OPTIMIZED 2X2 3DB MULTI-MODE INTERFERENCE COUPLER

Abstract

An optimized SOI 2×2 multimode interference (MMI) coupler is designed by use of the particle swarm optimization (PSO) algorithm. Finite Difference Time Domain (FDTD) simulation shows that, within a footprint of 9.4×1.6 μm.sup.2, <0.1 dB power unbalance and <1 degree phase error are achieved across the entire C-band. The excess loss of the device is <0.2 dB.

Claims

1-11. (canceled)

12. An optical coupler, comprising: a multi-mode region including: a length L between a first end and a second end; and a plurality of segments having widths, at least five of said segment widths from the first end to the second end being different one from the other; a plurality of first ports at the first end of the multi-mode region; and a plurality of second ports at the second end of the multi-mode region.

13. The coupler according to claim 12, wherein the multi-mode region comprises a few-mode region.

14. The coupler according to claim 12, wherein said segment widths vary in a symmetric pattern relative to a central distance L/2 along said length defining a bidirectional coupler.

15. The coupler according to claim 12, wherein said segment widths are at equally spaced locations along the length.

16. The coupler according to claim 12, wherein the second and fourth segment widths from the first end are greater than the third and fifth widths from the first end.

17. The coupler according to claim 13, wherein the plurality of first ports comprises two ports; and wherein the plurality of second ports comprises two ports.

18. The coupler according to claim 17, wherein said widths range from 1.439 μm to 1.6 μm.

19. The coupler according to claim 17, wherein the few-mode region and the plurality of first and second ports are within a footprint of 9.4×1.6 μm.sup.2

20. The coupler according to claim 17, wherein each of said ports is connected to said few-mode region by a taper connector.

21. The coupler according to claim 12, wherein at least one taper connector is connected to an edge of the few-mode region to smoothly transform input/output mode profiles.

22. A method of manufacturing an optical coupler, comprising: a multi-mode region including: a length L between a first end and a second end; and a plurality of segments having widths, at least five of said segment widths from the first end to the second end being different one from the other; a plurality of first ports at the first end of the multi-mode region; and a plurality of second ports at the second end of the multi-mode region; said method comprising: determining each width for the plurality of segments for a predefined set of design parameters, using a computerized optimization algorithm; and fabricating the optical coupler with the widths.

23. The method according to claim 22, wherein the multi-mode region comprises a few-mode region.

24. The method according to claim 22, wherein the computerized optimization algorithm comprises one of a particle swarm optimization algorithm.

25. The method according to claim 22, wherein said segment widths vary in a symmetric pattern relative to a central distance L/2 along said length defining a bidirectional coupler.

26. The method according to claim 22, wherein said segment widths are at equally spaced locations along the length.

27. The method according to claim 22, wherein the second and fourth segment widths from the first end are greater than the third and fifth widths from the first end.

28. The method according to claim 23, wherein the plurality of first ports comprises two ports; and wherein the plurality of second ports comprises two ports.

29. The coupler according to claim 28, wherein said widths range from 1.439 μm to 1.6 μm.

30. The method according to claim 28, wherein the few-mode region and the plurality of first and second ports are within a footprint of 9.4×1.6 μm.sup.2

31. The method according to claim 28, wherein each of said ports is connected to said few-mode region by a taper connector to smoothly transform input/output mode profiles.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The objects and features of the invention can be better understood with reference to the drawings described below, and the claims. The drawings are not necessarily to scale, emphasis instead generally being placed upon illustrating the principles of the invention. In the drawings, like numerals are used to indicate like parts throughout the various views.

[0017] FIG. 1 is a schematic diagram of a prior art MMI coupler.

[0018] FIG. 2 is a schematic of one embodiment of an improved MMI coupler constructed and operated according to principles of the invention.

[0019] FIG. 3 is a graph of the finite difference time domain simulated electrical field propagation in a device of FIG. 2.

[0020] FIG. 4 is a graph showing an insertion loss simulation for a device of FIG. 2.

[0021] FIG. 5 is a graph showing a phase error simulation for a device of FIG. 2.

[0022] FIG. 6 is a graph showing the experimental data for the loss vs. the number of cascaded devices, and a curve fit to the data.

[0023] FIG. 7 is a schematic diagram of a test structure comprising a Mach Zehnder interferometer used as a phase tuner to drive a MMI coupler.

[0024] FIG. 8 is a graph of the optical power in each of a top arm and a bottom arm as a function of phase tuning power.

[0025] FIG. 9 is a schematic diagram in which a first 2×2 multi-mode interference coupler is cascaded with a second 2×2 multi-mode interference coupler 200.

DETAILED DESCRIPTION

Acronyms

[0026] A list of acronyms and their usual meanings in the present document (unless otherwise explicitly stated to denote a different thing) are presented below.

[0027] AMR Adabatic Micro-Ring

[0028] APD Avalanche Photodetector

[0029] ARM Anti-Reflection Microstructure

[0030] ASE Amplified Spontaneous Emission

[0031] BER Bit Error Rate

[0032] BOX Buried Oxide

[0033] CMOS Complementary Metal-Oxide-Semiconductor

[0034] CMP Chemical-Mechanical Planarization

[0035] DBR Distributed Bragg Reflector

[0036] DC (optics) Directional Coupler

[0037] DC (electronics) Direct Current

[0038] DCA Digital Communication Analyzer

[0039] DRC Design Rule Checking

[0040] DSP Digital Signal Processor

[0041] DUT Device Under Test

[0042] ECL External Cavity Laser

[0043] E/O Electro-optical

[0044] FDTD Finite Difference Time Domain

[0045] FFE Feed-Forward Equalization

[0046] FOM Figure of Merit

[0047] FSR Free Spectral Range

[0048] FWHM Full Width at Half Maximum

[0049] GaAs Gallium Arsenide

[0050] InP Indium Phosphide

[0051] LiNO.sub.3 Lithium Niobate

[0052] LIV Light intensity (L)-Current (I)-Voltage (V)

[0053] MFD Mode Field Diameter

[0054] MMI Multi Mode Interference

[0055] MPW Multi Project Wafer

[0056] NRZ Non-Return to Zero

[0057] OOK On-Off Keying

[0058] PIC Photonic Integrated Circuits

[0059] PRBS Pseudo Random Bit Sequence

[0060] PDFA Praseodymium-Doped-Fiber-Amplifier

[0061] PSO Particle Swarm Optimization

[0062] Q Quality factor

[00001]$Q=2.Math.\pi \times \frac{\mathrm{Energy}.Math..Math.\mathrm{Stored}}{\mathrm{Energy}.Math..Math.\mathrm{dissipated}.Math..Math.\mathrm{per}.Math..Math.\mathrm{cycle}}=2.Math.\pi .Math..Math.f_r\times \frac{\mathrm{Energy}.Math..Math.\mathrm{Stored}}{\mathrm{Power}.Math..Math.\mathrm{Loss}}.$

[0063] QD Quantum Dot

[0064] RSOA Reflective Semiconductor Optical Amplifier

[0065] SOI Silicon on Insulator

[0066] SEM Scanning Electron Microscope

[0067] SMSR Single-Mode Suppression Ratio

[0068] TEC Thermal Electric Cooler

[0069] WDM Wavelength Division Multiplexing

[0070] An optimized Silicon-On-Insulator 2×2 MMI (multimode interference) coupler useful in manipulating optical signals is designed by particle swarm optimization (PSO) algorithm. FDTD simulation shows that, within a footprint of 9.4×1.6 μm.sup.2, <0.1 dB power unbalance and <1 degree phase error are achieved across the entire C-band. The excess loss of the device is <0.2 dB.

[0071] FIG. 1 is a schematic diagram of a prior art MMI coupler. The operating principle is defined by self-imaging theory. The prior art MMI coupler has four ports, P1 102, P2 104, P3 106 and P4 108. A waveguide mode optical signal is launched at one of these four ports (P1 in the case of FIG. 1), propagates in a rectangular piece of multimode region, and then two imaging mode optical signals with 90-degree phase difference emerge at the output ports (P2 and P4 in FIG. 1). The vertical port locations are fixed at ±¼ W.sub.MMI as required by self-imaging theory. In the conventional prior art practice, one tunes the width (W.sub.MMI 110) and length (L.sub.MMI 112) when designing a 3 dB MMI coupler.

[0072] As can be seen, there is not much design freedom for a typical rectangular shaped 2×2 MMI. A self-imaging point can be readily found by tuning W.sub.MMI and L.sub.MMI. However, to couple light out of the multimode region to a waveguide introduces excess loss. In addition, the geometry and symmetry property of the MMI will be altered during fabrication as a result of variations in processing such as may be caused by variations from run to run or even from wafer to wafer in lithography, etching, wafer thickness variation, and the like, affecting the power balance and the phase error. Simply changing the dimensions W.sub.MMI and L.sub.MMI of the device does not produce useful results.

[0073] We describe an MMI that is designed using an optimization algorithm. Some features of this device are now enumerated. The geometry of the multimode region is no longer a rectangle but is optimized by application of the particle swarm optimization (PSO) algorithm. Short tapers are introduced between the multimode region and input/output waveguides to better guiding the optical mode. A few-mode region is chosen to enhance the optical field coupling and shrink device footprint, which is different from the typical prior art MMI coupler in which the multimode region supports a large number of optical modes.

[0074] FIG. 2 is a schematic of one embodiment of an improved MMI coupler constructed and operated according to principles of the invention.

[0075] In FIG. 2, the width of the MMI coupler is digitized into several segments (8 segments in the embodiment illustrated in FIG. 2), identified by a width parameter Wi, where i is a positive integer. In FIG. 2 the widths are given as {W1, W2, W3, W4, W5, W4, W3, W2, W1}. In the embodiment illustrated in FIG. 2, the widths are taken at equally spaced locations along the length L.sub.MMI. In other embodiments, the widths can be determined at locations that are not equally spaced along the length L.sub.MMI. By defining the width parameter group, the geometric symmetry of the MMI coupler is maintained. Because of the geometric symmetry, the MMI coupler will work the same way in either direction. The input/output guiding tapers connect to the very edge of the multimode region to smoothly transform the input/output mode profiles. A gap 210 having a dimension Dgap is predefined between the top and the bottom waveguides, so that they are spaced apart by that distance. During optimization, L.sub.MMI is fixed. After the optimized MMI geometry has been obtained, one can tune or modify the length L.sub.taper to further reduce optical loss.

[0076] A design figure of merit (FOM) was set to be the total output power minus the unbalance of (or absolute difference between) the power of two output branches, as given by the following equation, in which the power is measured at ports p3 and p4, as shown in FIG. 2:

FOM=Power(p3)+Power(p4)−abs(Power(p3)−Power(p4)).

[0077] The SOI thickness during simulation is set at 220 nm. By choosing the parameters Dgap=0.2 μm, W_taper_wide=0.7 μm, W_taper_narrow=0.5 μm and L.sub.MMI=8 μm, PSO converges with a FOM=0.985. The width parameters for the embodiment shown in FIG. 2 are presented in Table 1.

TABLE-US-00001 TABLE 1 W.sub.1 W.sub.2 W.sub.3 W.sub.4 W.sub.5 Width (μm) 1.6 1.587 1.45 1.5 1.439 Distance from 0 1 2 3 4 edge of MMI (μm)

[0078] FIG. 3 is a graph of the finite difference time domain simulated electrical field propagation in a device of FIG. 2. As is clearly seen, the amplitude of the E-field is evenly distributed at the right side with minimal scattering loss.

[0079] FIG. 4 is a graph showing an insertion loss simulation for a device of FIG. 2. The detailed wavelength dependent performance of each output branch is shown in FIG. 4. Curve 410 is the curve for signal input at port p2 and signal output at port p3, while curve 420 is that for signal input at port p2 and signal output at port p4. Similar behavior would be expected for signal input at port p1 in place of port p2. Overall, the average excess loss in either branch is about 0.07 dB with a worst case of 0.13 dB and best case of 0.04 dB. These two branches are very well balanced, with <0.1 dB difference. The device also provides ultra broadband performance, with <0.1 dB variation across the C-band.

[0080] FIG. 5 is a graph showing a phase error simulation for a device of FIG. 2. As shown in FIG. 5, the phase difference is almost perfectly matched to 90-degree, within an error of 0.6 degree across C-band. The 2×2 MMI described is expected to provide high performance in loss and power balance, with a phase error that is very small.

Experimental Results

[0081] Insertion loss can be measured by cascading the devices. By cascading the devices with different numbers, one can accurately extract the insertion loss of the device. One application of such cascaded structures is to provide a test structure in the spare space of a large system to enable device characterizations in wafer scale fabrication.

[0082] FIG. 6 is a graph showing the experimental data for the loss in delivered power vs. the number of cascaded devices, and a curve fit to the data. The measured loss per device is about 0.11 dB at a wavelength around 1550 nm.

[0083] FIG. 7 is a schematic diagram of a test structure comprising a Mach Zehnder interferometer 710 used as a phase tuner to drive a MMI coupler 720. Imbalance and phase error can be measured by the MZI structure shown in FIG. 7. The imbalance can be measured by the extinction ratio of the MZI spectrum.

[0084] FIG. 8 is a graph of the optical power in each of a top arm and a bottom arm as a function of phase tuning power. As shown in FIG. 8, the extinction ratio is about 45 dB for both output arms, indicating imbalance of about 0.1 dB. By comparing the phases of bottom arm and top arm, the phase error is measured to be within 1 degree.

[0085] It is believed that apparatus constructed using principles of the invention and methods that operate according to principles of the invention can be used in the wavelength ranges (O band, E band, S band, C band, L band, and U band) described in Table II.

TABLE-US-00002 TABLE II Band Description Wavelength Range O band original 1260 to 1360 nm E band extended 1360 to 1460 nm S band short wavelengths 1460 to 1530 nm C band conventional (“erbium window”) 1530 to 1565 nm L band long wavelengths 1565 to 1625 nm U band ultralong wavelengths 1625 to 1675 nm

[0086] FIG. 9 is a schematic diagram in which a first 2×2 multi-mode interference coupler 200 (2×2 MMI A) is cascaded with a second 2×2 multi-mode interference coupler 200 (2×2 MMI A). As shown in FIG. 9 output port P4A of 2×2 MMI A is in optical communication with input port P1B of 2×2 MMI B by way of optical carrier 900, which in various embodiments can be an optical waveguide, or the two 2×2 multi-mode interference couplers can be close enough that one output is directly in optical communication with an input of a subsequent coupler. Note that port P4A can also be directly connected to port P2B (meanwhile P3A is connected to P1B) to form cascaded MMI structure. As illustrated in FIG. 9, any convenient number of couplers can be cascaded, if A=1, B=2, and N is a positive integer equal to or greater than 3.

[0087] It is believed that other coupler, such as to other M×M or N×M MMI coupler designs such as 3×3 MMI, 4×4 MMI, 2×4 MMI, and so forth can also be designed and constructed by direct extension of the methods to design and to fabricate the 2×2 MMI embodiment that has been described herein.

Design and Fabrication

[0088] Methods of designing and fabricating devices having elements similar to those described herein are described in one or more of U.S. Pat. Nos. 7,200,308, 7,339,724, 7,424,192, 7,480,434, 7,643,714, 7,760,970, 7,894,696, 8,031,985, 8,067,724, 8,098,965, 8,203,115, 8,237,102, 8,258,476, 8,270,778, 8,280,211, 8,311,374, 8,340,486, 8,380,016, 8,390,922, 8,798,406, and 8,818,141, each of which documents is hereby incorporated by reference herein in its entirety.

Definitions

[0089] As used herein, the term “optical communication channel” is intended to denote a single optical channel, such as light that can carry information using a specific carrier wavelength in a wavelength division multiplexed (WDM) system.

[0090] As used herein, the term “optical carrier” is intended to denote a medium or a structure through which any number of optical signals including WDM signals can propagate, which by way of example can include gases such as air, a void such as a vacuum or extraterrestrial space, and structures such as optical fibers and optical waveguides.

Theoretical Discussion

[0091] Although the theoretical description given herein is thought to be correct, the operation of the devices described and claimed herein does not depend upon the accuracy or validity of the theoretical description. That is, later theoretical developments that may explain the observed results on a basis different from the theory presented herein will not detract from the inventions described herein.

[0092] Any patent, patent application, patent application publication, journal article, book, published paper, or other publicly available material identified in the specification is hereby incorporated by reference herein in its entirety. Any material, or portion thereof, that is said to be incorporated by reference herein, but which conflicts with existing definitions, statements, or other disclosure material explicitly set forth herein is only incorporated to the extent that no conflict arises between that incorporated material and the present disclosure material. In the event of a conflict, the conflict is to be resolved in favor of the present disclosure as the preferred disclosure.

[0093] While the present invention has been particularly shown and described with reference to the preferred mode as illustrated in the drawing, it will be understood by one skilled in the art that various changes in detail may be affected therein without departing from the spirit and scope of the invention as defined by the claims.