edge-coupled to a waveguide 3406 on the PIC.

In some embodiments, the fiber-to-PIC connector may include a filter-based WDM demultiplexer and/or multiplexer. Such fiber-to-PIC connectors may convert a single row of N fibers to an Nx2M grating coupler array, where M is the number of wavelengths used.

The fiber-to-PIC connector may include a wavelength division multiplexer and/or a demultiplexer. Fig. 35 is a side view of an example of a fiber-to-PIC connector 3500 that receives a Wavelength Division Multiplexed (WDM) optical signal from optical fiber 3502 or transmits a WDM optical signal to optical fiber 3502. In this example, the WDM signal in optical fiber 3502 includes four wavelengths WL1, WL2, WL3, and WL4. Although the side view of fig. 35 shows one optical fiber 3502, it should be understood that there are more fiber staggers behind the fibers shown.

The first filter 3504 allows the light signal having the wavelength WL1 to pass therethrough, and reflects the light signals having the wavelengths WL2, WL3, and WL4. The second filter 3506 reflects the optical signal having the wavelength WL2 and allows the optical signals having the wavelengths WL3 and WL4 to pass. The third filter 3508 reflects the optical signal having the wavelength WL3 and allows the optical signal having the wavelength WL4 to pass. The wavelength independent mirror 3510 reflects an optical signal having a wavelength WL4.

When fiber-to-PIC connector 3500 is used as a demultiplexer, WDM optical signals having components with wavelengths WL1, WL2, WL3, and WL4 (two polarizations for each wavelength) are split by filters 3504, 3506, and 3508 into four optical signals, each having a wavelength. The four single wavelength optical signals pass through walk-off element 3512 and birefringent aperture plate 3514, producing eight optical signals having the same polarization state that are properly aligned with grating coupler 3516. When fiber-to-PIC connector 3500 is used as a multiplexer, optical signals having wavelengths WL1, WL2, WL3, and WL4 from grating coupler 3516 pass through birefringent aperture plate 3514 and walk-off element 3512 and are directed to fiber 3502 by filters 3504, 3506, 3508 and mirror 3510.

Fig. 36A shows a schematic diagram of an example of an arrangement of fiber ports 3600, a birefringent aperture plate 3602, and an array of grating couplers 3604. Fig. 36B is an enlarged view of the arrangement of fiber ports 3600. The reference numerals "P", "R" and "T" denote an optical energy fiber port, a receiver fiber port and a transmitter fiber port, respectively. Fig. 36C is an enlarged view of an array of grating couplers 3604. Reference numerals "P", "R" and "T" denote an optical energy grating coupler for coupling optical energy light, a receiver grating coupler for coupling in or receiving an optical signal, and an emitter grating coupler for coupling out or emitting an optical signal, respectively. For demultiplexing, the WDM optical signal from each fiber port 3600 is split into eight optical signals having four different wavelengths. For multiplexing, eight optical signals with four different wavelengths from grating coupler 3604 are multiplexed into a WDM optical signal and directed to fiber port 3600.

Fig. 37A is a schematic diagram illustrating an example of waveguide routing from a grating coupler to a modulator on the PIC. Fig. 37B is an enlarged view showing waveguide routing from the grating coupler to the input and output ports. Two rows of grating couplers 3702 process optical signals having a wavelength WL 1. The next two rows of grating couplers 3704 process the optical signal having wavelength WL 2. The next two rows of grating couplers 3706 process the optical signal having wavelength WL 3. The next two rows of grating couplers 3708 process the optical signal having wavelength WL 4.

The first set of modulators 3710 processes optical signals having wavelengths WL1 and WL 2. The second set of modulators 3712 processes optical signals having wavelengths WL3 and WL 4. Each modulator has an input port 3714 for receiving optical energy source light and an output port 3716 for outputting an emitted signal. In this example, the grating couplers processing wavelengths WL1 and WL2 are located on a first side of the grating coupler array near the first set of modulators 3710. The grating couplers processing wavelengths WL3 and WL4 are located on a second side of the grating coupler array near the second set of modulators 3712. This avoids crossing of the waveguides and makes it easier to design waveguide routing.

Referring to fig. 38, in some embodiments, a fiber-to-PIC connector 3800 is configured to couple to a plurality of rows of optical fibers 3802. The figure shows a side view of the fiber-to-PIC connector 3800, showing two fibers 3802 from two rows. It should be understood that there are more fiber staggering behind the fibers shown. Connector 3800 can convert K rows of N fibers (e.g., fiber (a), fiber (b)) to an Nx2MK grating coupler array, where M represents the number of wavelengths used. In this example, m=4 wavelengths are used.

The fiber-to-PIC connector 3800 includes a first filter 3804, the first filter 3804 allowing the optical signal having the wavelength WL1 to pass therethrough, and reflecting the optical signals having the wavelengths WL2, WL3, and WL 4. The second filter 3806 reflects the optical signal having the wavelength WL2 and allows the optical signals having the wavelengths WL3 and WL4 to pass. The third filter 3808 reflects an optical signal having a wavelength WL3 and allows an optical signal having a wavelength WL4 to pass. The wavelength independent mirror 3810 reflects an optical signal having a wavelength WL 4. Filters 3804, 3806, and 3808 are large enough to be able to process the light beams from two fibers 3802. In this example, the two input beams are split into 16 beams that are directed to a grating coupler.

Fig. 39 is a schematic diagram of an example of a fiber-to-PIC connector 3900 that includes filter-based wavelength-division demultiplexers and multiplexers that include broadband splitters and bandpass filters having different splitting ratios. The fiber-to-PIC connector 3900 includes a first broadband splitter having a split ratio of 25%:75% such that 25% of the light passes through the splitter to form beam 3904 and 75% of the light is reflected to form beam 3906. Beam 3904 passes through walk-off element 3908, and walk-off element 3908 splits beam 3904 into two polarized beams 3912a and 3912b. The polarized light beams 3912a, 3912b are filtered by a first bandpass filter 3910 that allows the wavelength WL1 to pass through.

Light beam 3906 is directed to second broadband beam splitter 3918 having a split ratio of 33%:67% such that 33% of light beam 3906 is reflected by beam splitter 3918 to form light beam 3914 and 67% of light beam 3906 passes through beam splitter 3918 to form light beam 3916. Beam 3914 has approximately 75% x 33% = 25% of the power of input beam 3922 from fiber 3924. Beam 3916 has approximately 75% x 67% = 50% of the power of input beam 3922. Beam 3914 passes through walk-off element 3908, and walk-off element 3908 splits beam 3914 into two polarized beams that are filtered by second bandpass filter 3920 that allows wavelength WL2 to pass through.

Light beam 3916 is directed to a third broadband beam splitter 3922 having a 50%:50% splitting ratio such that 50% of light beam 3916 is reflected by beam splitter 3922 to form light beam 3924 and 50% of light beam 3916 passes through beam splitter 3922 to form light beam 3926. Each of beams 3924 and 3926 has approximately 50% x 50% = 25% of the power of input beam 3922. Beam 3924 passes through walk-off element 3908, and walk-off element 3908 splits beam 3924 into two polarized beams filtered by third bandpass filter 3928 that allows wavelength WL3 to pass through.

Beam 3926 is directed toward steering mirror 3930, steering mirror 3930 reflecting 100% of beam 3926 toward walk-off element 3908, walk-off element 3908 splitting beam 3926 into two polarized beams filtered by fourth bandpass filter 3932 allowing wavelength WL4 to pass.

In some embodiments, bandpass filters 3910, 3920, 3928, and 3932 may be used in the multiplexers and demultiplexers of fig. 35 and 38 to reduce inter-channel crosstalk.

In some embodiments, a non-reciprocal optical element may be used to form the isolator. Referring to fig. 40, fiber-to-PIC connector 4000 includes an isolator so that light exiting the photonic integrated circuit (exit, blue arrow) does not trace back into the input (incident) optical path (red arrow) and thus does not couple back into the fiber. The principle is also applicable to gray (greyed-out) orthogonal polarization.

The incident light beam 4004 is split by the walk-off element 4002 into a first incident light beam 4008 having polarization a and a second incident light beam 4010 having polarization a'. Rotation of half-wave plate 4006 by-45 rotates the polarization of first incident light beam 4008 to have polarization B. The first incident beam 4008 then passes through +45° faraday rotator 4012 to have polarization C aligned with grating coupler 4016. The polarization directions A, B and C are shown in fig. 4014 at the upper part of the figure.

The outgoing beam initially has polarization C. The outgoing beam passes through +45° faraday rotator 4012 to have polarization B. The walk-off element 4002 changes the propagation direction of the outgoing light beam to a light beam 4018 that does not trace back the input (incident) light path and thus does not couple back into the optical fiber 4020.

Referring to fig. 41, which is similar to fig. 10B, an incident light beam 1008 from a fiber core 302 passes through a collimating lens 551 and is split by an walk-off crystal 1004 into two light beams 755a and 755B (see fig. 10A), which pass through a birefringent aperture plate 1006, a second lens 541, and a third lens 542, and are directed to a vertical coupling element 231 on photonic integrated circuit 210 at an incident angle θ. In some embodiments, the angle of incidence θ may be in a range, for example, from 1 ° to 30 °, from 1 ° to 16 °, or from 4 ° to 12 °, or about 8 °, depending on the design of the vertical coupling element 231, the vertical coupling element 231 may be, for example, a vertical grating coupler.

In some implementations, each third lens 542 is a rotationally symmetric spherical lens. To direct the light beams (e.g., 755a and 755 b) at a desired angle of incidence θ toward the grating coupler, the core 302, collimating lens 551, walk-off crystal 1004, birefringent aperture plate 1006, and second lens 541 are positioned relative to the third lens 542 such that each light beam propagates along a beam path that is parallel to the optical axis of the third lens 542, and the central axis of the beam path is spaced from the optical axis of the third lens 542 by an offset distance. The grating coupler 231 is positioned near the focal plane of the third lens 542, and the optical axis of the third lens 542 passes through the location on the grating coupler 231 where the light beam is intended to be focused, i.e., the focal point. The light beam is refracted by the third lens 542 toward the focal point at the incident angle θ. The offset distance and focal length of the third lens 542 are selected to achieve a desired angle of incidence θ. For example, increasing the offset distance of the third lens 542 and/or decreasing the focal length results in a greater angle of incidence θ.

In the above example, in order to achieve a desired offset between the central axis of the light beam and the optical axis of the third lens 542, the diameter of the light beam is made smaller than the diameter of the third lens 542. In an optical system in which the propagating beam is centered on the optical axis, a rotationally symmetric spherical lens shape provides sufficient area for use when arranged in an array. However, in the case of an off-axis beam as described above and shown in fig. 41, only a portion of the lens area is used, depending on the offset of the beam relative to the optical axis. In the case of large beam offsets, only a small fraction of the lens area is used, which limits the density of individual beams propagating through the lens array system.

One aspect of robust optical system design relates to the diameter of the beam passing through the system. Too small a beam diameter can cause the beam to diverge too quickly, which limits the longitudinal distance over which the collimated beam can travel. For example, a beam having a diameter of 100 μm at a wavelength of 1.5 μm produces a beam having a diameter of about 138 μm over a propagation distance of 5 mm. It is therefore important to make the collimated beam wide enough to propagate through the optical system without excessive widening (e.g., the fiber-to-PIC connector 1000 of fig. 10A, 10B, 12, 13A). In dense arrays, as in the example shown in fig. 20A-20C, the beam diameter is defined by the distance between adjacent beams within the array.

In some embodiments, to increase area usage, increase the density of individual beams propagating through the lens array system, and increase beam diameter as much as possible, a circular asymmetric or rotationally asymmetric optical lens array is provided, wherein each rotationally asymmetric optical lens has a surface profile similar to a truncated conventional rotationally symmetric optical lens, such that only the portion of the rotationally symmetric optical lens that intersects the offset beam is implemented. This allows unused areas of conventional lenses to be used with other rotationally asymmetric optical lenses in the array, thereby increasing the density of individual beams, as compared to using conventional rotationally symmetric optical lens arrays.

Examples of circularly asymmetric or rotationally asymmetric optical lens arrays configured to couple an incident beam array to an array of vertical coupling elements, e.g., grating couplers, on an active layer of a photonic integrated circuit at an incident angle, e.g., in the range of from 1 ° to 30 °, from 1 ° to 16 °, or from 4 ° to 12 °, or about 8 °, depending on the design of the vertical coupling elements, are described below. For example, if the grating coupler is designed to emit light at an angle θ relative to the normal direction of the main surface of the photonic integrated circuit, the circularly asymmetric optical lens is designed such that the incident light beam is focused onto the grating coupler at an angle of incidence equal to θ. The grating coupler is designed to reduce the amount of reflection at the interface between the grating coupler and an optical waveguide coupled to the grating coupler.

For example, the circularly asymmetric optical lens may be a free-form optical lens, which may be a free-form off-axis optical lens. The circularly asymmetric optical lens may be prepared by, for example, gray scale lithography and subsequent etching or 3D printing. The circularly asymmetric optical lens may be made of, for example, silicon, glass, or polymer-based materials.

Fig. 42 shows top and side views of an example of a circularly asymmetric (or rotationally asymmetric) optical lens 4200, the optical lens 4200 focusing an offset beam 4202 onto a vertical coupling element 4204 at an incident angle θ. The vertical coupling element 4204 may be, for example, a vertical grating coupler, which may be positioned near the focal plane 4214 of the circularly asymmetric optical lens 4200. The vertical coupling element 4204 is coupled to an optical waveguide 4210, the optical waveguide 4210 transmitting light received at the vertical coupling element 4204 to other portions of the photonic integrated circuit. An array of circular asymmetric optical lenses 4200 may be used, for example, in the fiber-to-PIC connector 1000.

The circularly asymmetric optical lens 4202 is circularly asymmetric or rotationally asymmetric with respect to the optical axis 4206. Circular asymmetric optical lens 4202 may be considered to implement only the portion of circular symmetric optical lens 4208 shown in phantom that intersects offset beam 4202. Circularly symmetric optical lens 4208 is rotationally symmetric with respect to optical axis 4206. In the example shown in fig. 42, the size of circularly asymmetric optical lens 4200 (measured in a direction parallel to the major surface of the photonic integrated circuit and in a plane aligned with optical axis 4206) is slightly larger than the radius of circularly symmetric lens 4208. In some embodiments, the size of the circularly asymmetric optical lens 4200 may be, for example, 80%, 70%, 60%, 50%, 40%, 30% or 20% of the diameter of the circularly symmetric lens 4208, or any value in the range of 20% to 80%, depending on the offset between the optical beam 4202 and the optical axis 4206. By using a smaller circularly asymmetric optical lens 4200 that occupies a smaller area than circularly symmetric lens 4208, the released space may be used by other circularly asymmetric optical lenses, thereby enabling a higher density of light beams 4202 to be coupled to a corresponding higher density of grating couplers 4204.

The above description relates to how the circularly asymmetric lens 4200 focuses the beam 4202 onto the grating coupler 4204 at an angle of incidence θ. The same principle operates inversely, i.e., grating coupler 4204 may emit light at an exit angle θ to circular asymmetric lens 4200, and circular asymmetric lens 4200 collimates the light to form a collimated light beam coupled to a corresponding fiber core (e.g., 302).

In the example of fig. 42, the circularly asymmetric optical lens 4200 has a circular outer circumference as viewed in the direction of the beam path, as if a circular cutout were cut from the circularly symmetric lens 4208. The circularly asymmetric optical lens does not necessarily have a circular outer circumference or footprint. The outer circumference or footprint of the circularly asymmetric optical lens may be, for example, circular, elliptical, square, rectangular, polygonal, or any arbitrary shape. The circumference of each circularly asymmetric optical lens 4200 is designed such that a circularly asymmetric lens 4200 may receive as many beams 4202 as possible to achieve a desired signal-to-noise ratio while adjacent circularly asymmetric optical lenses 4200 are packaged as densely as possible without causing interference or crosstalk between adjacent beams.

In some embodiments, the circumference or footprint of the circularly asymmetric optical lens 4200 corresponds to the cross-sectional shape of the offset beam 4202. For example, if offset beam 4202 has a circular or elliptical cross-section, circular asymmetric optical lens 4202 is also designed to have a corresponding circular or elliptical circumference or footprint.

The circularly asymmetric lens 4200 includes an upper surface having a contour or curvature that matches the contour or curvature of the surface of the circularly symmetric lens 4208. The surface profile or curvature of the circularly asymmetric lens 4200 may correspond to the curvature of a spherical lens. The surface profile or curvature of the circularly asymmetric lens 4200 may correspond to the curvature of an aspheric lens, e.g., to correct one or more optical aberrations. For example, in a wavelength division multiplexed system, beam 4202 carries multiple wavelengths, and a circularly asymmetric optical lens with an aspheric surface profile may be designed to correct chromatic aberration to maximize coupling efficiency at the grating coupler. Such circularly asymmetric optical lenses may be fabricated using techniques for producing free-form lenses.

Circular asymmetric lens 4200 has an edge 4212 along an outer circumference, edge 4212 being located outside of the region intersecting beam 4200. The edge 4212 may be parallel to the optical axis 4206, inclined at an angle with respect to the optical axis 4206, have a stepped shape, or any shape.

In some embodiments, beam 4202 is a gaussian beam, and the area of beam 4200 refers to a region having an intensity of at least half of the peak intensity of beam 4202. The edges of beam 4202 having an intensity less than half of the peak intensity may extend to or beyond the circumference of circular asymmetric lens 4200, thus taking into account reflection and/or refraction at edge 4212 in the design of the entire array of circular asymmetric lenses 4200. The geometry of each individual circularly asymmetric lens 4200 and the spacing between adjacent circularly asymmetric lenses 4200 are selected such that interference and/or crosstalk between light beams is negligible.

Fig. 43 is a schematic diagram showing top and side views of a circular asymmetric optical lens array 4200, circular asymmetric optical lens array 4200 focusing beam 4202 toward normal grating coupler 4204 at an angle of incidence θ that matches the design of grating coupler 4204 to increase coupling efficiency at grating coupler 4204 and reduce reflection at the interface between grating coupler 4204 and optical waveguide 4210.

The circularly asymmetric lens 4200a occupies a portion of the area of a conventional circularly symmetric lens 4208 shown in phantom. Circular asymmetric lens 4200 implements a portion of rotationally symmetric optical lens 4208 that intersects offset beam 4202. The released unused area is used by other circularly asymmetric lenses 4200. In this example, because adjacent circularly asymmetric lenses 4200b, 4200c, 4200d overlap with the area of circularly symmetric lens 4208, it is difficult to densely package beam 4202 as shown in fig. 43 if circularly symmetric lens array 4208 is used.

In this example, the distance between adjacent fiber cores 302 is about 250 μm, and the distance between the two nearest adjacent circular asymmetric lenses 4200 is aboutThe step height of the edge 4212 near the optical axis 4206 is about 14 μm. The walk-off direction of walk-off crystal 1004 is located in a diagonal direction indicated by arrow 4300. For example, a light beam from the fiber core 302 aligned with the circular asymmetric lens 4200e is split off the crystal 1004 into two light beams aligned with the circular asymmetric lens 4200e and the circular asymmetric lens 4200f, which focus the two light beams at a specified angle of incidence towards two corresponding grating couplers.

Fig. 44 is a schematic diagram showing a top view and a side view of circularly symmetric optical lens array 4208, circularly symmetric optical lens array 4208 focusing beam 4202 toward normal grating coupler 4204 at an angle of incidence θ that matches the design of grating coupler 4204 to increase coupling efficiency at grating coupler 4204 and reduce reflection at the interface between grating coupler 4204 and optical waveguide 4210. The diameter of beam 4202 is less than half the diameter of circularly symmetric optical lens 4208. If a circularly symmetric optical lens array 4208 is used for the fiber-to-PIC connector, the distance between the fiber cores 302 would be about 450 μm and the distance between two nearest neighbor circularly symmetric lenses 4208 would be aboutA comparison of fig. 43 and 44 shows that the advantage of the circular asymmetric lens array 4200 is that the circular asymmetric lens array 4200 allows the optical beam 4202 to be more densely packed without reducing the cross-sectional size of the optical beam, thereby allowing the optical fiber cores 302 to be more densely packed so that more optical fiber cores can be packaged in a given size of optical fiber cableAnd higher data throughput can be achieved for optical communication systems having limited areas (e.g., areas in the front and/or rear panels) allocated for connection to fiber optic cables.

Off-axis beams typically occur when grating couplers are used to couple light from or to Photonic Integrated Circuits (PICs). These devices implemented on PICs can in principle be designed for any emission angle. However, off-axis designs are generally preferred (e.g., emission at 8 degrees) because this allows for reduced reflection by the assembly. For many applications of PICs, low reflection or high return loss of the grating coupler is desirable in order to achieve the desired function.

Fig. 45A is a schematic diagram illustrating an example of a fiber-to-PIC connector 4500 using a circular asymmetric optical lens array 4200. The fiber-to-PIC connector 4500 is used to couple the fiber to a photonic integrated circuit having a grating coupler 4502, the grating coupler 4502 emitting a light beam at an exit angle θ, for example, in the range of 1 ° to 30 ° or about 8 °. When the optical wave signal is transmitted from the optical waveguide to the grating coupler 4502, the reflection may be smaller (compared to the example shown in fig. 45B).

In this example, similar to the example of fig. 43, the distance between adjacent fiber cores 302 is about 250 μm, and the distance between the two nearest adjacent circular asymmetric lenses 4200 is about Each circular asymmetric lens 4200 has a circular circumference (as viewed in the direction of the beam path) with a diameter of about 160 μm, and the geometric center (as viewed in the direction of the beam path) of the circular asymmetric lens 4200 is offset from the optical axis by a distance of about ∈>The radius of the surface curvature of the circularly asymmetric lens 4200 is about 610 μm. The distance between the circularly asymmetric lens 4200 and the focal plane 4214 is about 1mm. The parameter values listed above are merely examples, and other values may be used.

Fig. 45B is a schematic diagram illustrating an example of a fiber-to-PIC connector 4504 using a circularly symmetric optical lens array 4506. The fiber-to-PIC connector 4504 is used to couple the fiber to a photonic integrated circuit having a grating coupler 4508 that emits a light beam vertically (i.e., at an exit angle θ=0). When the lightwave signal is transmitted from the optical waveguide to the grating coupler 4508, the reflection may be greater (as compared to the example shown in fig. 45A).

In this example, the distance between adjacent fiber cores 302 is about 250 μm, and the distance between the two nearest adjacent circularly symmetric lenses 4506 is aboutEach circularly symmetric lens 4506 has a diameter of about 160 μm. The radius of curvature of the surface of the circularly symmetric lens 4506 is about 610 μm. The distance between the circular asymmetric lens 4506 and the focal plane 4214 is about 1mm. The parameter values listed above are merely examples, and other values may be used.

In some embodiments, the end face of the Fiber Array (FA) is polished at an angle (e.g., in the range of 4 ° to 12 ° or about 8 °) to achieve high return loss. Thus, on-axis designs are less preferred when the PIC is interfaced with the fiber array through a micro-optical assembly that includes a lens array and other optical components. On the other hand, achieving an off-axis design of collimated beams with the same diameter using conventional lens arrays limits the density of the array. Thus, a circular asymmetric lens array (e.g., 4200) is a solution that enables dense collimated beam arrays for high return loss interfaces from photonic integrated circuits to fiber arrays.

Fig. 46A and 46B show schematic diagrams demonstrating how a circular asymmetric lens array (e.g., 4200) that may be a free form lens achieves a larger beam diameter d2> d1 than a conventional circular symmetric lens array (e.g., 4208) design with the same beam density.

Fig. 46A and 46B illustrate how a circular asymmetric lens array achieves a larger beam diameter than a design using a conventional circular symmetric lens array with the same beam density.

Fig. 46A is a schematic diagram illustrating an example of a fiber-to-PIC connector 4600 coupling a grating coupler array 4602 to a fiber array 4604, with the end face polished at an angle of about 8 degrees. The fiber-to-PIC connector 4600 includes a first circular asymmetric lens array 4608 and a second circular asymmetric lens array 4610. For example, the surface curvature of the first circular asymmetric lens array 4608 may be the same as the surface curvature of the second circular asymmetric lens array 4610, except that the concave surface of the lens 4608 faces the concave surface of the lens 4610. The first circular asymmetric lens array 4608 receives light 4612 emitted from the grating coupler 4602 at an exit angle of 8 ° and collimates the light 4612 to produce a collimated light beam 4614. The second circular asymmetric lens array 4610 focuses the collimated light beam 4614 toward the fiber core of the fiber array 4604, with the path of the focused light beam 4616 being at an angle of 8 ° relative to the path of the collimated light beam 4614. Thus, the focused beam 4616 is aligned with the fiber cores of the fiber array 4604. The diameter of the collimated beam 4614 is d2.

Fig. 46B is a schematic diagram illustrating an example of a fiber-to-PIC connector 4620 coupling the grating coupler array 4602 to the fiber array 4604 with the end face polished at an angle of about 8 degrees. The fiber-to-PIC connector 4620 includes a first circularly symmetric lens array 4622 and a second circularly symmetric lens array 4624. The surface curvature of the first circularly symmetric lens array 4622 may be the same as the surface curvature of the second circularly symmetric lens array 4624, except that the concave surface of the lens 4622 faces the concave surface of the lens 4624. The first circularly symmetric lens array 4622 receives light 4626 emitted from the grating coupler 4602 at an exit angle of 8 ° and collimates the light 4612 to produce a collimated light beam 4628. Each light beam 4626 intersects only a portion of the corresponding circularly symmetric lens 4622. The second circularly symmetric lens array 4624 focuses the collimated light beam 4628 toward the fiber core of the fiber array 4604, with the path of the focused light beam 4630 being at an angle of 8 ° relative to the path of the collimated light beam 4628. Thus, the focused light beam 4628 is aligned with the fiber cores of the fiber array 4604. The diameter of collimated beam 4628 is d1. Because only portions of circularly symmetric lenses 4622 and 4624 are used to process (e.g., refract) beams 4626 and 4628, the diameter of beam 4628 is smaller, i.e., d1< d2, compared to beam 4614. The increased beam diameter d2 significantly relaxes the assembly tolerances of the micro-optical system, resulting in improved manufacturability.

Referring to fig. 47, in some embodiments, a circular asymmetric lens array may be combined with a conventional lens array (circular symmetric lens array) when a grating coupler array emitting at an angle of 8 degrees on the PIC is interfaced with a fiber array polished at an angle of 0 degrees. For example, the fiber-to-PIC connector 4700 is configured to couple the grating coupler array 4602 to the fiber array 4702 polished at 0 degrees. The fiber-to-PIC connector 4700 includes a first circularly asymmetric lens array 4608 that collimates the light from the grating coupler 4602, and a second circularly symmetric lens array 4624 that focuses the collimated light beam toward the fiber core of the fiber array 4702.

Referring to fig. 48, in some embodiments, two different (mirror-symmetrical) versions of the components of a circular asymmetric lens array are combined to interface a grating coupler array on a PIC that emits at an angle of 8 degrees with a fiber array polished at an angle of-8 degrees. For example, the fiber-to-PIC connector 4800 is configured to couple the grating coupler array 4602 to the fiber array 4802 polished at-8 degrees. The fiber-to-PIC connector 4800 includes a first circular asymmetric lens array 4608 that collimates light from the grating coupler 4602, and a second circular asymmetric lens array 4804 that focuses the collimated light beam toward the fiber core of the fiber array 4802. In this example, the second circularly asymmetric lens array 4804 is mirror symmetric with respect to the first circularly asymmetric lens array 4608.

Referring to fig. 49, in some embodiments, two identical circular asymmetric lens arrays are used for the smart connector assembly as previously described. In this example, the fiber-to-PIC connector 4900 includes a second circular asymmetric lens array 4610 that collimates light emitted from the fiber array 4604 polished at 8 °. The fiber-to-PIC connector 4900 includes an walk-off crystal 4902, which 4902 splits an incident collimated light beam 4904 into two beam components 4906a and 4906b that pass through a birefringent aperture plate 4908 (which may be a half-wave plate with an opening), one of which rotates its polarization by 90 °, such that the two beam components 4906a and 4906b have the same polarization direction. Beam components 4906a and 4906b are focused by circular asymmetric lens array 4608 at the appropriate angle of incidence to grating coupler array 4602.

In this example, the full density of collimated beams is required only for the first lens array 4608 that is closer to the grating coupler array 4602. Only each second lens is used in the second lens array 4610 facing the fiber array 4604. Thus, the second lens array may also be implemented as a conventional circularly symmetric lens array. In some embodiments, the fiber array is implemented with a 0 degree polish, and then the second lens array facing the fiber array is implemented as a conventional circularly symmetric lens array, as shown in the example of fig. 47.

Examples of using Circular Polarization Maintaining Fibers (CPMFs) to transmit light from a laser to a modulator are described below.

Referring to FIG. 54, in some embodiments, an optoelectronic data processing system 5400 includes a laser 5402, a first quarter wave plate 5404, a Circular Polarization Maintaining Fiber (CPMF) 5406, a second quarter wave plate 5408, a modulator 5410, and other data processing modules not shown in the figures. The laser 5402 provides optical energy light to the modulator 5410. The laser 5402 generates linearly polarized light, and the first quarter wave plate 5404 converts the linearly polarized light into circularly polarized light.

While the examples described in fig. 54-60 use one or more circular polarization maintaining fibers as a mechanism to couple a "laser" to a "modulator," the described techniques may generally be used to couple any first device that emits a linear polarization state to any second device that is expected to receive a linear polarization state. The first linearly polarizing means need not be a laser but may be any means of emitting linearly polarized light. The second linear polarization means need not be a modulator but may be any means that preferably accepts linearly polarized light. Examples of the second linear polarization means include: (a) a lithium niobate optical modulator; (b) a vertical grating coupler on the photonic integrated circuit; (c) a modulator integrated on the photonic integrated circuit. For example, the first linear polarization means may be a local oscillator generating a sequence of light pulses and the second linear polarization means may be a coherent light receiver. For example, the first linear polarization means may be a single polarized light emitter and the second linear polarization means may be a single polarized light receiver.

A circularly polarized light maintaining optical fiber (CPMF) 5406 transmits circularly polarized light from the first quarter-wave plate 5404 to the second quarter-wave plate 5408 while maintaining a polarization state of the circularly polarized light. The region between first quarter waveplate 5404 and second quarter waveplate 5408 forms a rotation invariant region. In this context, the term "rotation-invariant region" refers to a region comprising one or more sections of circular polarization-maintaining optical fiber connected to each other and to the first and second quarter-wave plates by optical connections. Because of the circular polarization maintaining properties of circular polarization maintaining fibers, these connections do not have to be rotationally aligned. The second quarter wave plate 5408 converts circularly polarized light into linearly polarized light which is passed to the modulator 5410. The modulator 5410 may be, for example, part of a photonic integrated circuit 2504 (fig. 27) or part of any other optoelectronic module.

An advantage of this design is that the circular polarization maintaining fiber 5406 can be mounted to the first quarter wave plate 5404 and the second quarter wave plate 5408 at any angle of rotation, thereby reducing alignment or packaging costs. As long as the linear polarization of the laser 5402 is properly oriented with respect to the first quarter wave plate 5404 and the second quarter wave plate 5408 is properly oriented with respect to the modulator 5410, the two ends of the circular polarization maintaining fiber 5406 can be randomly rotated without affecting the linear polarization orientation generated after the second quarter wave plate 5408. The critical rotational alignment is not at the end of the fiber, but at the quarter wave plates 5404 and 5408. Because the quarter wave plates 5404 and 5408 are attached to the laser 5402 and modulator 5410, respectively, during device assembly, this may generally be more convenient and/or less expensive than ensuring rotational alignment of the optical fibers at each fiber connector along the way.

Referring to fig. 55, in some embodiments, an optoelectronic data processing system 5500 includes a laser array 5502, a first single quarter wave plate 5504, a plurality of Circular Polarization Maintaining Fibers (CPMFs) 5506, a second single quarter wave plate 5508, a modulator array 5510, and other data processing modules not shown in the figures. The laser array 5502 includes multiple lasers with aligned polarizations and may have the same wavelength or different wavelengths. The laser array 5502 provides optical energy light to the modulator array 5510. Modulator array 5510 includes a plurality of modulators that may be disposed on, for example, a photonic integrated circuit. A Circular Polarization Maintaining Fiber (CPMF) 5506 transmits light from the laser array 5502 to the modulator array 5510, where the light from each laser is transmitted to a corresponding modulator through one of the polarization maintaining fibers 5506.

The first single quarter wave plate 5504 covers all laser outputs and converts the linearly polarized laser outputs into corresponding circularly polarized light. Each of the Circular Polarization Maintaining Fibers (CPMF) 5506 transmits circularly polarized light from the first single quarter-wave plate 5504 to the second single quarter-wave plate 5508. The second single quarter wave plate 5508 converts circularly polarized light from all circularly polarization maintaining fibers 5506 into linearly polarized light that is passed to the corresponding modulator. The second single quarter waveplate 5508 may be edge coupled or vertically coupled to a photonic integrated circuit that includes an array of modulators. The region between the first single quarter waveplate 5504 and the second single quarter waveplate 5508 forms a rotation invariant region.

Each of the lasers in the laser array 5502 generates linearly polarized light. The first laser generates first linearly polarized light and the second laser generates second linearly polarized light. When we say that the first and second lasers have aligned polarizations, by which is meant that the polarization direction of the first linearly polarized light is substantially parallel to the polarization direction of the second linearly polarized light. The term "substantially parallel" is intended to take into account tolerances in the manufacturing and/or assembly process. For example, in some cases, two directions may be "aligned" or "substantially parallel" when the angle between the two directions is within 10 °, or within 5 °, or within 1 °. Similarly, in some cases, two directions may be "substantially orthogonal" when the angle between the two directions is in the range of 80 ° to 100 °, or in the range of 85 ° to 95 °, or in the range of 89 ° to 91 °.

An advantage of this design is that each of the circular polarization maintaining fibers 5506 may be mounted to the first single quarter waveplate 5504 and the second single quarter waveplate 5508 at any angle of rotation, thereby reducing alignment or packaging costs.

Referring to fig. 56, in some embodiments, an optoelectronic data processing system 5600 includes a first photonic integrated circuit 5602 and a second photonic integrated circuit 5604. The first photonic integrated circuit 5602 includes a laser array 5606 and a first separate quarter wave polarization rotator array 5608. The second photonic integrated circuit 5604 includes a second separate quarter wave polarization rotator array 5610 and modulator array 5612. The laser array 5606 provides optical energy light to the modulator array 5612.

A plurality of Circular Polarization Maintaining Fibers (CPMF) 5614 are optically coupled to the first individual quarter-wave polarization rotator array 5608 and the second individual quarter-wave polarization rotator array 5610. The region between the first individual quarter wave polarization rotator array 5608 and the second individual quarter wave polarization rotator array 5610 forms a rotation invariant region. The circular polarization maintaining fiber 5614 may be edge coupled or vertically coupled to the photonic integrated circuit 5604. The laser 5606 generates linearly polarized light that is converted to circularly polarized light by a first separate quarter-wave polarization rotator 5608. The circularly polarized light is transmitted by the circularly polarization maintaining fiber 5614 to a second separate quarter wave polarization rotator 5610 that converts the circularly polarized light into linearly polarized light that is passed to the modulator 5612.

In the above example, the laser array 5606 and the separate quarter wave polarization rotator array 5608 may be integrated on a substrate or module other than a photonic integrated circuit. Similarly, separate quarter wave polarization rotator array 5610 and modulator array 5612 may be integrated on a substrate or module other than a photonic integrated circuit.

The advantage of this design is that the circular polarization maintaining fiber 5614 can be mounted to the first individual quarter wave polarization rotator 5608 and the second photonic integrated circuit 5604 at any rotation angle, thereby reducing alignment or packaging costs.

Examples of using a single circular polarization maintaining fiber to transmit photon supply light to two modulators are described below.

Referring to fig. 57, in some embodiments, an optoelectronic data processing system 5700 includes an optical energy source 5702 and a photonic integrated circuit 5704, wherein a circular polarization maintaining fiber 5706 is optically coupled between the optical energy source 5702 and the photonic integrated circuit 5704. The region between optical energy source 5702 and photonic integrated circuit 5704 forms a rotationally invariant region.

The optical energy source 5702 includes a first laser 5708a and a second laser 5708b. The first laser 5708a generates a first laser light having a first linear polarization and the second laser 5708b generates a second laser light having a second linear polarization. For example, the first linear polarization may be substantially orthogonal to the second linear polarization. The first and second lasers are combined at polarizing beam splitter 5710 to generate a first combined light 5712 having a first component and a second component, wherein the first component has a first linear polarization and the second component has a second linear polarization. The first quarter waveplate 5714 converts the first combined light 5712 into a second combined light 5716, the second combined light 5716 being in this example a circularly polarized combined light. The second combined light 5716 includes a first component having a first circular polarization (e.g., right-hand circular polarization) and a second component having a second circular polarization (e.g., left-hand circular polarization). The first component of the first combined light 5712 (with the first linear polarization) is converted by the quarter wave plate 5714 into the first component of the second combined light 5716 (with the first circular polarization). The second component of the first combined light 5712 (with the second linear polarization) is converted by the quarter wave plate 5714 into the second component of the second combined light 5716 (with the second circular polarization).

The photonic integrated circuit 5704 includes a quarter wave plate 5718, the quarter wave plate 5718 receiving the second combined light 5716 transmitted by the circular polarization maintaining fiber 5706 and converting the second combined light 5716 into a third combined light 5720. The third combined light 5720 includes a first component having a first linear polarization and a second component having a second linear polarization, where the second linear polarization is substantially orthogonal to the first linear polarization. The quarter waveplate 5718 converts the first component of the second combined light 5716 (with the first circular polarization) into the first component of the third combined light 5720 (with the first linear polarization). The quarter waveplate 5718 converts the second component of the second combined light 5716 (having the second circular polarization) into the second component of the third combined light 5720 (having the second linear polarization). Polarizing beam splitter 5722 splits third combined light 5720 into first light having a first linear polarization and second light having a second linear polarization. The first light is sent to a first modulator 5724a and the second light is sent to a second modulator 5724b.

An advantage of this design is that the circular polarization maintaining fiber 5706 can be mounted to the first quarter waveplate 5714 and the second quarter waveplate 5718 at any rotation angle, thereby reducing alignment or packaging costs.

Referring to fig. 58, in some embodiments, an optoelectronic data processing system 5800 includes an optical energy source 5802 and a photonic integrated circuit 5806, wherein a circular polarization maintaining fiber 5706 is optically coupled between the optical energy source 5802 and the photonic integrated circuit 5806. The region between the optical energy source 5802 and the photonic integrated circuit 5806 forms a rotationally invariant region. In optical energy source 5802, quarter wave plate 5804 is positioned between lasers 5708a and 5708b and polarizing beam splitter 5710. In photonic integrated circuit 5806, quarter wave plate 5808 is positioned between polarizing beam splitter 5722 and modulators 5724a and 5724 b. In this example, at optical energy source 5802, linearly polarized light from lasers 5708a and 5708b is converted to circularly polarized light before being combined by polarizing beam splitter 5710. At photonic integrated circuit 5806, polarizing beam splitter 5722 splits light from circular polarization maintaining fiber 5706 into two circularly polarized light rays that are converted by quarter wave plate 5808 into two linearly polarized light rays that are sent to modulators 5724a and 5724 b.

Referring to fig. 59, in some embodiments, an optoelectronic data processing system 5900 includes an optical energy source 5902 and a photonic integrated circuit 5906, with a circular polarization maintaining fiber 5706 optically coupled between the optical energy source 5902 and the photonic integrated circuit 5906. The region between optical energy source 5902 and photonic integrated circuit 5906 forms a rotationally invariant region. The optical energy source 5902 includes a first laser 5708a and a second laser 5708c having the same linear polarization, i.e., the polarization direction of the linearly polarized light generated by the first laser 5708a is substantially parallel to the polarization direction of the linearly polarized light generated by the second laser 5708 b. A polarization rotator 5904 is provided to rotate the polarization of light output from the second laser 5708c by 90 ° before combining with light from the first laser 5708a at the polarization beam splitter 5710.

At photonic integrated circuit 5906, a polarization rotator 5908 is provided to rotate one of the light output from the polarization beam splitter by 90 ° before sending it to modulator 5724 c. Modulator 5724c and modulator 5724a are configured to receive light having the same linear polarization. In contrast, in the examples of fig. 57 and 58, modulators 5724a and 5724b are configured to receive light having different linear polarizations (e.g., polarizations orthogonal to each other).

Referring to fig. 60, in some embodiments, an optoelectronic data processing system 6000 includes an optical energy source 6002 and a photonic integrated circuit 6004, wherein a circular polarization maintaining fiber 5706 is optically coupled between the optical energy source 6002 and the photonic integrated circuit 6004. The region between the optical energy source 6002 and the photonic integrated circuit 6004 forms a rotationally invariant region. At the optical energy source 6002, a quarter wave plate 6006 is positioned upstream of the polarizing beam splitter 5710 and downstream of the polarization rotator 5904 and the laser 5708 a. In photonic integrated circuit 6004, polarizing beam splitter 6008 is positioned downstream of polarizing beam splitter 5722 and upstream of polarization rotator 5908 and modulator 5724 a.

Fig. 61 is a side view of an example of a fiber-to-PIC connector 6100 coupling an input fiber array to a photonic integrated circuit. The fiber-to-PIC connector 6100 enables horizontal fiber attachment, i.e., where the optical fiber extends parallel to the top surface of the photonic integrated circuit. The figure shows a ray-tracing simulation using a design for a silicon lens array coupled to an input fiber array. The fiber-to-PIC connector 6100 enables an incident light beam to be directed at a vertical coupling element on a photonic integrated circuit at an angle of incidence θ1 that maximizes the coupling efficiency of the vertical coupling element. The angle of incidence θ1 can be in a range, for example, from 1 ° to 30 °, from 1 ° to 16 °, or from 4 ° to 12 °, or about 8 °, depending on the design of the vertical coupling element. The vertical coupling element may be, for example, a vertical grating coupler. The fiber-to-PIC connector 6100 also enables an outgoing light beam to be emitted from the vertical coupling element at an angle θ1 and coupled to horizontally oriented fibers of the input fiber array.

In some embodiments, fiber-to-PIC connector 6100 includes a first lens array 6102 having a lens array 6104, a birefringent beam shifting element (or walk-off element) 6106, a birefringent aperture plate 6108, a steering mirror 6110 having a reflective surface 6124, and a second lens array 6112. The first lens array 6102 may be made of, for example, silicon. Each lens in array 6102 is formed of a curved interface of another material having a different refractive index. The input fiber array 6128 is coupled perpendicularly to the first lens array 6102 at an angle of incidence of, for example, 0 °. The turning mirror 6110 can be, for example, a glass prism block.

The light beam 6116 from the input fiber of the fiber array 6128 is collimated by the corresponding lens 6104 in the first lens array 6102. The first lens array 6102 is attached to the birefringent beam displacement element 6106 using, for example, an optical adhesive. The birefringent beam shifting element 6106 splits the collimated beam into a first beam component 6118 having a first polarization and a second beam component 1028 having a second polarization. For example, the second polarization may be rotated 90 ° relative to the first polarization. The birefringent beam shifting element 6106 may be formed of, for example, suitably oriented calcite, yttrium vanadate (YVO ₄ ) Or a-BBO. The birefringent beam displacement element 6106 separates the second beam component 6120 from the first beam component 6118 by a distance.

In some embodiments, birefringent aperture plate 6108 may include regions with birefringent material and regions with openings (or regions without birefringent material), similar to birefringent aperture plate 1006 (fig. 13C), 1402 (fig. 14B), 1502 (fig. 15B), 1602 (fig. 16B), 1702 (fig. 17B), 1802 (fig. 18B), 1902 (fig. 19B), 2002 (fig. 20B), 2100 (fig. 21A), 2104 (fig. 21B), 2108 (fig. 21C), 2116 (fig. 21D), or 3602 (fig. 36). In some embodiments, birefringent aperture plate 6108 may include a region with birefringent material and a region with non-birefringent optically transparent material, similar to patterned birefringent plate 5004 (fig. 50) or 5104 (fig. 51).

The first beam component 6118 passes through an opening in the birefringent aperture plate 6108 or a non-birefringent optically transparent material and retains its polarization. The second beam component 6120 passes through the birefringent material of the birefringent aperture plate 6108, which rotates its polarization by 90 ° and has a polarization parallel to the first beam component 6118.

The reflective surface 6124 is at an angle θ2 relative to the plane of the major surface of photonic integrated circuit 6134. The steering mirror 6110 redirects the first beam component 6118 and the second beam component 6120 toward the photonic integrated circuit 6134. The angle θ2 is also referred to as the turning mirror angle. The first and second beam components 6118, 6120 pass through corresponding lenses of the second lens array 6112, which focus the first and second beam components 6118, 6120 at an incident angle θ1 towards corresponding grating couplers 6122 (only one shown in the figure). For example, θ2=41.2°, and θ1=8°. The angles θ1 and θ2 may also be other values, depending on the design of the grating coupler 6122, to maximize the coupling efficiency of the grating coupler 6122.

For example, the second lens array 6112 may be made of glass. The second lens array 6112 is attached to the lower surface 6126 of the glass prism block of the turning mirror 6110. The lower surface 6126 is a distance f1 from the grating coupler 6122, where f1 is approximately equal to the focal length of the lenses in the second lens array 6112. For example, the turning mirror 6110 includes one or more support portions 6132 defining a pocket 6132, and the turning mirror 6110 is secured to the photonic integrated circuit 6134 by applying glue inside the pocket 6132.

The input fiber array 6128 may include a two-dimensional arrangement (e.g., a two-dimensional array) of input fibers. The first lens array 6102 may have a two-dimensional arrangement of lenses 6104 corresponding to the two-dimensional arrangement of optical fibers in the input fiber array 6128. The photonic integrated circuit may have a two-dimensional arrangement (e.g., a two-dimensional array) of grating couplers. The fiber-to-PIC connector 6100 couples light beams from at least some of the fibers in the input fiber array 6128 to at least some of the grating couplers 6122 on the photonic integrated circuit 6134.

The dimensions of the first lens array 6102, the birefringent beam displacement element 6106, and the turning mirror 6110 depend on, for example, the size of the input fiber array 6128, the size of the area occupied by the grating couplers 6122, and the pitch between the grating couplers 6122. The thickness of the first lens array 6102 is t1, and the thickness of the birefringent beam displacement element 6106 is t2. For example, t1 may be about 1.4mm, and t2 may be about 1.8mm. For example, the air gap f1 may be about 0.4mm. Increasing t1 will increase the diameter of collimated beam components 6118 and 6120. Increasing t2 will increase the walk-off distance between the first beam component 6118 and the second beam component 6120. The distance between the pair of grating couplers receiving the first beam component 6118 and the second beam component 6120 is d1. A larger d1 corresponds to a larger t2, and conversely, a smaller d1 corresponds to a smaller t2.

For ease of illustration, in the example of fig. 61, the input fiber extends in the z-direction at the location where the input fiber is coupled to the first lens array 6102. For example, input fiber array 6128 includes rows and columns of fibers, with the row direction extending in the x-direction and the column direction extending in the y-direction. The input fiber array 6128 has a dimension in the y-direction of w1, which may be, for example, 2.75mm. The distance w1 is the distance between the optical axis of the core of the input fiber designated at the top row and the optical axis of the core of the input fiber positioned at the bottom row. The input fiber array 6128 has a footprint measured along a plane parallel to the x-y plane. The grating coupler has a footprint measured along a top surface of photonic integrated circuit 6134, where the top surface is parallel to the x-z plane. For example, the size of the footprint of the input fiber array 6128 in the y-direction generally corresponds to the size of the footprint of the grating coupler in the z-direction, and the size of the footprint of the input fiber array 6128 in the x-direction generally corresponds to the size of the footprint of the grating coupler in the x-direction.

In the example of fig. 61, the turning mirror angle (e.g., 41.2 °) is optimized such that the light beam propagates at a non-zero angle relative to the optical axis of the lenses of the second lens array 6112, such that the light beam is coupled to the grating coupler 6122 on the photonic integrated circuit 6134 at an optimized angle of, for example, 8 °. This allows for a larger beam diameter than using a combination of off-axis beams with conventional lenses. The birefringent beam displacement element 6106 may be placed at different positions in the propagation path of the beam. In some embodiments, the steering mirror 6110 and the second lens array 6112 are made from a single glass or silicon block. In some examples, the second lens array 6112 is fabricated separately from the turning mirror 6110 and attached to the turning mirror 6110.

Fig. 62 shows a fiber-to-PIC connector 6100 coupled between an input fiber array 6128 and a photonic integrated circuit 6134. The figure shows the propagation paths of a first input light beam 6202 from a first input optical fiber 6204 and a second input light beam 6206 from a second input optical fiber 6208. The first input optical fiber 6204 and the second input optical fiber 6208 are part of an input optical fiber array 6128. The first input beam 6202 is split into two beam components coupled to a first pair of grating couplers 6212. The second input beam 6206 is split into two beam components coupled to a second pair of grating couplers 6214. In the opposite direction, the light beams output from the first pair of grating couplers 6212 are coupled to the first input optical fiber 6204, and the light beams output from the second pair of grating couplers 6214 are coupled to the second input optical fiber 6208.

Fig. 63 shows an example of a fiber-to-PIC connector 6300 coupled to a photonic integrated circuit 6134. The fiber-to-PIC connector 6300 includes a first lens array 6302 having a two-dimensional arrangement (e.g., a two-dimensional array) of glass lenses 6304. The fiber-to-PIC connector 6300 includes a birefringent beam-shifting element (or walk-off element) 6106, a birefringent aperture plate 6108, a steering mirror 6110, and a second lens array 6112, similar to those of the fiber-to-PIC connector 6100. An air gap 6306 is provided between the first lens array 6104 and the birefringent beam displacement element 6106. The first lens array 6104 is attached to the birefringent beam displacement element 6106 by a coupling element 6308 at the edge of the first lens array 6104 using, for example, an optical adhesive. The thickness of the first lens array 6104 is t3, where t3 may be, for example, 0.6mm.

The figure shows the propagation paths of a first input light beam 6202 from a first input optical fiber 6204 and a second input light beam 6206 from a second input optical fiber 6208. The first input optical fiber 6204 and the second input optical fiber 6208 are part of an input optical fiber array 6128. The first input beam 6202 is split into two beam components coupled to a first pair of grating couplers 6212. The second input beam 6206 is split into two beam components coupled to a second pair of grating couplers 6214. In the opposite direction, the light beams output from the first pair of grating couplers 6212 are coupled to the first input optical fiber 6204, and the light beams output from the second pair of grating couplers 6214 are coupled to the second input optical fiber 6208.

One advantage of the silicon lens in the fiber-to-PIC connector 6100, comparing the examples shown in fig. 62 and 63, is that glue may be used directly over the lens due to the high refractive index of silicon. Because glass and glue can have similar refractive indices, an air gap is provided for the glass lens to maintain lens performance. Glass lenses may have other advantages, for example, because glass lenses are transparent to visible light, assembly becomes easier. The first lens array 6102 includes a small silicon lens array formed on a silicon block, and the first lens array 6302 includes a small glass lens array formed on a glass block. The silicon block and glass block shown in fig. 62 and 63 have different thicknesses due to different optical path lengths in materials having different refractive indices. The dimensions in these figures are selected such that the optical assembly has a magnification of about 1:1 for imaging the fiber mode to the grating coupler on the photonic integrated circuit.

Fig. 64 shows an example of a fiber-to-PIC connector 6400 coupled between an input fiber array 6128 and a photonic integrated circuit 6134. The fiber-to-PIC connector 6400 includes a first lens array 6302, a turning mirror 6110, a birefringent beam shifting element 6402, a birefringent aperture plate 6426, and a second lens array 6404. In this example, the first lens array 6302 is coupled to a first face 6406 of a turning mirror 6110. The second face 6408 of the turning mirror 6110 is coupled to the first face 6422 of the birefringent beam shifting element 6402. An air gap 6420 is provided between the lenses of the first lens array 6302 and the first face 6406 of the birefringent beam shifting element 6402. The birefringent beam shifting element 6402 has a second face 6424 coupled to a birefringent aperture plate 6426. The birefringent aperture plate 6426 is positioned between the second face 6424 of the birefringent beam shifting element 6402 and the second lens array 6404. The birefringent aperture plate 6426 may include a region having a birefringent material, and a region having an opening (either a region without a birefringent material or a region with a non-birefringent optically transparent material), similar to the birefringent aperture plate 6108 of fig. 61 to 63.

An air gap 6418 is provided between the birefringent aperture plate 6426 and the lenses of the second lens array 6404. For example, the second lens array 6404 includes a glass lens array formed on a glass block. The thickness of the glass block is t4, which may be, for example, 0.6mm. For example, the thickness of the glass block for the first lens array 6302 may be substantially the same as the thickness of the glass block for the second lens array 6404.

For example, the first input light beam 6202 from the first input optical fiber 6204 is redirected by the turning mirror 6110 toward the birefringent light beam shifting element 6402, the birefringent light beam shifting element 6402 separating the first input light beam 6202 into a first light beam component 6410 and a second light beam component 6412. After passing through the birefringent beam shifting element 6402, the second beam component 6412 has a polarization of, for example, 90 ° with respect to the polarization of the first beam component 6410. The birefringent aperture plate 6426 rotates the polarization of the second beam component 6412 by 90 °. After passing through the birefringent aperture plate 6426, the first and second beam components 6410, 6412 have parallel polarization. Corresponding lenses in the second lens array 6404 focus the first and second beam components 6410, 6412 to the first pair of grating couplers 6212 at an angle of incidence of, for example, 8 °. The second input light beam 6206 from the second input optical fiber 6208 is redirected by the turning mirror 6110 towards the birefringent light beam shifting element 6402, the birefringent light beam shifting element 6402 separating the second input light beam 6206 into a first light beam component 6414 and a second light beam component 6416 having, for example, orthogonal polarizations. After passing through the birefringent aperture plate 6426, the first and second beam components 6414, 6416 have parallel polarization. Corresponding lenses in the second lens array 6404 focus the first beam component 6414 and the second beam component 6416 to the second pair of grating couplers 6214 at an angle of incidence of, for example, 8 °. The light beams output from the first pair of grating couplers 6212 are coupled to a first input optical fiber 6204, and the light beams output from the second pair of grating couplers 6214 are coupled to a second input optical fiber 6208.

The following describes a mechanism for accurately aligning an optical fiber array with a photonic integrated circuit to improve optical coupling efficiency between the optical fiber array and the photonic integrated circuit. A new assembly method is used in which the final critical alignment and bonding is actively performed using a ferrule frame (ferruleframe) bonded to the optical subassembly stack. The ferrule frame is a high precision component that serves as a connection interface for the fiber array.

FIG. 65 shows an example of a co-packaged optical module 6500, where the co-packaged optical module 6500 may be part of an optoelectronic device or system, such as the communications device 101 ₁ To 101 ₆ Any one of them. The co-packaged optical module 6500 includes a photonic integrated circuit 6502, the photonic integrated circuit 6502 having vertical coupling elements 6504 disposed along a major surface 6506 of the photonic integrated circuit 6502. The co-packaged optical module 6500 includes a fiber-to-photonic integrated circuit optical subassembly 6508 for coupling light from the fiber array 6510 to the photonic integrated circuit 6502. The figure shows the optical path of a single light beam between a single vertical coupling element 6504 and a single core 6532 in optical fiber 6510. There may be multiple cores 6532 and multiple vertical coupling elements 6504. The plurality of cores and/or the plurality of vertical coupling elements may be arranged in a 1D array or a 2D array.

In this example, optical subassembly 6508 includes first lens array 6512, beam shifter 6514, half-wave plate 6516, and second lens array 6518. The figure shows one of the lenses in the first lens array 6512 and one of the lenses in the second lens array 6518.

Simulation of the coupling performance of this arrangement at 1550nm and 1310nm signal wavelengths shows that low losses can be achieved when the component is positioned accurately, for example with sub-micron accuracy. Different alignment accuracy may be required for different configurations of the optical fibers and optical subassemblies. A combination of active and passive alignment methods may be used to achieve the desired alignment accuracy.

Referring to fig. 66, the left portion of the figure shows a side view of a co-packaged optical module 6500, where the co-packaged optical module 6500 can achieve low coupling loss between the fiber array 6510 and the vertical coupling element 6504 (including, for example, 6504a and 6504b shown in the figure). The right-hand portion of the figure shows a top view of each component of the co-packaged optical module 6500, shown in positional order in the optical subassembly. For example, the vertical coupling element 6504 may be a grating coupler. The figure shows one of many stacking arrangements that may be used, where different stacking arrangements may have different stacking components depending on the application.

In some embodiments, photonic integrated circuit 6502 may be similar to photonic integrated circuit 210 of fig. 12, and vertical coupling element 6504 may be similar to vertical coupling element 231. Each vertical coupling element 6504 may comprise, for example, a single polarization vertical grating coupler, turning mirror, polarization diversity vertical grating coupler, vertical cavity surface emitting laser, surface normal modulator, photodiode, or any combination thereof. Second lens array 6518 may include lens array 6524. For example, second lens array 6518 may be similar to lens array 541, lens array 542, or a combination of lens array 541 and lens array 542. Half-wave plate 6516 may be similar to birefringent aperture plate 1006. For example, half-wave plate 6516 may be made of a birefringent material with holes 6526 filled with a non-birefringent medium, such as air, epoxy, or the like. Light passing through the birefringent material changes the polarization state, while light passing through the aperture does not change the polarization state. The beam shifter 6514 may be similar to the walk-off crystal 1004, the walk-off crystal 1004 splitting an incident beam into a corresponding pair of outgoing beams comprising respective light of two orthogonal polarization states of the incident beam. The first lens array 6512 may have an array of lenses 6528, and the lenses 6528 may be similar to the lenses 551. Fiber array 6510 may include an array of cores 6530, and cores 6530 may be similar to cores 302 and/or 1034.

In some embodiments, the light beam 6520 from the fiber core 6530 is projected by a lens in the first lens array 6512 to the beam shifter 6514, the beam shifter 6514 splitting the light beam into a first light beam component 6522a having an "x" polarization and a second light beam component 6522b having a "y" polarization, similar to the example shown in fig. 12. For example, the "x" polarization may be TE polarization and the "y" polarization may be TM polarization. Grating couplers 6504a and 6504b couple first and second beam components 6522a and 6522b to photonic integrated circuit 6502. In examples where the vertical coupling element 6504 comprises a single polarization vertical grating coupler, the half-wave plate 6516 may have spatially varying birefringence configured to convert incident polarized light into a grating coupler polarization state, e.g., having a polarization state that maximizes the coupling efficiency of the grating coupler.

As shown in fig. 67A and 67B, each of the optical subassemblies 6508 may have 6 degrees of mechanical motion, and the various subassemblies may not be perfectly aligned, e.g., the end faces of the subassemblies may not be perfectly parallel to each other. In this example, there are 4 optical elements between the fiber array 6510 and the photonic integrated circuit 6502, each having 6 degrees of freedom, which creates a large assembly space that can cause high photonic integrated circuit-to-fiber coupling losses. In addition, each optical element has manufacturing tolerances that may cause loss variations. Because the optical stack is constructed with components bonded together, passive alignment (e.g., vision system alignment device features or fiducial points) and active alignment (e.g., measuring the coupled light from the fiber to the photonic integrated circuit) can be used to position the components prior to bonding. The final alignment and bonding steps are probably the most critical in terms of positioning and angular assembly tolerances.

Fig. 68 shows an example of a process 6540 for assembling an optical stack 6544 comprising a photonic integrated circuit 6502, an optical subassembly 6508, and a ferrule frame 6542. Fig. 69 shows a top view of the ferrule frame 6542. This example is a new assembly method in which the final critical alignment and bonding is actively performed using ferrule frames 6542 bonded to the optical subassembly stack. Ferrule frame 6542 is a high precision component that interfaces with the connection of fiber array 6510. Ferrule frame 6542 is designed such that fiber array 6510 can be removably attached to ferrule frame 6542 and aligned to optimize the efficiency of light transfer between photonic integrated circuit 6502 and fiber cores 6530.

For example, process 6540 includes assembling optical subassembly 6508, attaching optical subassembly 6508 to photonic integrated circuit 6502, removably attaching fiber array 6510 to ferrule frame 6542, aligning ferrule frame 6542 with optical subassembly 6508 using an active alignment process, securely attaching ferrule frame 6542 to optical subassembly 6508 after the active alignment process is completed, and removing fiber array 6510 from ferrule frame 6542.

The process 6540 shown in fig. 68 is suitable for assembling an optical subassembly 6508 using the assembly shown in fig. 66. In some embodiments, the optical subassembly may have components other than those shown in fig. 66, and process 6540 may be modified accordingly.

Referring to fig. 68, in some embodiments, process 6540 includes a first step in which a lower surface of half-wave plate 6516 is attached or bonded to an upper surface of second lens array 6518. Arrow 6548 shows the location where glue or bonding material is placed to securely attach half-wave plate 6516 to second lens array 6518. In a second step, the lower surface of beam shifter 6514 is attached or bonded to the upper surface of half-wave plate 6516. Arrow 6550 shows the location where glue or bonding material is placed to securely attach beam shifter 6514 to half-wave plate 6516.

In a third step, a lower surface of the first lens array 6512 is attached or bonded to an upper surface of the beam shifter 6514. Arrow 6552 shows the location where glue or bonding material is placed to securely attach first lens array 6512 to beam shifter 6514. In the fourth step, the lower surface of the second lens array 6518 is attached or bonded to the main surface (upper surface) of the photonic integrated circuit 6502. Arrow 6554 shows the location where glue or bonding material is placed to securely attach second lens array 6518 to photonic integrated circuit 6502.

Steps 1 through 4 above include steps until the optical subassembly 6508 is attached to the photonic integrated circuit 6502. The order of stacking the components up to step 4 may vary. For example, the bottom surface of first lens array 6512 may be attached to the upper surface of beam shifter 6514, followed by the attachment of the bottom surface of beam shifter 6514 to the upper surface of half-wave plate 6516.

In some embodiments, different optical elements may be used to construct the optical stack. For example, element 6516 may not be present and element 6514 may be a non-birefringent optical element, such as a spacer block made of glass or silicon.

In some embodiments, the fiber array 6510 is attached to a fiber optic connector 6546, the fiber optic connector 6546 being configured to be removably attached to the ferrule frame 6542. Process 6540 includes a fifth step in which ferrule frame 6542 is positioned between fiber optic connector 6546 and optical subassembly 6508.

In a sixth step, the fiber optic connector 6546 is removably attached to the ferrule frame 6542 and laterally aligned using, for example, one or more alignment pins 6547. For example, alignment pins 6547 may be similar to alignment pins of SMP/MTP connectors. The structure is held together mechanically by, for example, one or more clamps, a snap-in mechanism as disclosed in U.S. patent publication US2022/0159860 (the entire contents of which are incorporated herein by reference), or a temporary/removable adhesive. In a seventh step, the ferrule frame 6542 is placed over the optical subassembly 6508 and the ferrule frame 6542 (with the fiber optic connector 6546 attached) is aligned relative to the optical subassembly 6508 using an active alignment process. In some examples, the ferrule frame 6542 defines an opening 6560 (see fig. 69), the opening 6560 being slightly larger than the outer circumference of the optical subassembly 6508. In a seventh step, ferrule frame 6542 is moved toward photonic integrated circuit 6502, wherein an upper portion of optical subassembly 6508 extends into opening 6560 until at least a portion of the bottom surface of element 6546 and/or the end face of fiber array 6510 is in contact with the upper surface of optical subassembly 6508. In the example shown in fig. 68 and 69, the ferrule frame 6542 defines two alignment holes 6562, the two alignment holes 6562 enabling two alignment pins to laterally align the fiber optic connector 6546 with the ferrule frame 6542.

In some embodiments, the active alignment process uses optical loopback (e.g., two grating couplers connected by an optical waveguide) on photonic integrated circuit 6502 to enable reference light to be injected into one fiber and looped-back light to be collected in the other fiber. For example, the active alignment process includes providing light from a first core 6530a of the fiber array 6510 and passing the light through the optical subassembly 6508 to the first vertical coupling element 6504. Light is transmitted from the first vertical coupling element 6504a to the second vertical coupling element 6504 through, for example, an optical waveguide on the photonic integrated circuit 6502. Light is transmitted from the second vertical coupling element 6504b through the optical subassembly 6508 to the second core 6530b of the fiber array 6510. The light received at the second core 6530b is measured and the position and/or orientation of the ferrule frame 6542 relative to the optical subassembly 6508 is adjusted to optimize the efficiency of light transfer between the cores 6530a, 6530b and the photonic integrated circuit 6502. For example, the first and second vertical coupling elements 6504a and 6504b may be the unused transmission grating couplers 2210a and 2210b of fig. 22. For example, the waveguide optically coupling the first vertical coupling element 6504a to the second vertical coupling element 6504b may be the waveguide 2212.

In an eighth step, the fiber connectors 6546 are removed from the ferrule frame 6542, leaving a finished optical stack 6544. When the communication device 101 ₁ To 101 ₆ When deployed in the field, as in a data center, the optical stack 6544 is ready to receive the fiber optic connectors 6546. This allows an operator to quickly connect the fiber optic cable to the communication device while accurately aligning the core to achieve efficient optical transfer between the core and the photonic integrated circuit.

Steps 5 to 7 above include final active alignment and bonding. The red arrow 6556 below the ferrule frame 6542 in step 7 indicates the location where the glue/bonding material is placed to securely attach the ferrule frame 6542 to the optical subassembly 6508 once optimal alignment is achieved. For example, the ferrule frame 6542 may be made of glass, metal, or plastic. For example, the ferrule frame 6542 may be made of a material that is transparent or translucent to Ultraviolet (UV) light, and a UV curable adhesive may be used to bond the ferrule frame 6542 to the optical subassembly 6508.

For example, in step 7, the position of ferrule frame 6542 may be adjusted along a plane substantially parallel to major surface 6506 of photonic integrated circuit 6502. The position of ferrule frame 6542 may be adjusted along the x-axis relative to major surface 6506 and/or along the y-axis relative to major surface 6506. Ferrule frame 6542 is rotatable about a z-axis relative to major surface 6506. In this example, the x-axis and y-axis are substantially parallel to major surface 6506, and the z-axis is substantially perpendicular to major surface 6506.

In some embodiments, adjusting the positioning of the ferrule frame 6542 relative to the optical subassembly 6508 may include (i) adjusting the distance of the end of the fiber optic connector 6546 relative to the optical subassembly 6508, and/or (ii) adjusting the tilt angle of the end face of the fiber optic connector 6546 relative to the optical subassembly 6508. For example, the precision of the alignment of the ferrule frame 6542 relative to the optical subassembly 6508 may be at least 10 μm accuracy, at least 1 μm accuracy, or at least 0.1 μm accuracy. In this context, a precision of at least 0.1 μm accuracy means that when ferrule frame 6542 is bonded to optical sub-module 6508, ferrule frame 6542 can be positioned within 0.1 μm of the optimal positioning to optimize light transfer between core 6530 and photonic integrated circuit 6502.

For example, the fiber array may include at least 10 cores, at least 50 cores, or at least 100 cores. Because cores 6530 are densely packed together, accurate alignment of cores 6530 with respect to optical subassembly 6508 is important for achieving efficient optical transfer between the cores and photonic integrated circuit 6502.

The mechanism that enables the fiber optic connector 6546 to be removably attached to the ferrule frame 6542 is made with high precision. In examples where the fiber optic connector 6546 is removably connected to the ferrule frame 6542 using alignment pins, the alignment pins and alignment holes are made with high precision such that the fiber optic connector 6546 can be attached to the ferrule frame 6542 with a precision in the range of, for example, 10nm or 100nm or 1 μm or 0.1 μm. In this context, an accuracy of 10nm means that the positioning of the fiber optic connector 6546 will always be within 10nm relative to the optimal positioning of the fiber optic connector 6546 as determined by the active alignment process in step 7 of process 6540 whenever the fiber optic connector 6546 is removably attached to the ferrule frame 6542.

In some embodiments, the fiber optic connector 6546 includes a first lens array 6512. In this example, optical subassembly 6508 includes beam shifter 6514, half-wave plate 6516, and second lens array 6518.

In the example shown in fig. 68, the lower surface of the fiber optic connector 4146 is substantially flush with the upper surface of the ferrule frame 4142. In some examples, the fiber optic connector 4146 may be designed such that a portion of the fiber optic connector 4146 extends into the opening 4160 of the ferrule frame 4142.

The process 6540 shown in fig. 68 may be performed manually by an operator or may be performed automatically by a computer-controlled machine. For example, the machine may include one or more robotic arms (or one or more holders holding the assembly, and one or more motor drives driving the one or more holders) that hold and move the various assemblies at the proper locations and orientations. The machine may include one or more sensors (e.g., cameras), pressure sensors, or distance sensors that capture images or video of the components during process 6540. The computer may implement machine vision to identify objects and ensure that each component is properly attached to the other components. The machine may include one or more laser sources that provide laser light during the active alignment process, such as step 7 in fig. 68. The machine may include one or more laser detectors that detect the return laser light during the active alignment process. The computer may implement an algorithm to process data derived from the detection signals generated by the one or more laser detectors and identify the optimal alignment positioning of the ferrule frame. In some embodiments, process 6540 is partially automated, wherein one or more steps are performed manually by an operator and one or more steps are performed automatically by a machine.

Techniques that enable a fiber array (e.g., 6510) to be removably attached to a ferrule frame (e.g., 6542) and aligned to optimize the efficiency of light transfer between a photonic integrated circuit (e.g., 6502) and a core (e.g., 6530) of the fiber array may also be applied to an optoelectronic device (e.g., 3300 of fig. 33A and 33B) having a fiber-to-PIC connector with a portion of an optical axis at an angle substantially different from a top surface normal of the photonic integrated circuit. The substantially different angle may be an angle between 10 degrees and 90 degrees from a top surface normal of the photonic integrated surface. If a portion of the optical axis is angled at about 90 degrees from the top surface normal, the portion of the optical axis is substantially parallel to the top surface of the photonic integrated circuit.

Fig. 71A and 71B are top and side views, respectively, of an example of an optoelectronic device 7100 in which the fiber-to-PIC connector has a portion parallel to the optical axis of the top surface of photonic integrated circuit 6502. The second lens array 6518 couples the light beam propagating in a direction perpendicular to the top surface of the PIC 6502, and the turning mirror 7102 changes the propagation direction of the light beam. The turning mirror 7102 has a reflective surface that reflects the light beam and redirects the light path between the optical fiber and the vertical coupling element on the photonic integrated circuit. In some embodiments, a birefringent aperture plate 7104 (which may be a half-wave plate with an aperture) is attached to the turning mirror 7102, and in some embodiments, walk-off crystal 7106 is attached to the birefringent aperture plate 7104. In some other embodiments, a non-birefringent element (e.g., a glass block) may be used in place of walk-off crystal 7106. The first lens array 7108 is attached to walk-off crystal 7106. The first lens array 7108, walk-off crystal 7106, birefringent aperture plate 7104, turning mirror 7102, and second lens array 6518 form an optical subassembly 7110. In some embodiments, birefringent aperture plate 7104 may not be present and walk-off crystal 7106 may be made of non-birefringent material such as glass or silicon.

A ferrule frame 7112 is provided, wherein the ferrule frame 7112 defines an opening 7114 (see fig. 71C), the opening 7114 being slightly larger than the outer circumference of the front portion of the optical subassembly 7110. The front portion of the optical subassembly 7110 extends into the opening 7114 of the ferrule frame 7112 and the ferrule frame 7112 is securely attached to the front portion of the optical subassembly 7110 by the application of epoxy, glue, or other bonding material. In fig. 71A and 71B, a first lens array 7108 is covered by a ferrule frame 7112 and is not visible in the figures. In the example shown in fig. 71A-71C, the ferrule frame 7112 defines two alignment holes 7116, the two alignment holes 7116 enabling alignment of the two alignment pins 7120 of the fiber optic connector 6546 with the ferrule frame 7112. The photonic integrated circuit 6502, the optical subassembly 7110, and the ferrule frame 7112 form an optical stack 7118 that is ready to receive the fiber optic connector 6546 when the optoelectronic device 7100 is deployed in the field (e.g., in a data center).

In the example shown in fig. 71A and 71B, the turning mirror 7102 is a right angle prism that turns the optical path by 90 °. In some examples, the turning mirror turns the optical path by an angle θ1, where 20+.θ1+.ltoreq.90 °. The longitudinal axis of walk-off crystal 7106 may be at an angle θ2 relative to the major surface of photonic integrated circuit 6502, where 0+.θ2+.ltoreq.70 °. This allows the fiber array to be oriented at an angle θ2 relative to the major surface of photonic integrated circuit 6502 when fiber connector 6546 is attached to ferrule frame 7112.

In some examples, the turning mirror turns the optical path by an angle θ1, where 90+.θ1+.160 °. The longitudinal axis of walk-off crystal 7106 may be at an angle θ2 relative to the major surface of photonic integrated circuit 6502, where-70+.θ2+.0°. This allows the fiber array to be oriented at an angle θ2 relative to the major surface of photonic integrated circuit 6502 when fiber connector 6546 is attached to ferrule frame 7112.

In some examples, there may be two or more turning mirrors or reflective surfaces that redirect the optical path two or more times.

Fig. 72 shows an example of a top view of a process 7200 for assembling an optical stack 7118 including a photonic integrated circuit 6502, an optical subassembly 7110, and a ferrule frame 7112. In step 1 of process 7200, optical subassembly 7110 is assembled by attaching the first surface of turning mirror 7102 to the upper surface of second lens array 6518, attaching the first surface of birefringent aperture plate 7104 to the second surface of turning mirror 7102, attaching the first surface of walk-off crystal 7106 to the second surface of birefringent aperture plate 7104, and attaching the first surface of first lens array 7108 to the second surface of walk-off crystal 7106. The optical subassembly 7110 is attached to a photonic integrated circuit 6502. In some embodiments, the birefringent aperture plate 7104 may not be present and a first surface of walk-off crystal 7106 may be attached to a second surface of steering mirror 7102. In some embodiments, the walk-off crystal may be replaced by a block of material made of a non-birefringent material (e.g., glass or silicon). Step 1 of process 7200 corresponds to steps 1 to 5 of process 6540 of fig. 68.

In step 2 of process 7200, fiber optic connector 6546 is removably attached to ferrule frame 7112 using alignment pins 7120. The alignment pin 7120 and the alignment hole 7116 are made with high precision such that the fiber optic connector 6546 can be attached to the ferrule frame 7112 with a precision in the range of, for example, 10nm or 100nm or 1 μm or 0.1 μm. Step 2 of process 7200 corresponds to step 6 of process 6540 of fig. 68.

In step 3 of process 7200, the ferrule frame 7112 is aligned with the optical subassembly 7110 using an active alignment process. The active alignment process performed in step 3 of process 7200 is similar to the alignment process performed in step 7 of process 6540 of fig. 68.

In step 4 of process 7200, after the active alignment process is completed, the ferrule frame 7112 is securely attached to the optical subassembly 7110 and the fiber connectors 6546 are removed from the ferrule frame 7112, thereby completing the assembly of the optical stack 7118. Step 4 of process 7200 corresponds to step 8 of process 6540 of fig. 68.

The process 7200 shown in fig. 72 is suitable for assembling the optical stack 7118 using the components shown in fig. 66 and the turning mirror shown in fig. 33A and 33B. In some embodiments, the optical subassembly may have components other than those shown in fig. 33A, 33B, and 66, and process 7200 may be modified accordingly.

In some embodiments, the process 7200 includes adjusting the position of the ferrule frame 7112 relative to the optical subassembly 7110, similar to the position of the ferrule frame 6542 relative to the optical subassembly 6508 described above. For example, the precision of the alignment of the ferrule frame 7112 relative to the optical subassembly 7110 may be at least 10 μm accuracy, at least 1 μm accuracy, or at least 0.1 μm accuracy.

Ferrule frame 7112 (fig. 71A-71C) need not be combined with walk-off crystals and/or half-wave plates. The ferrule frame 7112 may also be more generally used in an optical assembly or optical stack in which the walk-off crystal and half-wave plate are replaced by glass or silicon spacers (labeled (a) and (B) as shown in fig. 73-75 below), which may or may not have birefringent properties. As shown in fig. 73 to 75 below, turning mirrors may be implemented at various positions in the optical path.

Referring to fig. 73, in some examples, the optical stack 7300 includes a glass or silicon spacer 7302 positioned between the ferrule frame 7112 and the turning mirror 7304, wherein the glass or silicon spacer 7302 may or may not have birefringent properties. Steering mirror 7304 is attached to lens array 7306, and lens array 7306 is attached to photonic integrated circuit 7308.

Referring to fig. 74, in some examples, the optical stack 7400 includes a first glass or silicon spacer 7302 positioned between the ferrule frame 7112 and the turning mirror 7304, and a second glass or silicon spacer 7402 positioned between the turning mirror 7304 and the lens array 7306. Each of the glass or silicon spacers 7302 or 7402 may or may not have birefringent properties. In both optical stacks 7300 and 7400, the glass or silicon spacer 7302 is longer than the glass or silicon spacer 7402.

Referring to fig. 75, in some examples, the optical stack 7500 includes a first glass or silicon spacer 7502 positioned between the ferrule frame 7112 and the turning mirror 7304, and a second glass or silicon spacer 7504 positioned between the turning mirror 7304 and the lens array 7306. Each of the glass or silicon spacers 7502 or 7504 may or may not have birefringent properties. In this example, the glass or silicon spacer 7502 is shorter than the glass or silicon spacer 7504.

Fig. 46A, 46B, and 48 show examples of fiber arrays (e.g., 4604, 4802) having end faces polished at an angle of, for example, about 8 °. In these examples, the lenses (e.g., 4604, 4608, 4610, 4622, 4624) are adapted such that the fiber-to-PIC connectors output the light beams to the fiber array at an appropriate angle compatible with the angled end faces of the fiber array. The following describes a tilted fiber array ferrule adapter that may be used to adapt between any combination of different bevels between a fiber array connector and an optical assembly.

MPO connectors are commonly used to mate fiber optic cable connectors with fiber optic array cable connectors. The connector end faces of the two connector ferrules may be angled to reduce optical back reflection relative to the fiber end faces.

Fig. 70 is a schematic diagram of an example angled fiber array ferrule adapter 7000 coupled between an angled MPO-like fiber array ferrule of an MPO-like fiber connector 7002 and an optical assembly 7004. The MPO-like optical fiber connector 7002 is connected to a plurality of optical fibers 7024. The optical assembly 7004 may include, for example, the photonic integrated circuit 7010 and the fiber-to-PIC connector 7012 described above. For example, the fiber-to-PIC connector 7012 may include a first lens array 7014, a birefringent beam shifting element 7016, and a second lens array 7018. The optical assembly 7004 may generate and receive an output light beam to and from the MPO-like fiber connector 7002 through the tilted fiber array ferrule adapter 7004.

The MPO-like fiber connector 7002 has an inclined end 7006, and the inclined fiber array ferrule adapter 7000 has an end 7008 that is also inclined to the same angle as the end 7006 of the MPO-like fiber connector ferrule. The angled fiber array ferrule adapter 7000 may be made of nominally the same ferrule type as the MPO-like fiber connector 7002, with the fibers inserted into the ferrule holes and polished down to be nominally flush on both optical faces of the angled fiber array ferrule adapter 7000. The angled fiber array ferrule adapter 7000 enables a low loss connection between the MPO-like fiber connector 7002 and the optical assembly 7004 whose mating surfaces are not angled at the proper angle to mate directly with the MPO-like fiber connector 7002. In some embodiments, the fiber array ferrule adapter 7000 is permanently bonded to the optical subassembly 7004 at the interface 7020.

The MPO-like connector has optical fibers inserted into the ferrule bore and polished down to be nominally flush on the optical face of the MPO-like connector, similar to the optical fibers of a standard MPO connector. The MPO-like connector may have alignment pins or holes similar to those of a standard MPO connector. The cross-sectional profile of the MPO-like connector ferrule may be different from the cross-sectional profile of a standard MPO connector to accommodate various arrangements of optical fibers. For example, MPO-like connectors may be used to connect to fiber optic cables having a two-dimensional arrangement of optical fibers, such as a two-dimensional array of optical fibers. The two-dimensional array of optical fibers may have a plurality of rows and columns of optical fibers, for example, at least 2 rows and at least 4 columns of optical fibers, or at least 2 rows and at least 8 columns of optical fibers, or at least 3 rows and at least 8 columns of optical fibers.

The angled fiber array ferrule adapter 7000 may also be configured to couple between the angled MPO fiber array ferrule of a standard MPO fiber connector and an optical assembly whose output beam is configured to be compatible with the standard MPO fiber connector.

Although the figure shows an optical assembly 7004 having a 0 ° angled optical face, the fiber array ferrule adapter 7000 may be used to fit between any combination of different angled faces between the fiber array connector (e.g., 7002) and the optical assembly (e.g., 7004).

In some examples, the fiber array ferrule adapter 7000 is a male connector with alignment pins 7022 that mate with alignment holes in the angled MPO connector 7002. In some examples, the MPO connector 7002 is a male connector having alignment pins 7022 that mate with alignment holes in the angled fiber array ferrule adapter 7000. The interface of the fiber array ferrule adapter 7000 may have a configuration similar to a conventional ferrule-to-ferrule connector such that the fiber array ferrule adapter 7000 may mate with a standard MPO connector or MPO-like connector.

In some embodiments, a laser source configured to generate circularly polarized light may be used in the examples shown in fig. 27 and 54-60 without the use of a quarter-wave plate or a quarter-wave polarization rotator.

In an example, a first system includes a data processor configured to process data and a photonic integrated circuit configured to convert optical signals received from one or more optical fibers into electrical signals that are transmitted to the data processor.

In an example, a second system includes a data processor configured to process data and a photonic integrated circuit configured to convert electronic signals from the data processor into optical signals that can be output to one or more optical fibers.

In an example, a third system includes a data processor configured to process data and a photonic integrated circuit configured to convert optical signals received from one or more optical fibers into electrical signals that are transmitted to the data processor. The photonic integrated circuit is further configured to convert the electronic signals from the data processor into optical signals that can be output to the one or more optical fibers.

In some embodiments, each of the first system, the second system, and the third system may include a fiber-to-PIC connector optically coupled to the one or more optical fibers and also optically coupled to a coupling element of the photonic integrated circuit. The coupling element may be, for example, a grating coupler or an edge coupler. The fiber-to-PIC connector may include one or more of the features described above, such as one or more features of fiber-to-PIC connector arrangement 500 of fig. 5, one or more features of fiber-to-PIC connector arrangement 600 of fig. 6, one or more features of fiber-to-PIC connector arrangement 700 of fig. 7, one or more features of fiber-to-PIC connector 900 of fig. 9, one or more features of fiber-to-PIC connector 1000 of fig. 10A, 10B, 12, 13, one or more features of fiber-to-PIC connector 1100 of fig. 11A, one or more features of fiber-to-PIC connector 2300 of fig. 23, one or more features of fiber-to-PIC connector 2400 of fig. 24A, 24B, one or more features of fiber-to-PIC connector 2800 of fig. 28, one or more features of fiber-to-PIC connector 3310 of fig. 33C, one or more features of fig. 34A, 34B, one or more features of fiber-to-PIC connector 3400, one or more features of fig. 35 of fig. 33C, one or more features of fiber-to-PIC connector 39 of fig. 40, and one or more features of fig. 39-to-PIC connector 3900.

Each of the first system, the second system, and the third system may include one or more features or components described in the following documents: U.S. patent application 16/822,103 filed on 18 th 3/2020, U.S. patent application 16/847,705 filed on 14 th 4/2020, U.S. patent application 16/888,890 filed on 1 th 6/2020, U.S. provisional patent application 63/080,528 filed on 18 th 9/2020, U.S. provisional patent application 63/088,914 filed on 7 th 10/2020, U.S. provisional patent application 63/116,660 filed on 20 th 11/2020, and U.S. provisional patent application 63/146,421 filed on 5/2021/2/5. The entire contents of the above application are incorporated by reference.

It will be appreciated by one of ordinary skill in the relevant art that at least some of the embodiments described herein in the context of coupling light from one or more optical fibers 202 to PIC210 may operate equally to couple light from PIC210 to one or more optical fibers 202. This reversibility of coupling direction is a general feature of at least some of the embodiments described herein, including some of the embodiments that use polarization diversity.

The example optical systems disclosed herein should be considered only as some of many possible embodiments that may be used to perform polarization demultiplexing and independent array pattern scaling, array geometry rearrangement, spot size scaling, and angle of incidence adaptation using diffraction, refraction, reflection, and polarization dependent optical elements, 3D waveguides, and 3D print optical components. Other embodiments implementing a similar set of functions may be made and used by one of ordinary skill in the relevant art in view of this disclosure and without undue experimentation.

According to an example embodiment disclosed above, for example, in the summary section and/or with reference to any one or any combination of some or all of fig. 1-8, there is provided an apparatus comprising: one or more ofOr a plurality of optical fibers (e.g., 202, fig. 5) having a plurality of cores (e.g., 302, fig. 3A-3G); a photonic integrated circuit (e.g., 210, fig. 5) comprising a plurality (e.g., 230, fig. 5) of vertical coupling elements (e.g., 231, fig. 5) disposed along a major surface of the photonic integrated circuit; and a fiber optic connector (e.g., 240/250, fig. 5) connected between the one or more optical fibers and the photonic integrated circuit to transfer light between the one or more optical fibers and the photonic integrated circuit through the major surface, the fiber optic connector comprising optics configured to transfer light between the plurality of cores and the plurality of vertical coupling elements such that: a distance between a first pair of the cores (e.g., S _min Fig. 3A-3G) are optically scaled by a first scaling factor (e.g., a); and at least one of the cores has a diameter (e.g., D _core Fig. 3A through 3G) by a second scaling factor (e.g., C) different from the first scaling factor ₁ ) Is optically scaled.

In some embodiments of the above apparatus, the optics are further configured to transmit the light such that a distance between a second pair of the cores (e.g., S _max Fig. 3A through 3G) are optically scaled by a third scaling factor (e.g., B) that is different from the second scaling factor.

In some embodiments of any of the above devices, the optical device is configured to pass the light such that the third scaling factor is different from the first scaling factor.

In some embodiments of any of the above devices, the optical device is configured to transmit the light such that the first scaling factor is substantially equal to the third scaling factor.

In some embodiments of any of the above apparatuses, the optical device comprises: one or more first lenses (e.g., 551, fig. 5) located at a first offset distance from the major surface; a plurality of second lenses (e.g., 541, fig. 5) located at a second offset distance from the major surface, the second offset distance being less than the first offset distance; and a plurality of third lenses (e.g., 542, fig. 5) located at a third offset distance from the major surface, the third offset distance being less than the second offset distance.

In some embodiments of any of the above devices, the optical device includes at least one lens (e.g., 542, fig. 5) configured to communicate light with a single one of the cores and a single one of the vertical coupling elements.

In some embodiments of any of the above devices, the optical device includes a plurality of optical waveguides (e.g., 652, fig. 6) each optically connecting a respective one of the cores and a respective one of the vertical coupling elements.

In some embodiments of any of the above devices, at least some of the optical waveguides are tapered.

In some embodiments of any of the above devices, the optics include one or more polarizing beamsplitters (e.g., 810 and 820, fig. 8A and 8B).

In some embodiments of any of the above devices, the optics include one or more polarization rotating elements (e.g., 861, 862, fig. 8B).

In some embodiments of any of the above devices, the fiber optic connector includes a first connector member (e.g., 250, fig. 5) and a second connector member (e.g., 240, fig. 5) that are disconnectably connected to each other.

In some embodiments of any of the above devices, the optics are configured to generate a light spot (e.g., 560, fig. 5) at the mating surface between the first connector component and the second connector component, the light spot having a size at least twice the corresponding diameter of the fiber core.

In some embodiments of any of the above devices, the optical device is configured to transfer light between a first number of the cores and a second number of the vertical coupling elements, the second number being greater than the first number.

In some embodiments of any of the above, the one or more optical fibers comprise a multicore fiber.

In some embodiments of any of the above devices, each of the vertical coupling elements is selected from the group of elements consisting of: single polarization vertical grating coupler, turning mirror, polarization diversity vertical grating coupler, vertical cavity surface emitting laser, surface normal line modulator and photodiode.

According to another example embodiment disclosed above, for example, in the summary section and/or with reference to any one or any combination of some or all of fig. 1-8, there is provided a fiber optic connector comprising: a first connector member (e.g., 240, fig. 5) connectable at a first side (e.g., 555, fig. 5) of the first connector member to one or more optical fibers (e.g., 202, fig. 5) having a plurality of cores (e.g., 302, fig. 3A-3G), the first connector member having a second side (e.g., 556, fig. 5) opposite the first side; a second connector component (e.g., 250, fig. 5) connectable to the second side of the first connector component at one side of the second connector component (e.g., 545, fig. 5) and further connectable to a photonic integrated circuit (e.g., 210, fig. 2) at an opposite side of the second connector component (e.g., 546, fig. 5); and optics configured to transfer light between the first side of the first connector component and the opposite side of the second connector component such that: a distance between a first pair of the cores (e.g., S _min Fig. 3A-3G) are optically scaled by a first scaling factor (e.g., a); and at least one of the cores has a diameter (e.g., D _core Fig. 3A through 3G) by a second scaling factor (e.g., C) different from the first scaling factor ₁ ) Quilt is covered withOptically scaled.

As used herein, the term "relative" refers to the relative orientation and/or positioning of two corresponding sides or edges of a component, and should be construed to encompass any of the relative orientations/positions, wherein: (i) The two sides are substantially parallel to each other (e.g., within 15 degrees) but at different ends of the component; (ii) The two sides are not parallel to each other, i.e. may be oriented at a relative angle in the range between 15 degrees and 165 degrees; (iii) the two sides are substantially perpendicular to each other; (iv) At least one of the two sides is not strictly planar and has some features deviating from planar geometry; (v) points where the two sides do not contact each other; and (vi) the two sides have a common edge or contact area, for example, at the corners of the component. The sides 545, 546, 555, and 556 shown in fig. 5 should be considered as providing non-limiting illustrative examples of such sides.

In some embodiments of the above fiber optic connectors, the optics are further configured to transmit light such that a distance between a second pair of the cores (e.g., S _max Fig. 3A through 3G) are optically scaled by a third scaling factor (e.g., B) that is different from the second scaling factor.

In some embodiments of any of the above optical fiber connectors, the optical device is configured to transmit the light such that the third scaling factor is different from the first scaling factor.

In some embodiments of any of the above optical fiber connectors, the optical device is configured to transmit the light such that the first scaling factor is substantially equal to the third scaling factor.

In some embodiments of any of the above optical fiber connectors, the optical device comprises:

one or more first lenses (e.g., 551, fig. 5) located at a first offset distance from the opposite side of the second connector component; a plurality of second lenses (e.g., 541, fig. 5) located at a second offset distance from the opposite side of the second connector member, the second offset distance being less than the first offset distance; and a plurality of third lenses (e.g., 542, fig. 5) located at a third offset distance from the opposite side of the second connector member, the third offset distance being less than the second offset distance, the first, second, and third distances being measured with the first and second connector members connected to one another.

In some embodiments of any of the fiber optic connectors described above, the optical device includes at least one lens (e.g., 542, fig. 5) configured to communicate light with a single one of the cores and a single one of the vertical coupling elements of the photonic integrated circuit.

In some embodiments of any of the fiber optic connectors described above, the optical device includes a plurality of optical waveguides (e.g., 652, fig. 6) each disposed to optically connect a respective one of the cores with a respective one of the vertical coupling elements of the photonic integrated circuit.

In some embodiments of any of the above optical fiber connectors, at least some of the optical waveguides are tapered.

In some embodiments of any of the fiber optic connectors described above, the optical device includes one or more polarizing beam splitters (e.g., 810 and 820, fig. 8A and 8B).

In some embodiments of any of the fiber optic connectors described above, the optical device includes one or more polarization rotating elements (e.g., 861, 862, fig. 8B).

In some embodiments, the ferrule frame may have two or more components, wherein the relative position and orientation between the different components may be precisely adjusted. For example, a first component may be coupled to an optical subassembly (e.g., 6508 of fig. 68 or 7110 of fig. 71A), a second component may be removably attached to a fiber optic connector (e.g., 6546), and the relative positions of the first and second components may be precisely adjusted, for example, using screws. For example, the ferrule frame may have a precision multi-axis positioner for accurately positioning the fiber optic connector.

While this disclosure includes references to illustrative embodiments, this description is not intended to be construed in a limiting sense. Various modifications of the described embodiments, as well as other embodiments within the scope of the disclosure, which are apparent to persons skilled in the art to which the disclosure pertains are deemed to lie within the principle and scope of the disclosure, e.g., as expressed in the following claims.

Unless expressly stated otherwise, each numerical value and range should be construed as being approximate, as if the numerical value or range were preceded by the word "about" or "approximately.

It will be further understood that various changes in the details, materials, and arrangements of parts which have been described and illustrated for the purpose of explaining the nature of the present disclosure may be made by those skilled in the art without departing from the scope of the present disclosure, for example, as expressed in the following claims.

The use of reference numerals and/or indicia in the claims is intended to identify one or more possible embodiments of the claimed subject matter in order to facilitate interpretation of the claims. Such use should not be construed as necessarily limiting the scope of those claims to the embodiments shown in the corresponding figures.

Although elements in the following method claims (if any) are recited in a particular order with corresponding labeling, those elements are not necessarily intended to be limited to practice in the particular order, unless the claim recitations otherwise imply a particular order for implementing some or all of those elements.

Reference herein to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the disclosure. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments necessarily mutually exclusive of other embodiments. The above applies to the term "embodiment".

Unless otherwise indicated herein, the use of ordinal adjectives "first," "second," "third," etc., to refer to one object from a plurality of like objects, merely indicate that different examples of like objects are being referred to, and are not intended to imply that like objects so referred to must be in a corresponding sequence or order, either temporally, spatially, in ranking, or in any other manner.

Also for purposes of this specification, the terms "couple," "coupled," "connected," or "connected" refer to any manner known in the art or later developed in which energy is allowed to pass between two or more elements, and the insertion of one or more additional elements is contemplated, although not required. In contrast, the terms "directly coupled," "directly connected," and the like, imply that no such additional elements are present.

The description and drawings merely illustrate the principles of the disclosure. It will thus be appreciated that those skilled in the art will be able to devise various arrangements that, although not explicitly described or shown herein, embody the principles of the disclosure and are included within its spirit and scope. Furthermore, all examples cited herein are in principle explicitly intended for pedagogical purposes only to aid the reader in understanding the principles of the disclosure and the concepts contributed by the inventor to furthering the art, and are to be construed as being without limitation to such specifically recited examples and conditions. Moreover, all statements herein reciting principles, aspects, and embodiments of the disclosure, as well as specific examples thereof, are intended to encompass equivalents thereof.

As used in this application, the term "circuitry" may refer to one or more or all of the following: (a) Circuit implementations that are hardware-only (e.g., implementations in analog and/or digital circuitry only); (b) A combination of hardware circuitry and software, such as (if applicable): (i) A combination of analog and/or digital hardware circuitry and software/firmware, and (ii) any portion of a hardware processor (including digital signal processors), software, and memory with software that work together to cause a device, such as a cell phone or server, to perform various functions; and (c) hardware circuitry and/or a processor, such as a microprocessor or a portion of a microprocessor, that requires software (e.g., firmware) to operate, but may not be present when software is not required to operate. This definition of circuitry applies to all uses of this term in this application, including in any claims. As a further example, as used in this application, the term circuitry also encompasses hardware-only circuitry or processor (or multiple processors) or an implementation of a hardware circuit or portion of a processor and its (or their) accompanying software and/or firmware. For example and if applicable to particular claim elements, the term circuitry also encompasses baseband integrated circuits or processor integrated circuits for a mobile device or similar integrated circuits in a server, a cellular network device, or other computing or network device.

The terms "upper", "lower", "top" and "bottom" refer to the relative positions shown in the figures. It should be understood that the systems, devices, and components described in this document may be used in a variety of orientations. Thus, a system, device, or component having a surface described in this document as an "upper surface" or "top surface" may be arbitrarily oriented during operation of the system, device, or component such that the surface faces in any direction, e.g., downward or sideways.

It will be appreciated by those of ordinary skill in the art that any block diagrams herein represent conceptual views of illustrative circuitry embodying the principles of the disclosure.

Although the invention is defined in the appended claims, it should be understood that the invention may also be defined in accordance with the following examples:

example 1: a method, the method comprising:

providing a photonic integrated circuit comprising a plurality of vertical coupling elements disposed along a major surface of the photonic integrated circuit;

attaching an optical subassembly to the photonic integrated circuit;

removably connecting a fiber optic connector to the ferrule frame, wherein the fiber optic connector is attached to the fiber array;

Aligning the ferrule frame with the optical subassembly using an alignment process; and

after the active alignment process, the ferrule frame is securely attached to the optical subassembly.

Example 2: the method of embodiment 1, the alignment process comprising an active alignment process comprising transmitting light between at least one optical fiber and the photonic integrated circuit.

Example 3: the method of embodiment 2, the active alignment process comprising:

Example 4: the method of any one of embodiments 1-3, comprising removing the fiber optic connector from the ferrule frame.

Example 5: the method of any one of embodiments 1-4, the ferrule frame comprising an opening to allow light from the fiber array to be transmitted to the optical subassembly.

Example 6: the method of embodiment 5 comprising passing a portion of the fiber optic connector through an opening of the ferrule frame and positioning an end of the fiber optic connector adjacent the optical subassembly.

Example 7: the method of embodiment 5 or 6, comprising passing a portion of the fiber optic connector through an opening of the ferrule frame, and positioning an end of the fiber optic connector near the optical subassembly.

Example 8: the method of any one of embodiments 1-7, removably connecting the fiber array to the ferrule frame comprises at least one of: (i) Aligning the fiber array with the ferrule frame using one or more alignment pins; (ii) Using one or more clamps to secure the fiber array to the ferrule frame; (iii) Using one or more magnets to connect the fiber array to the ferrule frame; or (iv) using a removable adhesive to connect the fiber array to the ferrule frame.

Example 9: the method of any one of embodiments 1-8, the array of optical fibers comprising a two-dimensional array of optical fibers.

Example 10: the method of embodiment 9, the two-dimensional array of optical fibers comprising at least two rows of optical fibers.

Example 11: the method of any one of embodiments 1-10, the fiber array comprising at least 10 cores.

Example 12: the method of embodiment 11, the fiber array comprising at least 50 cores.

Example 13: the method of embodiment 12, the fiber array comprising at least 100 cores.

Example 14: the method of any of embodiments 1-13, the optical subassembly comprising a first lens array, and the active alignment process comprising projecting light from the array of optical fibers through the first lens array to a corresponding vertical coupling element, comprising passing light from at least one of the optical fibers through a corresponding lens to the corresponding vertical coupling element.

Example 15: the method of embodiment 14, the optical subassembly comprising a second lens array, and the active alignment process comprising projecting light from the fiber array through the first lens array and the second lens array to the at least one vertical coupling element.

Example 16: the method of embodiment 15, the optical subassembly comprising a beam shifter, and the active alignment process comprising projecting light from the fiber array through the first lens array, the beam shifter, and the second lens array to the at least one vertical coupling element.

Example 17: the method of embodiment 16, the optical subassembly comprising a half-wave plate, and the active alignment process comprising projecting light from the fiber array through the first lens array, the beam shifter, the half-wave plate, and the second lens array to the at least one vertical coupling element.

Example 18: the method of embodiment 15, the optical subassembly comprising a spacer block disposed between the first lens array and the second lens array along an optical path, and the active alignment process comprising projecting light from the fiber array through the first lens array, the spacer block, and the second lens array to the at least one vertical coupling element.

Example 19: the method of embodiment 18, the optical subassembly comprising a half-wave plate, and the active alignment process comprising projecting light from the fiber array through the first lens array, the spacer block, the half-wave plate, and the second lens array to the at least one vertical coupling element.

Example 20: the method of any one of embodiments 1-13, the fiber optic connector comprising a first lens array, and the active alignment process comprising projecting light from the fiber optic array through the first lens array to a corresponding vertical coupling element, comprising passing light from at least one of the optical fibers through a corresponding lens to the corresponding vertical coupling element.

Example 21: the method of embodiment 20, the optical subassembly comprising a second lens array, and the active alignment process comprising projecting light from the fiber array through the first lens array and the second lens array to the at least one vertical coupling element.

Example 22: the method of embodiment 21, the optical subassembly comprising a beam shifter, and the active alignment process comprising projecting light from the fiber array through the first lens array, the beam shifter, and the second lens array to the at least one vertical coupling element.

Example 23: the method of embodiment 22, the optical subassembly comprising a half-wave plate, and the active alignment process comprising projecting light from the fiber array through the first lens array, the beam shifter, the half-wave plate, and the second lens array to the at least one vertical coupling element.

Example 24: the method of embodiment 21, the optical subassembly comprising a spacer block disposed between the first lens array and the second lens array along an optical path, and the active alignment process comprising projecting light from the fiber array through the first lens array, the spacer block, and the second lens array to the at least one vertical coupling element.

Example 25: the method of embodiment 24, the optical subassembly comprising a half-wave plate, and the active alignment process comprising projecting light from the fiber array through the first lens array, the spacer block, the half-wave plate, and the second lens array to the at least one vertical coupling element.

Example 26: the method of any one of embodiments 1-25, the active alignment process comprising adjusting the position of the ferrule frame relative to the optical subassembly to maximize an overall efficiency of light transfer between the fiber array and the photonic integrated circuit.

Example 27: the method of any one of embodiments 1-26, at least one half of the ferrule frame being made of at least one of glass, metal, or plastic by weight.

Example 28: the method of any one of embodiments 1-26, the ferrule frame comprising a material that is transparent or translucent to Ultraviolet (UV) light, and fixedly connecting the ferrule frame to the optical subassembly comprises attaching the ferrule frame to the optical subassembly using a UV-curable adhesive.

Example 29: the method of any one of embodiments 1-28, adjusting the position of the ferrule frame relative to the optical subassembly comprising adjusting the position of the ferrule frame along a plane substantially parallel to a major surface of the photonic integrated circuit.

Example 30: the method of embodiment 29, adjusting the position of the ferrule frame along the plane substantially parallel to a major surface of the photonic integrated circuit comprising at least one of: (i) Adjusting the position of the ferrule frame along an x-axis relative to a major surface of the photonic integrated circuit; (ii) Adjusting the position of the ferrule frame along a y-axis relative to a major surface of the photonic integrated circuit; or (iii) rotating the ferrule frame about a z-axis relative to a major surface of the photonic integrated circuit;

Wherein the x-axis and the y-axis are substantially parallel to a major surface of the photonic integrated circuit and the z-axis is substantially perpendicular to the major surface of the photonic integrated circuit.

Example 31: the method of any one of embodiments 1-30, adjusting the position of the ferrule frame relative to the optical subassembly comprises adjusting a distance of an end of the fiber optic connector relative to the optical subassembly.

Example 32: the method of any one of embodiments 1-31, adjusting the position of the ferrule frame relative to the optical subassembly comprises adjusting an angle of inclination of an end face of the fiber optic connector relative to the optical subassembly.

Example 33: the method of any of embodiments 1-32, aligning the ferrule frame with the optical subassembly comprises aligning the ferrule frame with the optical subassembly with an accuracy of at least 10 μιη accuracy.

Example 34: the method of embodiment 33, aligning the ferrule frame with the optical subassembly comprising aligning the ferrule frame with the optical subassembly with an accuracy of at least 1 μιη accuracy.

Example 35: the method of embodiment 34, aligning the ferrule frame with the optical subassembly comprising aligning the ferrule frame with the optical subassembly with an accuracy of at least 0.1 μm accuracy.

Example 36: the method of any one of embodiments 1-35, each of the vertical coupling elements comprising at least one of: single polarization vertical grating couplers, turning mirrors, polarization diversity vertical grating couplers, vertical cavity surface emitting lasers, surface normal line modulators, or photodiodes.

Example 37: the method of any one of embodiments 16, 17, 22, 23, and 26-36, the beam shifter comprising a polarization dependent optical element.

Example 38: the method of any one of embodiments 1-37, the optical subassembly comprising a turning mirror that turns a first optical path between an optical fiber and a corresponding vertical coupling element,

wherein the first optical path includes a first optical path section between the vertical coupling element and a reflective surface of the turning mirror, and a second optical path section between the reflective surface of the turning mirror and the optical fiber, the second optical path section being at an angle θ1 with respect to the first optical path section, and θ1 is in a range of 20 ° to 160 °.

Example 39: the method of embodiment 38, θ1 is in the range of 45 ° to 110 °.

Example 40: the method of embodiment 39, θ1 is in the range of 80 ° to 100 °.

Example 41: the method of embodiment 40, after the ferrule frame is securely connected to the optical subassembly, the ferrule frame is oriented such that when the fiber optic connector is removably connected to the ferrule frame, at least some of the fibers in the array of fibers output a light beam in a direction substantially parallel to a major surface of the photonic integrated circuit, the output light beam traveling at an angle θ2 relative to the major surface of the photonic integrated circuit, and 0 ° - θ2-10 °.

Example 42: the method of any one of embodiments 38-41, the optical subassembly comprising a beam shifting element disposed between the steering mirror and the ferrule frame.

Example 43: the method of any one of embodiments 38-42, the optical subassembly comprising a lens array disposed between the turning mirror and the photonic integrated circuit.

Example 44: the method of any one of embodiments 1 to 43, comprising controlling a machine using a computer to align the ferrule frame with the optical subassembly using the active alignment process.

Example 45: an apparatus, the apparatus comprising:

a photonic integrated circuit comprising a plurality of vertical coupling elements disposed along a major surface of the photonic integrated circuit;

an optical subassembly attached to the photonic integrated circuit;

a ferrule frame configured to enable a fiber optic connector to be removably connected to the ferrule frame and aligned with the optical subassembly;

wherein the fiber optic connector is connected to an array of optical fibers and the optical subassembly is configured to transfer light between the array of optical fibers and the vertical coupling element on the photonic integrated circuit;

wherein the ferrule frame is aligned with the optical sub-assembly using an active alignment process in which light is transferred between at least one optical fiber of the array of optical fibers and the photonic integrated circuit through the optical sub-assembly and at least one vertical coupling element of the plurality of vertical coupling elements, and the position of the ferrule frame relative to the optical sub-assembly is adjusted based on at least one characteristic of the light transferred between the at least one optical fiber and the photonic integrated circuit; and is also provided with

Wherein the ferrule frame is securely attached to the optical subassembly after the active alignment process.

Example 46: the apparatus of embodiment 45, the ferrule frame enabling the fiber array to be aligned with the optical subassembly with an accuracy of at least 10 μm.

Example 47: the apparatus of embodiment 45, the ferrule frame enabling the fiber array to be aligned with the optical subassembly with an accuracy of at least 1 μm.

Example 48: the apparatus of embodiment 45, the ferrule frame enabling the fiber array to be aligned with the optical subassembly with an accuracy of at least 0.1 μm.

Example 49: the apparatus of any of embodiments 45-48, the optical subassembly comprising a first lens array, and a ferrule module configured to align the fiber array with the lens array.

Example 50: the apparatus of embodiment 49, the optical subassembly comprising a second lens array, and the second lens array positioned along a beam path between the first lens array and the vertical coupling element.

Example 51: the apparatus of embodiment 50, the optical subassembly comprising a beam shifter.

Example 52: the apparatus of embodiment 51, the optical subassembly comprising a half-wave plate positioned between the beam shifter and the second lens array.

Example 53: the apparatus of embodiment 51, the optical subassembly comprising a birefringent plate having an aperture, and the birefringent plate positioned between the beam shifter and the second lens array.

Example 54: the apparatus of embodiment 50, the optical subassembly comprising a spacer block disposed along the beam path between the first lens array and the second lens array.

Example 55: the apparatus of embodiment 54, the optical subassembly comprising a half-wave plate positioned along the beam path between the spacer block and the second lens array.

Example 56: the apparatus of any of embodiments 45-48, the fiber optic connector comprising a first lens array, and a ferrule module configured to align the first lens array with the optical subassembly.

Example 57: the apparatus of embodiment 56, the optical subassembly comprising a second lens array positioned along a beam path between the first lens array and the vertical coupling element.

Example 58: the apparatus of embodiment 57, the optical subassembly comprising a beam shifter.

Example 59: the apparatus of embodiment 58, the optical subassembly comprising a half-wave plate positioned between the beam shifter and the second lens array.

Example 60: the apparatus of embodiment 58, the optical subassembly comprising a birefringent plate having an aperture, and the birefringent plate being positioned between the beam shifter and the second lens array.

Example 61: the apparatus of embodiment 57, the optical subassembly comprising a spacer block disposed along the beam path between the first lens array and the second lens array.

Example 62: the apparatus of embodiment 61, the optical subassembly comprising a half-wave plate positioned along the beam path between the spacer block and the second lens array.

Example 63: the apparatus of any of embodiments 45-48, each optical fiber comprising one or more cores, the optical subassembly comprising at least one lens configured to communicate light with a single one of the cores and a single one of the vertical coupling elements.

Example 64: the apparatus of any of embodiments 45-48, each optical fiber comprising one or more cores, the optical subassembly comprising a plurality of optical waveguides, each optical waveguide optically connecting a respective one of the cores and a respective one of the vertical coupling elements.

Example 65: the apparatus of embodiment 64, at least some of the optical waveguides being tapered.

Example 66: the apparatus of any one of embodiments 45 to 65, the optical subassembly comprising one or more polarizing beam splitters.

Example 67: the apparatus of any one of embodiments 45 to 66, the optical subassembly comprising one or more polarization rotating elements.

Example 68: the apparatus of any of embodiments 45-67, each optical fiber comprising one or more cores, the optical subassembly configured to transfer light between a first number of the cores and a second number of the vertical coupling elements, and the second number being greater than the first number.

Example 69: the apparatus of any one of embodiments 45-68, each of the vertical coupling elements comprising at least one of: single polarization vertical grating couplers, turning mirrors, polarization diversity vertical grating couplers, vertical cavity surface emitting lasers, surface normal line modulators, or photodiodes.

Example 70: the apparatus of any one of embodiments 45-69, the ferrule frame being made of at least one of glass, metal, or plastic by weight.

Example 71: the apparatus of any one of embodiments 45-69, the ferrule frame comprising a material that is transparent or translucent to Ultraviolet (UV) light, and a UV curable adhesive for securely attaching the ferrule frame to the optical subassembly.

Example 72: the apparatus of any one of embodiments 51-53, 58-60, and 63-71, the beam shifter comprising a polarization dependent optical element.

Example 73: the apparatus of any one of embodiments 45 to 72, the optical subassembly comprising a turning mirror that turns a first optical path between an optical fiber and a corresponding vertical coupling element,

wherein the first optical path includes a first optical path section between the vertical coupling element and a reflective surface of the turning mirror, and a second optical path section between the reflective surface of the turning mirror and the optical fiber, the second optical path section being at an angle θ1 with respect to the first optical path section, and θ1 is in a range of 30 ° to 150 °.

Example 74: the apparatus of embodiment 73, θ1 being in a range of 45 ° to 110 °.

Example 75: the apparatus of embodiment 74, θ1 being in the range of 80 ° to 100 °.

Example 76: the apparatus of embodiment 75, the ferrule frame oriented such that when the fiber optic connector is removably connected to the ferrule frame and the apparatus is in operation, at least some of the optical fibers in the array of optical fibers output a light beam in a direction substantially parallel to a major surface of the photonic integrated circuit, the output light beam traveling at an angle θ2 with respect to the major surface of the photonic integrated circuit, and 0 Σ θ2 Σ 10 °.

Example 77: the apparatus of any one of embodiments 73-76, the optical subassembly comprising a beam shifting element disposed between the turning mirror and the ferrule frame.

Example 78: the apparatus of any one of embodiments 73-77, the optical subassembly comprising a lens array disposed between the turning mirror and the photonic integrated circuit.

Example 79: an apparatus, the apparatus comprising:

a plurality of photonic integrated circuits, each photonic integrated circuit comprising a plurality of coupling elements;

A plurality of optical subassemblies, each optical subassembly attached to a corresponding photonic integrated circuit of the plurality of photonic integrated circuits;

a plurality of ferrule frames, each ferrule frame configured to enable a corresponding fiber optic connector to be removably connected to the ferrule frame and aligned with a corresponding one of the optical subassemblies;

wherein each fiber optic connector is connected to an array of optical fibers and the corresponding optical subassembly is configured to pass light between the array of optical fibers and the corresponding coupling element on the corresponding photonic integrated circuit;

wherein each ferrule frame enables alignment of the fiber array with the corresponding optical subassembly with an accuracy of at least 10 μm accuracy.

Example 80: the apparatus of embodiment 79, each ferrule frame enabling the fiber array to be aligned with the corresponding optical subassembly with an accuracy of at least 1 μm accuracy.

Example 81: the apparatus of embodiment 80, each ferrule frame enabling the fiber array to be aligned with the corresponding optical subassembly with an accuracy of at least 0.1 μm accuracy.

Example 82: the apparatus of any one of embodiments 79 to 81, each optical subassembly comprising a turning mirror that turns a first optical path between an optical fiber and a corresponding coupling element,

Wherein the first optical path includes a first optical path section between the coupling element and a reflective surface of the turning mirror, and a second optical path section between the reflective surface of the turning mirror and the optical fiber, the second optical path section being at an angle θ1 with respect to the first optical path section, and θ1 is in a range of 80 ° to 100 °.

Example 83: an apparatus, the apparatus comprising:

a storage device that stores instructions; and

at least one data processor configured to execute the instructions and to perform a process comprising controlling a machine to align a ferrule frame with an optical subassembly using an active alignment process;

wherein the optical subassembly is optically coupled to a photonic integrated circuit; and is also provided with

Wherein the ferrule frame is configured to enable a fiber optic connector to be removably connected to the ferrule frame, the fiber optic connector being attached to an array of optical fibers, and the ferrule frame is configured to align the optical fibers with the optical subassembly to transmit light between the optical fibers and the photonic integrated circuit.

Claims

1. A method, the method comprising:

attaching an optical subassembly to the photonic integrated circuit;

2. The method of claim 1, wherein the alignment process comprises an active alignment process comprising transmitting light between at least one optical fiber and the photonic integrated circuit.

3. The method of claim 2, wherein the active alignment process comprises:

4. The method of claim 1, comprising removing the fiber optic connector from the ferrule frame.

5. The method of claim 1, wherein the ferrule frame includes an opening to allow light from the fiber array to be transmitted to the optical subassembly.

6. The method of claim 5, comprising passing a portion of the optical subassembly through an opening of the ferrule frame and positioning an end of the fiber optic connector adjacent the optical subassembly.

7. The method of claim 1, comprising passing a portion of the fiber optic connector through an opening of the ferrule frame and positioning an end of the fiber optic connector adjacent the optical subassembly.

8. The method of claim 1, wherein removably connecting the fiber array to the ferrule frame comprises at least one of: (i) Aligning the fiber array with the ferrule frame using one or more alignment pins; (ii) Using one or more clamps to secure the fiber array to the ferrule frame; (iii) Using one or more magnets to connect the fiber array to the ferrule frame; or (iv) using a removable adhesive to connect the fiber array to the ferrule frame.

9. The method of claim 1, wherein the array of optical fibers comprises a two-dimensional array of optical fibers.

10. The method of claim 9, wherein the two-dimensional array of optical fibers comprises at least two rows of optical fibers.

11. The method of any one of claims 1 to 10, wherein the fiber array comprises at least 10 cores.

12. The method of claim 11, wherein the fiber array comprises at least 50 cores.

13. The method of claim 12, wherein the fiber array comprises at least 100 cores.

14. The method of any one of claims 1 to 10, wherein the optical subassembly comprises a first lens array, and the active alignment process comprises projecting light from the array of optical fibers through the first lens array, including passing light from at least one of the optical fibers through a corresponding lens to a corresponding vertical coupling element.

15. The method of claim 14, wherein the optical subassembly comprises a second lens array, and the active alignment process comprises projecting light from the fiber array through the first lens array and the second lens array to the at least one vertical coupling element.

16. The method of claim 15, wherein the optical subassembly comprises a beam shifter, and the active alignment process comprises projecting light from the fiber array through the first lens array, the beam shifter, and the second lens array to the at least one vertical coupling element.

17. The method of claim 16, wherein the optical subassembly comprises a half-wave plate, and the active alignment process comprises projecting light from the fiber array through the first lens array, the beam shifter, the half-wave plate, and the second lens array to the at least one vertical coupling element.

18. The method of claim 15, wherein the optical subassembly comprises a spacer block disposed between the first lens array and the second lens array along an optical path, and the active alignment process comprises projecting light from the fiber array through the first lens array, the spacer block, and the second lens array to the at least one vertical coupling element.

19. The method of claim 18, wherein the optical subassembly comprises a half-wave plate, and the active alignment process comprises projecting light from the fiber array through the first lens array, the spacer block, the half-wave plate, and the second lens array to the at least one vertical coupling element.

20. The method of any one of claims 1 to 10, wherein the fiber optic connector comprises a first lens array, and the active alignment process comprises projecting light from the fiber optic array through the first lens array, including passing light from at least one of the optical fibers through a corresponding lens to a corresponding vertical coupling element.

21. The method of claim 20, wherein the optical subassembly comprises a second lens array, and the active alignment process comprises projecting light from the fiber array through the first lens array and the second lens array to the at least one vertical coupling element.

22. The method of claim 21, wherein the optical subassembly comprises a beam shifter, and the active alignment process comprises projecting light from the fiber array through the first lens array, the beam shifter, and the second lens array to the at least one vertical coupling element.

23. The method of claim 22, wherein the optical subassembly comprises a half-wave plate, and the active alignment process comprises projecting light from the fiber array through the first lens array, the beam shifter, the half-wave plate, and the second lens array to the at least one vertical coupling element.

24. The method of claim 21, wherein the optical subassembly comprises a spacer block disposed between the first lens array and the second lens array along an optical path, and wherein the active alignment process comprises projecting light from the fiber array through the first lens array, the spacer block, and the second lens array to the at least one vertical coupling element.

25. The method of claim 24, wherein the optical subassembly comprises a half-wave plate, and the active alignment process comprises projecting light from the fiber array through the first lens array, the spacer block, the half-wave plate, and the second lens array to the at least one vertical coupling element.

26. The method of any one of claims 1 to 10, wherein the active alignment process comprises adjusting the position of the ferrule frame relative to the optical subassembly to maximize an overall efficiency of light transfer between the fiber array and the photonic integrated circuit.

27. The method of any one of claims 1 to 10, wherein at least half of the ferrule frame is made of at least one of glass, metal or plastic by weight.

28. The method of any one of claims 1-10, wherein the ferrule frame comprises a material that is transparent or translucent to Ultraviolet (UV) light, and fixedly connecting the ferrule frame to the optical subassembly comprises attaching the ferrule frame to the optical subassembly using a UV-curable adhesive.

29. The method of any one of claims 1-10, wherein adjusting the position of the ferrule frame relative to the optical subassembly comprises adjusting the position of the ferrule frame along a plane substantially parallel to a major surface of the photonic integrated circuit.

30. The method of claim 29, wherein adjusting the position of the ferrule frame along the plane substantially parallel to a major surface of the photonic integrated circuit comprises at least one of: (i) Adjusting the position of the ferrule frame along an x-axis relative to a major surface of the photonic integrated circuit; (ii) Adjusting the position of the ferrule frame along a y-axis relative to a major surface of the photonic integrated circuit; or (iii) rotating the ferrule frame about a z-axis relative to a major surface of the photonic integrated circuit;

31. The method of any one of claims 1 to 10, wherein adjusting the position of the ferrule frame relative to the optical subassembly comprises adjusting a distance of an end of the fiber optic connector relative to the optical subassembly.

32. The method of any one of claims 1 to 10, wherein adjusting the position of the ferrule frame relative to the optical subassembly comprises adjusting an angle of inclination of an end face of the fiber optic connector relative to the optical subassembly.

33. The method of any one of claims 1-10, wherein aligning the ferrule frame with the optical subassembly comprises aligning the ferrule frame with the optical subassembly with an accuracy of at least 10 μιη accuracy.

34. The method of claim 33, wherein aligning the ferrule frame with the optical subassembly comprises aligning the ferrule frame with the optical subassembly with an accuracy of at least 1 μιη.

35. The method of claim 34, wherein aligning the ferrule frame with the optical subassembly comprises aligning the ferrule frame with the optical subassembly with an accuracy of at least 0.1 μιη accuracy.

36. The method of any one of claims 1 to 10, wherein each of the vertical coupling elements comprises at least one of: single polarization vertical grating couplers, turning mirrors, polarization diversity vertical grating couplers, vertical cavity surface emitting lasers, surface normal line modulators, or photodiodes.

37. The method of claim 16, wherein the beam shifter comprises a polarization dependent optical element.

38. The method of any one of claims 1 to 10, wherein the optical subassembly includes a turning mirror that turns a first optical path between an optical fiber and a corresponding vertical coupling element,

39. The method of claim 38, wherein θ1 is in the range of 45 ° to 110 °.

40. The method of claim 39, wherein θ1 is in the range of 80 ° to 100 °.

41. The method of claim 40, wherein after the ferrule frame is securely attached to the optical subassembly, the ferrule frame is oriented such that when the fiber optic connector is removably attached to the ferrule frame, at least some of the fibers in the array of fibers output a light beam in a direction substantially parallel to a major surface of the photonic integrated circuit, the output light beam traveling at an angle θ2 relative to the major surface of the photonic integrated circuit, and 0 ° - θ2-10 °.

42. The method of claim 38, wherein the optical subassembly comprises a beam shifting element disposed between the steering mirror and the ferrule frame.

43. The method of claim 38, wherein the optical subassembly comprises a lens array disposed between the turning mirror and the photonic integrated circuit.

44. The method of any one of claims 1 to 10, comprising controlling a machine using a computer to align the ferrule frame with the optical subassembly using the active alignment process.

45. An apparatus, the apparatus comprising:

an optical subassembly attached to the photonic integrated circuit;

46. The apparatus of claim 45, wherein the ferrule frame enables alignment of the fiber array with the optical subassembly with an accuracy of at least 10 μm.

47. The apparatus of claim 45, wherein the ferrule frame enables alignment of the fiber array with the optical subassembly with an accuracy of at least 1 μm.

48. The apparatus of claim 45, wherein the ferrule frame enables alignment of the fiber array with the optical subassembly with an accuracy of at least 0.1 μm.

49. The apparatus of any one of claims 45 to 48, wherein the optical subassembly comprises a first lens array and a ferrule module is configured to align the fiber array with the lens array.

50. The apparatus of claim 49, wherein the optical subassembly comprises a second lens array, and the second lens array is positioned along a beam path between the first lens array and the vertical coupling element.

51. The apparatus of claim 50, wherein the optical subassembly comprises a beam shifter.

52. The apparatus of claim 51, wherein the optical subassembly comprises a half-wave plate positioned along the beam path between the beam shifter and the second lens array.

53. The apparatus of claim 51, wherein the optical subassembly comprises a birefringent plate having an aperture, and wherein the birefringent plate is positioned along the beam path between the beam shifter and the second lens array.

54. The apparatus of claim 50, wherein the optical subassembly comprises a spacer block disposed along the beam path between the first lens array and the second lens array.

55. The apparatus of claim 54, wherein the optical subassembly comprises a half-wave plate positioned along the beam path between the spacer block and the second lens array.

56. The apparatus of any one of claims 45 to 48, wherein the fiber optic connector comprises a first lens array, and a ferrule module is configured to align the first lens array with the optical subassembly.

57. The apparatus of claim 56, wherein the optical subassembly comprises a second lens array positioned along a beam path between the first lens array and the vertical coupling element.

58. The apparatus of claim 57, wherein the optical subassembly comprises a beam shifter.

59. The apparatus of claim 58, wherein the optical subassembly comprises a half-wave plate positioned along the beam path between the beam shifter and the second lens array.

60. The apparatus of claim 58, wherein the optical subassembly comprises a birefringent plate having an aperture, and the birefringent plate is positioned along the beam path between the beam shifter and the second lens array.

61. The apparatus of claim 57, wherein the optical subassembly comprises a spacer block disposed along the beam path between the first lens array and the second lens array.

62. The apparatus of claim 61, wherein the optical subassembly comprises a half-wave plate positioned along the beam path between the spacer block and the second lens array.

63. The apparatus of any one of claims 45 to 48, wherein each optical fiber comprises one or more cores, the optical subassembly comprising at least one lens configured to communicate light with a single one of the cores and a single one of the vertical coupling elements.

64. The apparatus of any one of claims 45 to 48, wherein each optical fiber comprises one or more cores and the optical subassembly comprises a plurality of optical waveguides, each optical waveguide optically connecting a respective one of the cores and a respective one of the vertical coupling elements.

65. The apparatus of claim 64, wherein at least some of the optical waveguides are tapered.

66. The apparatus of any one of claims 45 to 48, wherein the optical subassembly comprises one or more polarizing beam splitters.

67. The apparatus of any one of claims 45 to 48, wherein the optical subassembly comprises one or more polarization rotating elements.

68. The apparatus of any one of claims 45 to 48, wherein each optical fiber comprises one or more cores, the optical subassembly is configured to transfer light between a first number of the cores and a second number of the vertical coupling elements, and the second number is greater than the first number.

69. The apparatus of any one of claims 45 to 48, wherein each of the vertical coupling elements comprises at least one of: single polarization vertical grating couplers, turning mirrors, polarization diversity vertical grating couplers, vertical cavity surface emitting lasers, surface normal line modulators, or photodiodes.

70. The apparatus of any one of claims 45 to 48, wherein at least half of the ferrule frame is made of at least one of glass, metal, or plastic by weight.

71. The apparatus of any one of claims 45 to 48, wherein the ferrule frame comprises a material that is transparent or translucent to Ultraviolet (UV) light, and a UV curable adhesive is used to securely attach the ferrule frame to the optical subassembly.

72. The apparatus of claim 51, wherein the beam shifter comprises a polarization dependent optical element.

73. The apparatus of any one of claims 45 to 48, wherein the optical subassembly includes a turning mirror that turns a first optical path between an optical fiber and a corresponding vertical coupling element,

74. The apparatus of claim 73, wherein θ1 is in the range of 45 ° to 110 °.

75. The apparatus of claim 74, wherein θ1 is in the range of 80 ° to 100 °.

76. The apparatus of claim 75 wherein the ferrule frame is oriented such that when the fiber optic connector is removably connected to the ferrule frame and the apparatus is in operation, at least some of the fibers in the array of fibers output a light beam in a direction substantially parallel to a major surface of the photonic integrated circuit, the output light beam traveling at an angle θ2 relative to the major surface of the photonic integrated circuit, and 0 ° - θ2-10 °.

77. The apparatus of claim 73, wherein the optical subassembly comprises a beam shifting element disposed between the turning mirror and the ferrule frame.

78. The apparatus of claim 73, wherein the optical subassembly comprises a lens array disposed between the turning mirror and the photonic integrated circuit.

79. An apparatus, the apparatus comprising:

a plurality of ferrule frames, each ferrule frame configured to enable a corresponding fiber optic connector to be removably connected to the ferrule frame and aligned with a corresponding optical subassembly of the plurality of optical subassemblies;

80. The apparatus of claim 79, wherein each ferrule frame enables the fiber array to be aligned with the corresponding optical subassembly with an accuracy of at least 1 μιη accuracy.

81. The apparatus of claim 80, wherein each ferrule frame enables the fiber array to be aligned with the corresponding optical subassembly with an accuracy of at least 0.1 μιη accuracy.

82. The apparatus of any one of claims 79 to 81, wherein each optical subassembly comprises a turning mirror that turns a first optical path between an optical fiber and a corresponding coupling element,

83. An apparatus, the apparatus comprising:

a storage device that stores instructions; and