Patent Application
20250237828
This disclosure relates to configuring index of refraction in a photonic integrated circuit.
Complementary metal-oxide-semiconductor (CMOS) processes and other fabrication techniques can be used to fabricate electronic integrated circuits (EICs) that operate using electrical signals (e.g., voltage signals and/or current signals). Similar fabrication techniques can be used to fabricate photonic integrated circuits (PICs) in a silicon photonic platform or in other integrated photonic platforms. A silicon on insulator platform is an example of a silicon photonic platform that can be used to make opto-electrical active devices, optical passive devices, and optical waveguides in a silicon layer. In a silicon on insulator platform, the optical signals can be transmitted by optical waveguides and can be confined within the silicon layer, for example, because there is an underlying buried oxide (BOX) layer made up of thermal silicon dioxide (i.e., silicon oxidized using a thermal process) and an overlying silicon dioxide cladding surrounding the silicon layers. In such examples, the index contrast between the high index of refraction of silicon and the low index of refraction of silicon dioxide can be responsible for the confinement. Some advantages of silicon photonic platforms are the ability to make both active and passive devices, and the ability to make compact PICs due to the high index contrast between silicon and silicon dioxide.
In one aspect, in general, a method for fabricating a photonic integrated circuit comprises: forming a first waveguide portion; forming a first set of index-modifying elements located in proximity to the first waveguide portion, each configured to modify an index of refraction of the first waveguide portion based on a respective electrical signal; forming an electrical power source configured to supply one or more electrical signals; forming a first set of modifiable electrical elements, each configured to be modifiable at least once between a conductive configuration that conducts electricity and a nonconductive configuration that does not conduct electricity, and each electrically connected to the electrical power source and to one or more respective index-modifying elements in the first set of index-modifying elements; transmitting a first optical wave portion through the first waveguide portion; determining first optical wave information associated with the first optical wave portion after it traverses the first waveguide portion; and modifying one or more modifiable electrical elements in the first set of modifiable electrical elements based at least in part on the first optical wave information.
Aspects can include one or more of the following features.
Where the first optical wave information is associated with at least one of a first phase or a first optical power of the first optical wave portion.
Where modifying one or more modifiable electrical elements in the first set of modifiable electrical elements modifies at least one of a real component or an imaginary component of the index of refraction of the first waveguide portion.
Where modifying one or more modifiable electrical elements in the first set of modifiable electrical elements comprises applying a laser pulse or applying electrical power above a threshold value.
Where modifying one or more modifiable electrical elements in the first set of modifiable electrical elements comprises adding or removing one or more (1) wire bonds, (2) flip-chip bumps, or (3) redistribution layers.
The method further comprises forming a second waveguide portion non-overlapping with the first waveguide portion; forming a second set of index-modifying elements located in proximity to the second waveguide portion, each configured to modify an index of refraction of the second waveguide portion based on a respective electrical signal; forming a second set of modifiable electrical elements, each configured to be modifiable at least once between a conductive configuration that conducts electricity and a nonconductive configuration that does not conduct electricity, and each electrically connected to the electrical power source and to one or more respective index-modifying elements in the second set of index-modifying elements; transmitting a second optical wave portion through the second waveguide portion; determining second optical wave information associated with the second optical wave portion after it traverses the second waveguide portion; and modifying one or more modifiable electrical elements in the first set of modifiable electrical elements or in the second set of modifiable electrical elements, based at least in part on the first optical wave information and the second optical wave information.
Where the second optical wave information is associated with at least one of a second phase or a second optical power of the second optical wave portion.
Where the first optical wave information and the second optical wave information are collectively associated with a difference in phase or a difference in optical power between the first optical wave portion and the second optical wave portion.
In another aspect, in general, an article of manufacture comprises: a first waveguide portion; a first set of index-modifying elements located in proximity to the first waveguide portion, each configured to modify an index of refraction of the first waveguide portion based on a respective electrical signal; an electrical power source configured to supply one or more electrical signals; a first set of modifiable electrical elements, each configured to be modifiable at least once between a conductive configuration that conducts electricity and a nonconductive configuration that does not conduct electricity, and each electrically connected to the electrical power source and to one or more respective index-modifying elements in the first set of index-modifying elements; a closed electrical path, where electrical current flows from the electrical power source and through (1) a first modifiable electrical element in the first set of modifiable electrical elements that is configured to be in a conductive configuration, and (2) a first index-modifying element in the first set of index-modifying elements; and an open electrical path, where no electrical current flows through (1) a second modifiable electrical element in the first set of modifiable electrical elements that is configured to be in a nonconductive configuration, and (2) a second index-modifying element in the first set of index-modifying elements.
Aspects can include one or more of the following features.
Where the first of modifiable electrical elements comprises at least one of a wire bond, a flip-chip bump, a redistribution layer, or a fuse-like element.
Where the first set of index-modifying elements comprise at least one of a thermal phase-shifter, a PIN junction, or a non-centrosymmetric crystal.
Where each index-modifying element in the first set of index-modifying elements is located in proximity to a respective sub-portion of the first waveguide portion and modifies a respective index of refraction of the respective sub-portion of the first waveguide portion.
The article of manufacture further comprises a second waveguide portion; a second set of index-modifying elements located in proximity to the second waveguide portion, each configured to modify an index of refraction of the second waveguide portion based on a respective electrical signal; and a second set of modifiable electrical elements, each configured to be modifiable at least once between a conductive configuration that conducts electricity and a nonconductive configuration that does not conduct electricity, and each electrically connected to the electrical power source and to one or more respective index-modifying elements in the second set of index-modifying elements.
Where the first waveguide portion and the second waveguide portion form a portion of (1) a Mach-Zehnder interferometer, (2) a Michelson interferometer, (3) a multi-mode interferometer, (4) an interferometer comprising three or more optical paths, or (5) an N×M coupler comprising N input ports and M output ports.
Where the first waveguide portion and the second waveguide portion are each optically coupled to a first 2×2 coupler and to a second 2×2 coupler, and the first 2×2 coupler and the second 2×2 coupler each comprise two input optical ports and two output optical ports.
Where the first set of index-modifying elements comprises a first index-modifying element configured to modify the index of refraction of the first waveguide portion by a first amount for an applied voltage across the first index-modifying element; and a second index-modifying element configured to modify the index of refraction of the first waveguide portion by a second amount for an applied voltage across the second index-modifying element that is equal to the first applied voltage across the first index-modifying element; where the second amount is twice as large as the first amount.
Where the first set of index-modifying elements further comprises a third index-modifying element configured to modify the index of refraction of the first waveguide portion by a third amount that is twice as large as the second amount.
Where a third index-modifying element in the first set of index-modifying elements comprises a first portion at a first voltage; and a second portion at a second voltage that is equal to a voltage of a respective modifiable electrical element in the first set of modifiable electrical elements.
Where a fourth index-modifying element in the first set of index-modifying elements comprises a third portion at a third voltage; and a fourth portion at a fourth voltage that is equal to a voltage of a respective modifiable electrical element in the first set of modifiable electrical elements.
Where the third voltage is equal to either a ground voltage of the electrical power source or to the second voltage.
Aspects can have one or more of the following advantages.
The index-modifying configurations (IMCs) disclosed herein can be statically configured to provide a modification to the index of refraction of waveguides, thereby providing enhanced control over the phase of optical waves guided by the waveguides. In some examples, the IMCs can have lower power consumption by utilizing electrical power sources that are also used to electrically power other components on the die or on a companion die.
IMCs can also be utilized to increase the known good die yield of dies produced from semiconductor wafers. In some examples, IMCs can reduce die failure rates that occur due to extinction ratio metrics that characterize, for example, the transmitting ports of photonic integrated circuits (PICs). Furthermore, IMCs can control the operating point of interferometric devices by balancing optical losses in different optical paths (e.g., by modifying the imaginary component of the index of refraction), so as to improve the extinction ratio, to improve the common-mode rejection ratio, or to adjust the optical power splitting ratio (e.g., to achieve higher accuracy of the hybrid angle).
Other features and advantages will become apparent from the following description, and from the figures and claims.
The disclosure is best understood from the following detailed description when read in conjunction with the accompanying drawings. It is emphasized that, according to common practice, the various features of the drawings are not to-scale. On the contrary, the dimensions of the various features are arbitrarily expanded or reduced for clarity.
One challenge in designing photonic integrated circuits (PICs) can be the unpredictability of the phase of optical waves during or after propagation through passive optical components or subcircuits. For example, nanometer scale variations in the spatial dimensions of a waveguide (e.g., in the width or the height) can modify the index of refraction of the waveguide, such that the phase of optical waves guided by the waveguide can substantially deviate from modelled behavior. In general, the absolute phase of an optical wave may be irrelevant while the difference in phase between two optical modes (e.g., two optical modes in the same waveguide, or one mode in a first waveguide and a second mode in a second waveguide) can be important. However, the small variations in the spatial dimensions of one or more waveguides can similarly result in a relative phase difference between two optical modes that is often larger than 2π after sub-micrometer propagation.
In general, there are several approaches that can be used to achieve a desired phase or loss associated with one or more waveguides in a PIC that guide optical waves.
In a first approach, dies that perform below a threshold for one or more metrics (e.g., extinction ratio) can be yielded out during a wafer-level test, thereby resulting in increased cost and decreased manufacturing efficiency.
In a second approach, a voltage or current output by a biasing circuit can be dynamically modified so as to adjust one or more index-modifying elements (AIEs) to correct imbalances in phase modulation that result in phase variation of an optical carrier (e.g., chirp) at the output of a modulator. For example, different PN junction biases in each optical path (i.e., arm) lead to different optical losses, making such balancing possible. Dynamically modifying the output of the biasing circuit, however, can lead to increases in complexity, power consumption, and an increased footprint size. Furthermore, the additional bias applied by the biasing circuit can have an impact on the Vπ of the modulator (i.e., the amount of voltage required to produce a phase change of π), which can affect the chirp resulting from the modulator and may require compromises in other characteristics (e.g., in RF bandwidth, optical bandwidth, impedance, or RF phase velocity). Additionally, the applicability of this approach is limited to PICs comprising a modulator.
In a third approach, high-intensity laser pulses can be applied to one or more waveguide materials (e.g., silicon) so as to perform static modifications to the index of refraction of the waveguide. To perform such modifications, dedicated equipment and process lines may be required, which can be costly or time consuming in a high-volume application. Furthermore, such modifications can increase optical losses in addition to creating a phase shift, which may be undesirable in some applications.
In a fourth approach, the propagation length over which two paths of an optical wave propagate is reduced to nearly zero, thereby reducing the change in relative phase between the two paths. Such a technique, however, can lead to distortions in the optical waves that result in optical losses and can be otherwise limiting by requiring short propagation lengths.
In a fifth approach, PICs can be cointegrated with electronic integrated circuits (EICS) by utilizing flip-chip assemblies or wirebond assemblies, for example. In such examples, one or more digital-to-analog converters (DACs) can be used to modify the index of refraction of a waveguide, such as by applying a thermo-optic effect, a carrier-based effect (e.g., the plasma dispersion effect), or an electro-optic effect (e.g., the Pockels effect). In general, the number of DACs that can be utilized may be limited by: (1) the number of input and outputs available on the PIC (e.g., the number of wire bonds, flip-chip bumps, or through-silicon vias), (2) the footprint the DACs occupy on a companion die, an application-specific integrated circuit (ASIC), or a circuit board, or (3) the power consumption, cost, and complexity the DACs add to an application. Compared to applications that utilize a static electrical power source, a DAC presents additional overhead that can be detrimental for certain applications.
Some of the example approaches for controlling phase and/or loss disclosed herein include index-modifying configurations (IMCs) comprising one or more modifiable electrical elements (MEEs) that are modifiable at least once between a conductive configuration that conducts electricity and a nonconductive configuration that does not conduct electricity. Examples of MEEs include wire bonds, flip-chip bumps, redistribution layers, and fuse-like elements. The IMCs disclosed herein also comprise one or more index-modifying elements (IMEs) that each modify an index of refraction of a waveguide portion based on a respective electrical signal. As used herein, the index of refraction comprises both a real component, which is associated with the phase velocity of an optical wave, and an imaginary component, which is associated with the attenuation (i.e., loss) of an optical wave. The IMCs disclosed herein enable static modification of the real and/or imaginary components of the index of refraction of a waveguide within a PIC. As used herein, a waveguide can be any structure that guides or confines any number of optical waves. In some examples, the waveguide can be a portion of (1) a Mach-Zehnder interferometer, (2) a Michelson interferometer, (3) a multi-mode interferometer, (4) a Fabry-Perot cavity, (5) a ring resonator, (6) a grating, (7) an electro-optic modulator, (8) an interferometer comprising three or more optical paths (also referred to as a multi-leg interferometer), or (9) an N×M coupler comprising N input ports and M output ports.
In some examples, the MEEs of an IMC are fuse-like elements that can be modified from a conductive state to a nonconductive state by applying electrical power through them above a threshold value. After such a modification, the MEEs prevent electrical power from flowing to respective IMEs that modify the phase shift of an optical wave (i.e., the real component of the index of refraction of a waveguide) and/or the optical losses of an optical wave (i.e., the imaginary component of the index of refraction of a waveguide). In some examples, the MEEs can be modified once (i.e., in a “set and forget” manner) during wafer-level testing or during assembly, thereby enabling the production of self-configurable PICs for applications that utilize a static configuration of the index of refraction of a waveguide.
IMCs can take advantage of the availability of fixed (i.e., statically configured) voltages on a PIC or on an EIC electrically connected to the PIC. Examples of such fixed voltages include electrical power sources that provide electrical power to photodiodes, modulators, transimpedance amplifiers (TIAs), drivers, or other companion chips (e.g., voltages such as Vc or Vdd). Examples of IEs include thermal phase-shifters (TPSs), PIN junctions, and non-centrosymmetric crystals (e.g., composed of barium borate or potassium niobate). The voltages provided to the AIEs may be applied (or not applied) in a binary manner by utilizing one or more MEEs that can be modified, for example, by performing laser cuts, laser ablation, or by applying electrical power above a threshold value so as to create an open circuit by blowing one or more fuses that may be composed of small cross-sections of metal traces or metal vias. In some examples, the configurations of the MEEs (i.e., which MEEs are set to a conductive configuration and which are set to a nonconductive configuration) can be determined during or following wafer characterization. In other examples (e.g., where a PIC and one or more voltage sources are cointegrated as separate chips), assembly steps may involve selectively adding or removing wire bonds, flip-chip bumps, or redistribution layers (RDLs), for example, so as to toggle one or more MEEs between a conductive configuration and a nonconductive configuration.
In general, cascading two or more IMEs in series or in parallel can enable high resolution control of the index of refraction being modified. For example, one or more IMEs can be toggled on or off so as to each generate 0.1π phase shifts over a full range of 2π, thus allowing for phase control of an optical wave over a range of 2π and with a resolution of 0.1π. Alternatively, IMEs of different lengths or different efficiencies may be used so as to achieve a desired resolution over a given range with fewer required IMEs (e.g., by being arranged such that the IMEs provide n-bit control of the index of refraction or of the phase shift itself). For example, an IMC with IMEs configured to provide n-bit control can comprise n IMEs that each provide a phase shift that is twice the phase shift of a preceding IME (e.g., a first IME provides a phase shift of 0.25π, a second IME provides a phase shift of 0.5π, and a third IME provides a phase shift of π).
Referring again to
Referring again to
Referring again to
ICs can also be used to modify the splitting ratio of a Mach-Zehnder interferometer (MZI) by controlling the relative phase between the two arms of the MZI. For example, the splitting ratios between a receiver port and a transmitter port of a PIC can be modified by utilizing one or more IMCs (e.g., located in one or both arms of an MZI).
As photonic platforms continue to expand and coherent communication becomes increasingly prevalent in data centers, the importance of yield may increase, such that techniques and designs that increase yield can be highly beneficial. While component improvement may play a role in increasing yield, it can also be important to consider the impact of circuit yield. In the context of such considerations, the ability to electrically disconnect (i.e., turn off) faulty sections of a die without discarding the entire unit can increase yield and efficiency.
Referring again to
Referring again to
In general, IMCs can be configured using a variety of techniques.
While the disclosure has been described in connection with certain embodiments, it is to be understood that the disclosure is not to be limited to the disclosed embodiments but, on the contrary, is intended to cover various modifications and equivalent arrangements included within the scope of the appended claims, which scope is to be accorded the broadest interpretation so as to encompass all such modifications and equivalent structures as is permitted under the law.
