WO2019111051A2

WO2019111051A2 - Active optical switch system with simultaneously activated electro-wetting on dielectric optical switches

Info

Publication number: WO2019111051A2
Application number: PCT/IB2018/001505
Authority: WO
Inventors: Herbert D'HEER; Dries Van Thourhout; Cristina LERMA ARCE; Saurav Kumar
Original assignee: Dheer Herbert; Dries Van Thourhout; Lerma Arce Cristina; Saurav Kumar
Priority date: 2017-12-04
Filing date: 2018-12-04
Publication date: 2019-06-13
Also published as: WO2019111051A3

Abstract

An optical switch unit has first and second elementary optical switches activatable by a single liquid droplet. The optical switch unit includes a first input and a first output. The optical switch unit may be included in an optical circuit that includes a plurality of inputs and a plurality of outputs, wherein at least one of the first and second elementary optical switches lies on a selectable path between a first input of the optical switch unit and a first output of the optical switch unit. A method of operating an optical switch unit includes changing states of first and second elementary switches by moving a liquid droplet, resulting in a change in switch state of the optical switch unit. The optical switch unit may include an elementary switch having two waveguides disposed horizontally and another elementary switch having two waveguides disposed vertically.

Description

ACTIVE OPTICAL SWITCH SYSTEM WITH SIMULTANEOUSLY ACTIVATED ELECTRO- WETTING ON DIELECTRIC OPTICAL SWITCHES

Cross-Reference To Related Applications

This application is being filed on December 4, 2018 as a PCT International Patent Application and claims the benefit of U.S. Patent Application Serial No. 62/594,339, filed on December 4, 2017, and claims the benefit of U.S. Patent Application Serial No.

62/632,040, filed on February 19, 2018, the disclosures of which are incorporated herein by reference their entireties.

Field of the Invention

The present invention is generally directed to optical communications, and more specifically to active optical switch systems based on electro-wetting-on-dielectric (EWOD) optical switches.

Background of the Invention

Optical fiber networks are becoming increasingly prevalent in part because service providers want to deliver high bandwidth communication and data transfer capabilities to customers. As optical networks become more complex, it has become increasingly important to manage optical signals in the network. Many optical signal management functions, such as redirecting signals to bypass faulty components, or opening new channels to facilitate the addition of more users of the network, can be accomplished using active optical switches, such as electro-wetting on dielectric (EWOD)-activated optical switches. Such active optical switches are based on the principles of microfluidics: two fluids with different refractive indices, wherein one fluid typically is a liquid, are moved relative to an adiabatic waveguide coupler. Depending on the location of the fluids relative to the waveguide coupler, the coupler switches between two states, either facilitating the transition of a propagating optical signal from one waveguide to another, in which case the switch is said to be in a“cross-state,” or prohibiting the transition of the signal between the waveguides, in which case the switch is said to be in the“bar state.”

EWOD switches may be implemented in integrated switch networks, on an optical chip, having a large number of ports. In many cases it is desirable to increase the density of the switches so as to reduce the size of the optical chip. There are, however, competing issues that lead to the increase of chip size and complexity. In some applications, the performance of a single optical switch provides satisfactory discrimination between the bar and cross states, in other words the crosstalk between output ports when the switch is in the two states may be sufficiently low. Other applications, however, may require a lower crosstalk than is achievable with a single switch, and so multiple switches may be used to achieve the required level of performance. Additionally, in some cases, an integrated switch network may need to handle two orthogonal polarizations separately, and so the number of EWOD switches in such situations double. Furthermore, the number of electrical elements, such as drivers, electrical traces, electrodes and the like, becomes large for switches with a large number of ports.

It is desirable, therefore, to arrange and to control optical switches in the optical chip in a manner that reduces the number of optical and electrical components, and that also reduces chip size and cost. Summary of the Invention

One embodiment of the invention is directed to an optical switch unit that has a first elementary optical switch configured to be activatable by a first liquid droplet and a second elementary optical switch configured to be activatable by the first liquid droplet. A first input is optically coupled to at least one of the first and second elementary optical switches, and a first output is optically coupled to at least one of the first and second elementary optical switches.

Another embodiment of the invention is directed to an optical circuit that has a plurality of inputs, a plurality of outputs and a plurality of optical switch devices disposed between the plurality of inputs and the plurality of outputs. The optical switch devices are configurable to selectively connect at least one input of the plurality of inputs to at least one output of the plurality of outputs. At least one of the optical switch devices comprises an optical switch unit having a first elementary optical switch configured to be activatable by a first liquid droplet and a second elementary optical switch configured to be activatable by the first liquid droplet. At least one of the first and second elementary optical switches lies on a selectable path between a first input of the optical switch unit and a first output of the optical switch unit.

Another embodiment of the invention is directed to a method of operating an optical switch unit. The method includes changing a state of a first elementary switch of the optical switch unit by moving a liquid droplet and simultaneously changing a state of a second elementary switch of the optical switch unit by moving the liquid droplet.

Changing the state of the first elementary switch and changing the state of the second elementary switch results in changing a switch state of the optical switch unit.

Another embodiment of the invention is directed to an optical switch unit that has a first waveguide disposed at a first distance from a substrate. A second waveguide is positioned over the first waveguide at a second distance from the substrate, the second distance being greater than the first distance. The optical switch unit also includes at least a third waveguide disposed at the second distance from the substrate. A first low-index layer is disposed at a third distance from the substrate, the third distance being greater than the second distance. The first low index layer has a refractive index less than the refractive index of the second waveguide. A recess is formed in the first low index layer over at least the second waveguide to form a liquid-waveguide coupling region over the second waveguide. The optical switch unit also includes a liquid droplet movable into the recess of the first low index layer to affect coupling of optical signals between the first and second waveguides.

Another embodiment of the invention is directed to an optical switch circuit that has a plurality of optical switch units arranged on a substrate and a plurality of waveguides connecting between selected pairs of optical switch units. At least one of the optical switch units comprises a first microfluidically controlled elementary switch having a first waveguide and a second waveguide disposed horizontally relative to the first waveguide, and also has a second microfluidically controlled elementary switch having a third waveguide and a fourth waveguide disposed vertically relative to the third waveguide.

Another embodiment of the invention is directed to an optical switch unit that has a first waveguide, a second waveguide proximate the first waveguide and a third waveguide proximate the second waveguide. A low index layer is provided above the first, second and third waveguides. The low index layer has a recess over at least one of the first, second and third waveguides. A liquid droplet is movable into the recess of the low index layer to affect coupling of optical signals between i) the first and second waveguides and ii) the second and third waveguides.

Another embodiment of the invention is directed to an optical switch unit that has a first waveguide, a second waveguide proximate the first waveguide, a third waveguide proximate the second waveguide, and a fourth waveguide proximate the third waveguide. A low index layer is provided above the first, second, third, and fourth waveguides. The low index layer has a recess over at least two of the first, second, third and fourth waveguides. A liquid droplet is movable into the recess of the low index layer to affect coupling of optical signals between at least one of i) the first and second waveguides and ii) the third and fourth waveguides.

Another embodiment of the invention is directed to a method of making an optical chip. The method includes providing a first layer of relatively high index material over a first layer of relatively low index material and forming a first waveguide in the first layer of relatively high index material. A second layer of relatively low index material is provided over the first layer of relatively high index material having the first waveguide. The second layer of relatively low index material is then planarized. A third layer of relatively low index material is provided on the second layer of relatively low index material, the third layer of relatively low index material having a thickness so that the second layer of relatively low index material and third layer of relatively low index material together have a desired separation thickness. A second layer of relatively high index material is provided over the third layer of relatively low index material. A second waveguide is formed in the second layer of relatively high index material above the first waveguide. A separation distance between the first waveguide and the second waveguide is the desired separation thickness.

Another embodiment of the invention is directed to an optical device that has a substrate, a first low index layer on the substrate and a first waveguide, the first low index layer being positioned between the first waveguide and the substrate. The device has second low index layer, the first waveguide being positioned between the second and first low index layers. There is a second waveguide substantially above the first waveguide, at least a portion of the second waveguide above the first waveguide being parallel to the first waveguide, the second low index layer being positioned between the second waveguide and the first waveguide.

Another embodiment of the invention is directed to a waveguide device that includes a first optical waveguide in a first waveguide layer disposed above a substrate, a second optical waveguide in a second waveguide layer, at least part of the second optical waveguide being disposed below the first optical waveguide, between the first optical waveguide and the substrate. This arrangement permits coupling of light between the first and second waveguides. In some embodiments, a third optical waveguide is disposed in the first waveguide layer and is disposed so as to permit optical coupling between the first and third optical waveguides. The above summary of the present invention is not intended to describe each illustrated embodiment or every implementation of the present invention. The figures and the detailed description which follow more particularly exemplify these embodiments. Brief Description of the Drawings

The invention may be more completely understood in consideration of the following detailed description of various embodiments of the invention in connection with the accompanying drawings, in which:

FIG. 1 schematically illustrates an embodiment of an active optical switch system; FIG. 2 schematically illustrates a cross-sectional view through a portion of an electro-wetting on dielectric (EWOD) active optical switch system;

FIG. 3 A schematically illustrates a first type of optical switch unit configured using three elementary optical switches operating in a first mode, according to an embodiment of the present invention;

FIG. 3B schematically illustrates the optical switch unit configured using three elementary optical switches operating in a second mode, according to an embodiment of the present invention;

FIG. 3C schematically illustrates an equivalent switch unit for the switch unit illustrated in FIGs. 3 A and 3B;

FIG. 4 schematically illustrates a cross-sectional view through a portion of an active optical switch unit having three simultaneously switched optical switches, according to an embodiment of the present invention;

FIG. 5 A schematically illustrates another optical switch unit configured using three elementary optical switches operating in a first mode, according to another embodiment of the present invention;

FIG. 5B schematically illustrates the optical switch unit configured using three elementary optical switches operating in a second mode, according to another embodiment of the present invention;

FIG. 5C schematically illustrates an equivalent switch unit for the switch unit illustrated in FIGs. 5A and 5B;

FIGs. 6A and 6B schematically illustrate cross-sectional views through a portion of an active optical switch system having three simultaneously switched optical switches, in a first switch state (FIG. 6A) and in a second switch state (FIG. 6B), according to another embodiment of the present invention; FIG. 7 A schematically illustrates an optical switch unit configured using four elementary optical switches, according to another embodiment of the present invention;

FIG. 7B schematically illustrates an equivalent switch unit to the switch unit illustrated in FIG. 7A;

FIG. 7C schematically illustrates an optical switch unit configured using four elementary optical switches, according to another embodiment of the present invention;

FIG. 8A schematically illustrates an optical switch unit configured using two elementary optical switches, according to another embodiment of the present invention;

FIG. 8B schematically illustrates an equivalent switch unit to the switch unit illustrated in FIG. 8A;

FIG. 9A illustrates cross-talk of the switch unit shown in FIG. 8A in a first switch state;

FIG. 9B illustrates cross-talk of the switch unit shown in FIG. 8A in a second switch state;

FIG. 10A schematically illustrates an optical circuit using a first arrangement of the optical switch units of the type illustrated in FIG. 8A, according to the present invention;

FIG. 10B schematically illustrates an optical circuit using a second arrangement of the optical switch units of the type illustrated in FIG. 8A, according to the present invention;

FIG. 11 A schematically illustrates a microfluidic arrangement for activating an elementary EWOD optical switch, according to an embodiment of the present invention;

FIG. 11B schematically illustrates a microfluidic arrangement for simultaneously activating two elementary EWOD optical switches, according to another embodiment of the present invention;

FIG. 12A schematically illustrates a microfluidic arrangement for simultaneously activating two elementary EWOD optical switches in the same switch state, according to another embodiment of the present invention;

FIG. 12B schematically illustrates a microfluidic arrangement for simultaneously activating two elementary EWOD optical switches to be in different switch states, according to another embodiment of the present invention;

FIGs. 13 A and 13B schematically illustrate a second type of switch unit using two elementary switches in a parallel arrangement, in a first switch state and a second switch state respectively, according to an embodiment of the present invention; FIG. 14 schematically illustrates a third type of switch unit having parallel optical paths, where each optical path includes at least two elementary switches in a serial arrangement, according to another embodiment of the invention;

FIG. 15 schematically illustrates an embodiment of a polarization-preserving optical circuit using optical switch units that include elementary optical switches for different polarized signals, according to the present invention;

FIG. 16A schematically illustrates an embodiment of a Pi-loss optical switch array implemented in a single layer;

FIG. 16B schematically illustrates an embodiment of a PI-loss optical switch array implemented in two layers, according to an embodiment of the present invention;

FIG. 16C schematically illustrates an embodiment of a cross-bar optical switch network;

FIG. 17A schematically illustrates a cross-sectional view of an elementary optical switch having waveguides arranged in vertical relationship, according to an embodiment of the invention;

FIGs. 17B and 17C schematically illustrate different plan views of the elementary optical switch having waveguides arranged in vertical relationship, according to different embodiments of the invention;

FIGs. 18A and 18B schematically illustrate an optical switch unit formed from two elementary switches, according to an embodiment of the present invention, respectively with a liquid droplet not over and over the elementary switches;

FIG. 19A schematically illustrated an optical switch unit formed from two elementary switches, one of the elementary switches having horizontal coupling and the other elementary switch having vertical coupling, according to an embodiment of the present invention;

FIG. 19B schematically illustrates a cross-sectional view of an embodiment of part of the optical switch unit of FIG. 19 A;

FIG. 20A schematically illustrates an arrangement of two optical switch units, each optical switch unit being formed from one elementary switch having horizontal coupling and another elementary switch having vertical coupling, according to an embodiment of the present invention;

FIG. 20B schematically illustrates a cross-sectional view of an embodiment of part of the optical switch unit of FIG. 20 A; FIG. 21 A schematically illustrates an optical switch unit formed from two elementary switches, where each of the elementary switched has vertical coupling, according to an embodiment of the present invention;

FIG. 21B schematically illustrates a cross-sectional view of an embodiment of part of the optical switch unit of FIG. 21 A;

FIG. 22 presents a graph showing measured crossover transmission loss for waveguides in a single layer;

FIG. 23 presents a graph showing measured crossover transmission loss for waveguides in separate layers;

FIG. 24 presents calculated cross state insertion losses and crosstalk for a vertically coupled elementary switch;

FIG. 25 schematically illustrates a two input, two output generic optical switch;

FIG. 26 presents a graph showing performance of a vertically coupled elementary switch in the bar state;

FIG. 27 presents a graph showing performance of a vertically coupled elementary switch in the cross state;

FIGs. 28A-28G schematically illustrate fabrication steps for making a vertically coupled elementary switch, according to an embodiment of the present invention;

FIGs. 29A-29D schematically illustrate embodiments of an elementary switch that includes three waveguides, according to the present invention; and

FIGs. 30A-30D schematically illustrate embodiments of an elementary switch that includes four waveguides, according to the present invention.

While the invention is amenable to various modifications and alternative forms, specifics thereof have been shown by way of example in the drawings and will be described in detail. It should be understood, however, that the intention is not to limit the invention to the particular embodiments described. On the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention as defined by the appended claims.

Detailed Description

The present invention is directed to systems, devices, and methods that can provide benefits to optical communication networks. More particularly, the invention is directed to active optical switch systems and devices employing electro-wetting on dielectric (EWOD)-activated optical switches. In some applications where it is desired to achieve higher performance than is available from a single EWOD switch, e.g. a level of crosstalk lower than produced by a single switch, it may be useful to operate multiple switches together. For example, two switches arranged in series in many situations will produce a level of crosstalk that is less than the crosstalk produced by a single switch. Multiple elementary EWOD switches can be grouped in what is further called a switch unit with the purpose to reduce the crosstalk. To help reduce the footprint of the switch network, it is important that these switch units are compact.

Additionally, an integrated switch network that needs to handle two orthogonal polarization states separately can use twice the number of elementary switches used by a switch network that uses unpolarized light. In these switch networks it is also important that multiple elementary switches can be grouped in one compact switch unit in order to reduce the footprint of the optical circuit.

As the optical circuit on an optical chip increases in complexity, the number of electrical elements associated with the EWOD switches, such as drivers, electrical traces, electrodes and the like, also increases. The number of independent electrical elements can be reduced if the number of independently controlled elementary switches is reduced, for example, using an architecture where elementary switches of a switch unit share the electrical elements.

An exemplary embodiment of a single, or elementary, active optical switch 100 is schematically illustrated in FIG. 1. The active optical switch 100 incorporates a first waveguide 101 and a second waveguide 102. The first and second waveguides 101, 102 are situated physically closer to one another in a waveguide-light coupling region 103, a region where light propagating along one of the waveguides 101, 102 may couple to the other waveguide 102, 101. Whether light couples between the waveguides 101, 102 depends on the effective refractive index experienced by the light as it propagates along the waveguides 101, 102. The effective refractive index can be altered by positioning a fluid of greater or lesser refractive index close to the waveguide-light coupling region 103 and the waveguide-fluid coupling region 109, discussed further below.

In many embodiments, the active optical switch includes two fluids that are moveable to change the state of the switch. The figure shows a first fluid 104 positioned over the waveguide-light coupling region 103 and the waveguide-fluid coupling region 109. A second fluid 105 is shown generally filling the remaining space of the fluid channel 106. The first and second fluids 104, 105 may be in either a liquid or gaseous phase. The first fluid 104 has a first refractive index and the second fluid 105 has a second refractive index, different from the first refractive index. The first and second fluid 104, 105 may move within fluid channel 106, so for example, the first fluid 104 may move away from waveguide-light coupling region 103 and waveguide-fluid coupling region 109 to the location shown as l04a, with the second fluid 105 generally filling the remaining space in the fluid channel 106. One or more of the inner surfaces of the fluid channel 106 may be coated with anti -wetting coatings, such as coatings shown as 107, 108, to assist in controlling the position of first and second fluids 104, 105 with respect to the waveguide- light coupling region 103 and waveguide-fluid coupling region 109.

In the illustrated embodiment, an optical signal transmitted into the first waveguide 101 is coupled to the second waveguide 102 when the first fluid 104 is positioned close to the waveguide-light coupling region 103 and the waveguide-fluid coupling region 109. This is referred to as the switch’s“cross state.” An optical signal transmitted into the first waveguide 101 is output from the first waveguide 101 when the first fluid l04a is positioned away from the waveguide-light coupling region 103 and the waveguide-fluid coupling region 109, and instead the second fluid 105 is positioned near coupling regions 103, 109. This is referred to as the switch’s“bar state.” Microfluidic optical switches have previously been described, for example in U.S. Provisional Patent Application No. 62/094,506,“Integrated Optical Switching and Splitting for Optical Networks,” filed on December 19, 2014, in U.S. Provisional Patent Application No. 62/116,784, entitled “Remote Control and Power Supply for Optical Networks,” filed on February 16, 2015, and in WO 2015/092064A1,“Adiabatic Coupler,” published on June 25, 2015, all of which are incorporated herein by reference.

A cross-sectional view through a portion of an exemplary embodiment of an elementary active optical switch system 200 is schematically illustrated in FIG. 2. In this embodiment, optical fluids are moved in a fluid channel relative to waveguides using the technique of electro-wetting. A first fluid 201 and a second fluid 202 are disposed within a fluid channel 203 formed between two structures 204, 205. Either of the first and second fluids 201, 202 may be in a liquid or gaseous phase, although at least one of them is a liquid. In the illustrated embodiment, the first fluid 201 is liquid and, in a preferred embodiment, the second fluid 202 is also a liquid. The first fluid 201 has a first refractive index and the second fluid 202 has a second refractive index, different from the first refractive index. The first structure 204 is provided with a common electrode 213, insulated from channel 203 by a first dielectric layer 206, which provides at least partial electrical insulation between the common electrode 213 and the fluids 201, 202 and the fluid channel 203. A first anti -wetting layer 208 may be deposited on the first dielectric layer or substrate 206 to facilitate movement of fluids 201, 202 in the fluid channel 203. The second structure 205 is provided with multiple electrodes 214, 215 that can be activated with an applied voltage independently of each other. A fluidic driving mechanism, generally 212, comprises the common electrode 213 and the independently addressable electrodes 214, 215. In the illustrated embodiment, two independently addressable electrodes 214, 215 are shown, but it will be appreciated that other embodiments of the invention may include a larger number of independently addressable electrodes. It will further be appreciated that the multiple independently addressable electrodes 214, 215 may be located in the first structure 204, while the common electrode could be located in the second structure 205. It will also be appreciated that, in alternative embodiments, it may not be necessary to insulate each electrode from the fluids in an EWOD-type switch, which may require only one electrode to be insulated from the fluids of the switch. Alternative embodiments also may have independently addressable electrodes and a common electrode located in the same substrate, for example, structure 204. Alternative embodiments may also have only independently addressable electrodes, without a common electrode incorporated into the active optical switch system, wherein the independently addressable electrodes are located, for example, in structure 204. In the illustrated embodiment, a second dielectric layer or substrate 207, having an upper surface 217, at least partially insulates electrodes 214, 215 from the fluids 201, 202 and the fluid channel 203. In the illustrated embodiment, the surface 217 is also the bottom surface of the fluid channel 203. A second anti -wetting layer 209 may be deposited on the second dielectric layer or substrate 207, for example on the shared surface 217, to facilitate movement of fluids 201, 202 in the fluid channel 203.

The second substrate 207 contains a first waveguide 210 and a second waveguide 211. An etched recess 216 of the second substrate 207 above the second waveguide 211 exposes the second waveguide 211 at or close to the upper surface 217 of the second substrate 207, on which the second anti -wetting layer 209 may be deposited. The etched recess 216 defines a liquid-waveguide region 207a of the second substrate 207, in which the refractive index of the fluid located above the second waveguide 211 can affect the propagation constant of light passing along the second waveguide 211. The first waveguide 210 is located away from the etched recess 216 of the second substrate 207 and away from the liquid-waveguide coupling region 207a, remaining isolated within the second substrate 207 so that the refractive index of the fluid above the first waveguide 210 has substantially no impact on the propagation constant for light passing along the first waveguide 210. The anti -wetting layers 208, 209 are typically very thin, preferably less than 1 pm and more preferably less than 100 nm, so as to avoid shielding the waveguide 211 from effects of the different refractive indices of the fluids 201, 202.

In the illustrated embodiment, the first fluid 201 has a relatively higher refractive index than the second fluid 202. The first fluid 201 is located within the fluid channel 203 and in the etched recess 216, so that the relatively higher refractive index of the first fluid 201 affects the effective refractive index experienced by light propagating along the second waveguide 211. According to the illustrated embodiment, light can couple between the second and first waveguides 211, 210 when the first fluid 201 is in the etched recess 216. In other words, when the first fluid 201 is in the etched recess 216, the switch is in the cross state. In another switch state, when the first fluid 201 is outside of the etched recess 216, and the second fluid 202 with a relatively lower refractive index is in the etched recess 216, the effective refractive index experienced by light propagating along the second waveguide 211 is changed, preventing coupling of light between waveguides 211, 210, and the switch is in the bar state. It will be appreciated that in alternative embodiments, the first fluid may have a lower refractive index than the second fluid, so that the first fluid could induce the switch to assume the cross state when the first fluid is in the etched region. Alternative embodiments may also employ a first fluid of relatively higher refractive index than the second fluid, and which induces a bar state when in the etched region, and vice versa.

The electro-wetting (EW) effect occurs when an applied potential difference induces a change in the contact angle of a liquid at a surface. In the illustrated

embodiment, when an electric field is generated between, for example, electrodes 213 and 215, the surface tension of liquid 201 lying between the electrodes 213 and 215, can be reduced, allowing it to“wet” the surface it contacts. As in the embodiment illustrated in FIG. 2, because the EW effect is applied to liquid 201 separated from electrodes 213, 214, 215 by dielectric layers 207, 206, this configuration is referred to as electro-wetting on dielectric (EWOD). As discussed above, it will be appreciated that only one electrode need be insulated from the fluids of the switch so as to qualify as an EWOD-type switch.

In the illustrated embodiment, the fluidic driving mechanism 212 selectively applies electric potentials to the electrodes 213, 214, 215 of optical switch 200 to move fluids 201, 202 inside fluid channel 203. For example, in a configuration (not shown) where fluid 201 is above the first waveguide 210, i.e. not in the etched region 216, voltages may be applied to the second electrode 215, together with common electrode 213. Such activation of electrodes 213, 215 may result in fluid 201 moving from a location above the first waveguide 210 to the location shown in FIG. 2, above the second waveguide 211 and in the etched region 216. The movement of fluid 201 causes corresponding movement of fluid 202 inside fluid channel 203. In this way the bar state and cross state of optical switch system 200 can be set.

The use of the EW effect to move liquid droplets is well known, and the use of microfluidics in the control of optical waveguide devices has been described in

W02015/092064A1,“Adiabatic Coupler,” filed on December 21, 2014, incorporated herein by reference, in U.S. Provisional Patent Application NO. 62/094,506,“Integrated Optical Switching and Splitting for Optical Networks,” filed on December 19, 2014, and in U.S. Provisional Patent Application No. 62/116,784, entitled“Remote Control and Power Supply for Optical Networks,” filed on February 16, 2015, both of which have been incorporated by reference. But it will be appreciated that other conformations and configurations of electrode and fluid or liquid can be used to move fluids 201, 202. It will further be appreciated that such approaches can be used to move two or more liquids. For example, if a channel contains two immiscible liquids, separated at an interdiquid interface, movement of one of the liquids via the EW effect can result in both liquids being moved in the channel. The second liquid can be moved along the channel by the EW forces acting on the first liquid, even though the second liquid does not itself exhibit EW behavior. For example, liquids that respond well to EW typically are polar in nature, but the second liquid may be non-polar, yet still be moved as a result of an EW force applied via a polar liquid. The EW technique can also be used to move liquid droplets around a network of microchannels, so long as electrodes are suitably positioned along the different channels.

An approach to solving the problems discussed above, is to employ more than one single EWOD switch, referred to here as an elementary switch, in a switch unit, where the elementary switches within the switch unit are controlled in such a manner that the switch states of the different elementary switches in the switch unit are not independent of each other. Instead, there is a defined set of allowed switch states of the elementary switches in the switch unit, which results in the switch unit performing a well-defined switching function. The elementary switches are typically used as either 1 x 2 or 2 x 2 switches as building blocks for the switch unit architectures. A switch unit, as the term is used herein, is a unit having at least two switch states and that contains a combination of elementary switches. Typically, the states of all of the component elementary switches change when the state of the switch unit is changed.

There are different types of switch unit, sometimes referred to as a dilated switch. In a first type of switch unit, the elementary switches are optically connected in a manner that permits most of the light from one elementary switch to pass to at least another elementary switch. This kind of switch unit employs elementary switches generally in a series configuration and can be used to reduce the crosstalk. In a second type of switch unit, elementary switches are grouped together in a generally parallel configuration and there are independent light paths through the switch unit. This kind of switch unit can be used, for example, in handling light in different polarization states. In a third type of switch unit, which includes elementary switches arranged in a series and parallel configuration, some of the elementary switches are optically connected together in a first group and other elementary switches are optically connected together in a second group that has no optical communication with the first group.

An example of the first type of switch unit is schematically illustrated in FIGs. 3 A and 3B. The switch unit 300 includes a first elementary switch 302 having an input II.

The outputs from the first elementary switch 302 are respectively coupled to a second elementary switch 304 and a third elementary switch 306. One of the outputs from the second elementary switch 304 is labeled 01 and one of the outputs from the third elementary switch is labeled 02. In the switch state shown in FIG. 3 A, all the elementary switches 302, 304, 306 are in the bar state, and so light entering the first elementary switch 302 via II is directed to the second elementary switch 304. In this figure, light propagating within an elementary switch is shown as a dashed line. The light entering the second elementary switch 304 is directed to output 01.

In the switch state shown in FIG. 3B, all the elementary switches 302, 304, 306 are in the cross state, and so light entering the first elementary switch 302 via II is directed to the third elementary switch 306. The light entering the third elementary switch 306 is directed to output 02.

In this arrangement, the switch unit 300 has two states - light entering input II is either switched to output 01 or to output 02. An equivalent switch 310 is schematically illustrated in FIG. 3C. In this arrangement, all the elementary switches 302, 304, 306 are in the same state and are switched together to change the state of the switch unit 300.

A logic table for the switch unit 300 is as follows:

where“I^st layer” refers to the first elementary switch 302 in the switch unit 300 and“2^nd layer” refers to either the second or third elementary switch 304, 306.

An advantage of this configuration of cascaded elementary switches is that the light output at either of the outputs, 01 or 02, has passed through two elementary switches, either 302 and 304 or 302 and 306, and so has reduced crosstalk relative to a single stage switch comprising only a single elementary switch.

One approach to implementing a switch unit as shown in FIGs. 3A-3C is schematically illustrated in FIG. 4. FIG. 4 shows a cross-section through an optical chip showing three sets of waveguides 402a and 402b, 404a and 404b, and 406a and 406b in a layer 400, corresponding to three optical switches. The upper surface 408 of the layer 400 is modified to expose one waveguide of each pair of waveguides, viz. waveguides 402b, 404b and 406b. Above the surface 408 is a fluid channel 410 that contains a first liquid 412 and a second liquid 414. This configuration is similar to that shown above in FIG. 2, except that it shows multiple switches covered by the same droplet of the first liquid 412. In the configuration shown in FIG. 4, the first liquid 412 is shown above all of the pairs of waveguides 402a and 402b, 404a and 404b, and 406a and 406b, with the result that all corresponding elementary switches are in the same switch state, e.g. the bar state.

Replacement of the first liquid 412 above the waveguides 402a and 402b, 404a and 404b, and 406a and 406b by the second liquid 414 results in each of the corresponding switches changing state e.g. from the bar state to the cross state or vice-versa. Thus, all three switches are said to be switched simultaneously. In the context of this disclosure, “simultaneous” switching of optical switches means that the switches are actively controlled by the movement of one droplet of liquid. It is understood that, over a short time, the switches will change state sequentially as the droplet of the first liquid 412 is moved past the waveguides 402a and 402b, 404a and 404b, and 406a and 406b, and so there is a finite time over which the liquid droplet moves from one equilibrium position to another, and over which the elementary switches in the switch unit convert from one switch state to another. However, such change in state is considered to be simultaneous because it takes place within the time for the single droplet of liquid to be moved from a first equilibrium position corresponding to one state, e.g. II - 01, to a second equilibrium position corresponding to second state, e.g. II - 02.

It will be appreciated that the number of switches being activated by the same liquid droplet is not limited to the number shown in the drawing, but may be two, four or more, depending on the size of the droplet and the separation between elementary switch waveguides. : In this disclosure, a switch is activated by a first liquid when the first liquid is in close proximity to a waveguide of the switch so as to affect the effective refractive index of that waveguide, thus controlling whether the switch is either in cross or bar state.

The elementary switches need not all be in the same state within a switch unit, as is exemplified in the switch unit 500 shown in FIGs. 5A and 5B. The switch unit 500 generally operates in a manner similar to the switch unit 300, by directing light from a single input, II to either of two outputs 01 and 02. In this embodiment, however, the first elementary switch 502 is in a different switch state from the second elementary switches 504 and 506. In the FIG. 5A, the switch unit 500 is shown with the first elementary switch

502 in a bar state, so that light entering at input II is directed to the second elementary switch 504, which is in a cross state and directs the light to 01. In FIG. 5B, the switch unit 500 is shown with the first elementary switch 502 in a cross state, so that light entering at input II is directed to the third elementary switch 506, which is in a bar state and directs the light to 02. An equivalent switch 510 is shown in FIG. 5C, where light entering at II is directed to either 01 or to 02.

A logic table for the switch unit 500 is as follows:

where“I^st layer” refers to the first elementary switch 502 in the switch unit 500, as it is the first elementary switch experienced by a light signal upon entering the switch unit 500. “2^nd layer” refers to the second and third elementary switches 504, 506, as they are the second elementary switch an optical signal passes through after entering the switch unit 500. One approach to implementing a switch unit as shown in FIGs. 5A-5C is schematically illustrated in FIGs. 6 A and 6B. FIGs. 6 A and 6B show a cross-section through an optical chip showing three sets of waveguides 602a and 602b, 604a and 604b, and 606a and 606b in a layer 600, corresponding to three optical switches. The upper surface 608 of the layer 600 is modified to expose one waveguide of each pair of waveguides, viz. waveguides 602b, 604b and 606b. Above the surface 608 is a fluid channel 610 that contains a first liquid 612 and a second liquid 614. In the configuration shown in FIG. 6A, the first liquid 612 is shown above two of the pairs of waveguides,

602a and 602b, and 604a and 604b, while the second liquid 614 is above the third pair of waveguides, 606a and 606b, with the result that two of the corresponding switches are in the same switch state, e.g. the bar state, while the other switch is in the opposite state, e.g. the cross state.

Replacement of the first liquid 612 above the waveguides 602a and 602b, and 604a and 604b by the second liquid 614, and replacement of the second liquid 612 above the waveguides 606a and 606b with the first liquid 612, as shown in FIG. 6B, results in each of the corresponding switches changing state e.g. from the bar state to the cross state or vice-versa. Thus, all three switches are switched simultaneously.

The present invention contemplates other designs that use different numbers of elementary switches and/or different configurations, in addition to those shown here. For example, a switch unit may have four elementary switches, where two elementary switches are covered by the first liquid while the other two switches are covered by the second liquid.

Another embodiment of switch unit 700 is schematically illustrated in FIG. 7A.

The switch unit 700 employs four elementary switches 702, 704, 706 and 708. The elementary switches 702, 704 are in the first layer of the switch unit 700 and the elementary switches 706, 708 are in the second layer of the switch unit 700. In this configuration, each output from each of the elementary switches 702, 704 in the first layer is connected to an input of each of the elementary switches 706, 708 in the second layer.

The switch unit 700 uses one input, II, of the first elementary switch 702 and one input, 12, of the second elementary switch. It also uses one output 01 of the third elementary switch 706 and one output 02 of the fourth elementary switch 708. In this switch unit 700 light input at the first input II can be delivered to either of the outputs 01, 02, and light input at 12 is delivered to the other of the outputs 02, 01. An equivalent optical switch circuit 710 is schematically illustrated in FIG. 7B. A logic diagram for switch unit 700, where all the elementary switches are in the same state, is as follows:

By using different inputs to and/or outputs from the elementary switches, a switch unit using a similar architecture may be implemented where the elementary switches in the first layer are in a different state from the switches in the second layer. For example, a switch unit 750, schematically illustrated in FIG. 7C, can be implemented to have the same equivalent function as switch unit 700. In such a case, the logic table for the switch unit 750 is as follows:

Another embodiment of switch unit 800 is schematically illustrated in FIG. 8A.

The switch unit 800 employs two elementary switches 802, 804, in a series arrangement. The first elementary switch 802 is in the first layer of the switch unit 800 and the second elementary switch 804 is in the second layer of the switch unit 800. In this configuration, an output from the first elementary switch 802 is an output of the switch unit while the other output of the first elementary switch 802 is connected to an input of the second elementary switch 804. In this embodiment of switch unit 800, no signal passes from II to 01

The logic diagram for this switch unit 800 is as follows:

In the table, the term“n/a” means that the state of the switch does not affect the output to which the light is directed. However, if both elementary switches 802, 804 are in the bar state, light can propagate from II to 02 at the same time as light propagates from 12 to 01. If both elementary switches 802, 804 are in the cross state, then light passes from 12 to 02. The equivalent switch 810 is shown in FIG. 8B.

Some of the configurations of the switch unit discussed herein may operate with reduced crosstalk relative to single elementary switches. Crosstalk is an unwanted signal that is transmitted out of one output port when the signal is intended to be transmitted out of another output port. For example, with regard to the elementary EWOD switches discussed herein, when the elementary switch is in the bar state, a certain portion of the signal, XTb, is output via the cross state output. Likewise, when the elementary switch is in the cross state, a certain portion of the signal, XTc, is output via the bar state port.

The crosstalk of the switch unit 800, shown in FIG. 8A, is discussed with reference to FIGs. 9 A and 9B. In FIG. 9 A, the elementary switches 802, 804 of the switch unit 800 are each in the bar state. The dashed line shows the crosstalk signal 902 exiting the elementary switch 802 to the second elementary switch 804. The power of the crosstalk signal 902 is XTb (measured in dB down from the main output signal). The crosstalk signal 902 is then fed into the second elementary switch 804, which is also in the bar state. A fraction of the crosstalk signal 902, XTb, becomes a second crosstalk signal 904 within the second elementary switch 804, and is transmitted out of the output 02 of the second elementary switch 804. Thus, the crosstalk signal output at 02, resulting from an input applied at II, is 2XTb.

In FIG. 9B, the elementary switches 802, 804 of the switch unit 800 are in the cross state. The dashed line shows the crosstalk signal 912 exiting the elementary switch 802. The power in the crosstalk signal 912 is XTc, measured in dB down from the power of the main output signal. The crosstalk signal 912 is then fed to the output 01. Thus, in this state, output 01 receives a signal from 12 whose magnitude is XTc. Likewise, the output 02 receives a crosstalk signal 914, whose magnitude is XTc, from input II. Thus, the crosstalk for this configuration of switch unit 800 is no greater than the crosstalk of a single elementary switch and, under certain switching conditions, is less.

Switch units as discussed herein may be arranged in a network to form switch circuits. For example, FIG. 10A shows an embodiment of a 4x4 cross-bar network 1000 formed using a 4 x 4 matrix of switch units 800. The rows of switch units 800 are designated (1), (2), (3) and (4), while the columns are designated (a), (b), (c), and (d). Thus, for example, the third switch unit from the left in the top row is designated switch unit (l)(c).

In the cross-bar arrangement, switch units generally use the output from the switch unit above as one input and an output from the switch unit to the left as another input. Further, in the cross-bar arrangement, there is generally only one switch unit per row in the bar state and only one switch unit per column in the bar state. For example, if the following switch units are in the bar state: (l)(c), (2)(a), (3)(d) and (4)(b), then the connections between the input and output ports are as follows:

The figure illustrates the optical paths taken through the switch units of optical circuit 1000 by the four optical signals between the circuit inputs and outputs.

It will be appreciated that changing of the states of certain switch units in the network 1000 can result in a remapping of the input ports II, 12, 13, 14 to different output ports 01, 02, 03, 04. In addition, different cross-bar networks can be assembled using other types of switch units, not only the switch units 800 illustrated in FIG. 8A.

An embodiment of a different type of switch network, a path-independent loss network 1050, is schematically illustrated in FIG. 10B. The path-independent loss network 1050 is, in this embodiment, implemented using the type of switch unit 800 illustrated in FIG. 8A, and is shown as a 4 x 4 network, although other sizes of network may be used.

In this type of network, the outputs from a switch unit in a given column and row generally are directed to a switch unit in the next column, but one row up, and to a switch unit in the next column, but one row down. For example, the switch unit in the second column and second row, (2)(b), has an output connected to the switch unit at position (l)(c) and another output connected to the switch unit at (3)(c). If a switch unit cannot connect to a switch unit in an upper or lower row, it connects to the next switch unit in the same row. For example, the switch unit at position (l)(b) connects to the switch units at (l)(c) and (2)(c). Likewise, the switch unit at position 4(c) connects output to the switch units at (3)(d) and (4)(d). Also, in this embodiment of switch network, a switch unit is inverted relative to the vertically and horizontally adjacent neighboring switch units.

Thus, for example, the switch unit at (2)(b) is inverted relative to the adjacent switch units at positions (l)(b), (2)(a), (2)(c) and 3(b).

In the illustrated embodiment of network 1050, when the switch units 800 are in the following states:

the mapping from the input ports to the output ports is

The signals propagating through the switch units 800 are shown as solid lines. The effect of crosstalk can be reduced in this configuration. An example of this is presented in FIG. 10B, which shows crosstalk signals 1052 as dashed lines. In this embodiment, the crosstalk signals 1052 are directed to unused output channels 1054.

The elementary switches in a switch unit having multiple elementary switches may be configured relative to a microfluidic management system such that no, or very little, extra space is required compared to a switch unit having only one single elementary switch. FIG. 11 A schematically illustrates a single elementary switch 1102 of a switch unit 1100, having two inputs II and 12, and two outputs 01 and 02. A liquid barrier 1104, i.e. a barrier to liquid, partially surrounds the switch unit 1100 to contain the liquid droplet 1106. An exemplary configuration of a switch unit 1120 formed of two elementary switches 1122 is schematically illustrated in FIG. 11B. The two elementary switches 1122 are covered by a single liquid droplet 1126 and the switch unit 1120 is delineated from neighboring switch units by a liquid barrier 1124. The switch unit 1120 has two inputs II and 12 and two outputs 01 and 02 and, in the illustrated embodiment, is configured like switch unit 800 illustrated in FIG. 8A, although it may be configured with different numbers of elementary switches 1122, for example like the switch unit embodiments illustrated in FIGs. 5A and 7A, or in some other arrangement.

As has been discussed above, multiple elementary switches within a switch unit may be activated using a single droplet simultaneously to be in the same state, or one or more of the elementary switches may be activated simultaneously to be in the opposite state from one or more of the other elementary switches in the switch unit. These different approaches to activation are now discussed with reference to FIGs. 12A and 12B. FIG. 12A schematically illustrates a switch unit 1200 having multiple elementary switches 1202 covered, and activated by, a liquid droplet 1204 in one switch state. The switch unit includes a liquid wall 1206 to contain the liquid droplet 1204. This approach is similar to that for the switch unit 1120 schematically illustrated in FIG. 11B. Thus, when the liquid droplet 1204 covers the elementary switches 1202, they are in one state, and when the liquid droplet is moved to a position away from the elementary switches 1202, for example as droplet 1204’ shown by a dashed line, the elementary switches 1202 are activated to another state. In the embodiment illustrated in FIG. 12B, a switch unit 1220 includes two elementary switches l222a and l222b partially surrounded by a liquid wall 1226. A different liquid is above each elementary switch l222a, l222b. In the illustrated example, a droplet 1224 of a first liquid is above elementary switch l222a, while a second fluid, which may be an ambient fluid surrounds the droplet 1224 and covers the second elementary switch l222b. Thus, in this configuration, the first elementary switch l222a is in a first state while the second elementary switch l222b is in a second state, different from the first state. For example, the first elementary switch l222a may be in a bar state while the second elementary switch l222b is in the cross state. When the droplet 1224 is moved to a second position, shown in dashed lines as position 1224’, the droplet covers the second elementary switch l222b while the second fluid covers the first elementary switch l222a, and so the states of the two elementary switches l222a, l222b are said to be simultaneously changed. For example, the first elementary switch l222a may change from the bar state to the cross state while the second elementary switch l222b changes from the cross state to the bar state.

This approach has an additional advantage in network configurations where, in a group of switch units, only one switch unit needs to be in a first state, while all the others need to be in a second state. For example, in a cross-bar network configuration, each row has only one switch unit in the bar-state and only one column has a switch unit in the bar state. When the switch unit is designed such that the liquid droplet produces the bar state, it may be possible to share one droplet among the switch units in one row or column.

When the switch unit is designed such that the liquid droplet changes the state of the switch unit to the cross state, then all switch units in each row and column, except one, are covered by a liquid droplet. The droplet associated with the“uncovered” switch unit may be moved to a storage location which can be shared among all switch units in a row or column. The concepts of sharing a single liquid droplet in a row or column, and the use of common storage space are explained further in U.S. Provisional Patent Application No. 62/331,777, titled“Integrated Optical Switch Network with high Performance and Compact Configuration” and filed on May 4, 2016, which is incorporated herein by reference. The sharing of liquid droplets in a row or column, and the use of common storage space can result in the further reduction of the footprint of the optical circuit.

Another type of switch unit, referred to above as a second type of switch unit, includes elementary switches that are not optically connected to each other, i.e. lie on parallel optical paths through the switch unit, but which are switched simultaneously. One embodiment of such a switch unit 1300 is schematically illustrated in FIGs. 13A and 13B. In the configuration shown in FIG. 13A, the two elementary switches l302a and l302b of the switch unit are in the bar state. Thus, light entering the switch unit 1300 at input II is output from the bar output of the first elementary switch l302a, i.e. output Ola. Likewise, light entering the switch unit 1300 at input 12 is output from the bar output of the second elementary switch 1302b, i.e. at output 02a.

When the switch unit 1300 is in its other state, the elementary switches l302a, l302b are both in the cross state. Thus, light entering the first elementary switch l302a via input II is output from the switch unit via output Olb, while light entering the second elementary switch l302b via input 12 is output from the switch unit 1300 via output 02b. Thus, the elementary switches 1302a and 1302b are switched simultaneously, for example using one of the simultaneous switch activation techniques discussed above.

It will be appreciated that other embodiments of this second type of switch unit, where there are separate, independent optical paths for different optical signals through the switch unit, but where the optical switches along each switch path are switched

simultaneously, may be employed. For example, additional elementary switches may be added on parallel optical paths through the switch unit. In other embodiments, one or more of the elementary switches may be in a different state from one or more of the remaining elementary switches in the switch unit. For example, in a switch unit containing two elementary switches, one of the elementary switches may be in the cross state while the other is in the bar state, and vice versa.

One particular application for this second type of switch unit is in providing an optical circuit capable of handling orthogonally polarized light signals substantially equally, using elementary switches that have been optimized for only one polarization state. An embodiment of an optical circuit 1500 for handling orthogonally polarized light signals is schematically illustrated in FIG. 15. In this embodiment, the switch units 1502, illustrated by dashed lines, each contain two elementary switches 1504 in the manner described above for switch unit 1300 shown in FIGs. 13A and 13B. An input optical signal is split into two orthogonal polarization states in the input PSR unit 1506 and one of the polarized components has its polarization rotated so that its polarization is parallel to the other polarization component, also in the input PSR unit 1506. Thus, the input PSR 1506 produces two separate optical signals having the same polarization. The optical circuit 1500 may contain switch units 1502 comprising a single type of elementary switch 1504 optimized to operate with light in a single polarization state, the polarization state of the signals. Furthermore, the liquid activation approaches discussed above may be implemented so that the elementary switches 1504 in a switch unit 1502 are operated simultaneously to be in the same switch state.

In the illustrated embodiment, the optical circuit 1500 is configured as a path- independent loss network so that light signals incoming to particular input ports can be directed to selected output ports in a manner similar to that discussed above with reference to the optical circuit shown in FIG. 10B. An incoming optical signal is split into two orthogonally polarized signals, and then the polarization of one of the signals l508a is rotated to be the same as the polarization of the other optical signal l508b. The polarized optical signals l508a, l508b propagate through the optical circuit 1500 in parallel with each other to an output PSR unit 1510, where the polarization of one of the signals l508a is rotated to be orthogonal to that of the other polarized signal l508b, and the orthogonal polarization signals can then be combined in the output PSR and output as a mixed polarization output signal 1512. In this manner, the polarization-dependent losses through the switches of the optical circuit 1500 are the same for both polarization signals , and so the mixed polarization output signal 1512 preserves the polarization state of the mixed polarization signal input to the optical circuit 1500.

The illustrated embodiment shows only a single optical signal entering the optical circuit 1500 at input 12 and leaving the optical circuit at output 03. It will be understood that additional optical signals may be applied to the other inputs II, 13, 14 of the optical circuit 1500, and that these signals may be respectively mapped to an output selected from 01, 02 and 04. It will further be appreciated that the mapping of inputs to outputs may be altered through the selective activation of specific switch units.

In other embodiments, there may be multiple, independent optical paths through a switch unit, where each independent optical path is itself switchable. An embodiment of such third type of switch unit 1400 is schematically illustrated in FIG. 14. In this embodiment, the switch unit 1400 includes two independent switch paths like those shown for the single switch unit 800 shown in Fig. 8A. The first independent switch path uses inputs II and 12, elementary switches l402a, l402b, and has outputs 01 and 02.

Likewise, the second independent switch path uses inputs 13 and 14, elementary switches l402c and l402d, and has output 03 and 04. Each independent switch path operates in the manner described above for switch unit 800, but the relationship between the switch states of the first independent switch path and the second independent switch path remains constant, since the elementary switches l402a, l402b, l402c and l402d are all switched simultaneously.

Another embodiment of the invention is directed to optical circuits that contain waveguides in more than one layer. Multi-layer waveguide structures can reduce circuit losses by eliminating crossovers. An example of a single layer Pi-loss network 1600 is schematically presented in FIG. 16A. The illustrated embodiment comprises a 4 x 4 array of optical switches 1602. The network 1600 includes inputs 1604 to switches 1602 on the left side of the array and outputs 1606 from switches 1602 at the right side of the array.

The optical switches 1602 couple light between waveguides located in the same plane, and so the coupling between waveguides in the switches 1602 may be called“horizontal coupling” or“planar coupling.” Coupling waveguides 1608 couple between sequential switches 1602 in the network. The coupling waveguides 1608 are all in the same plane so they intersect at crossovers 1610. Since the loss at each crossover 1610 can be as high as 0.2 dB - 0.3 dB, the signal loss on traversing the switch network 1600 can be significant, especially for larger arrays.

An embodiment of a two-layer Pi-loss network 1620 is schematically illustrated in FIG. 16B. The illustrated embodiment comprises a 4 x 4 array of optical switches 1622. The network 1620 includes inputs 1624 to switches 1622 on the left side of the array and outputs 1626 from switches 1622 at the right side of the array. In this case, the optical switches 1622 couple light vertically between waveguides, rather than coupling between two in-plane waveguides as in the optical switches 1602 in the previous network 1600. Such optical switches may be referred to as vertically coupled optical switches. In the illustrated embodiment, the waveguides 1628 connecting to, and between, switches 1622 are located in one of two different waveguide layers, illustrated by solid lines for waveguides l628a in the first waveguide layer and dotted lines for waveguides l628b in the second waveguide layer. Thus, the waveguides l628a and l628b are vertically separated at those regions 1630 where they cross-over each other, and so cross-over losses are reduced relative to the network 1600 illustrated in FIG. 16 A, where waveguides are all in the same plane. At the top and bottom rows of the network 1620, passive vertical couplers 1632 may be used to couple light between waveguides l628a in the first waveguide layer and waveguides l628b in the second waveguide layer. Where the optical switches 1622 are optical switch units comprising more than one elementary optical waveguide switch, as discussed above, then the two-waveguide layer optical circuit 1620 can provide lower insertion losses and improved crosstalk compared to a single- waveguide layer circuit.

An embodiment of another type of optical network, a crossbar network 1640 is schematically illustrated in FIG. 16C. The illustrated embodiment comprises a 4 x 4 array of optical switches 1642. The network 1640 includes inputs 1644 to switches 1642 on the left side of the array and outputs 1646 from switches 1642 at the right side of the array. This type of array may be implemented using planar coupling optical switches or vertically coupling, multilayer, optical switches. The waveguides 1648 coupling between different optical switches 1642 do not include crossovers in this network 1640. However, this type of network also benefits from the use of dilated switches with reduced crosstalk.

An embodiment of a vertically coupled elementary optical switch 1700 is illustrated in FIGs. 17A and 17B. FIG. 17A schematically illustrates a cross-section through a vertically coupled optical switch 1700. The vertically coupled elementary optical switch 1700 may be formed to include a low index growth layer 1704 over a substrate 1702. Over the growth layer 1704 is a first waveguide layer 1706, having a higher refractive index than the growth layer 1704. An interlayer 1708 separates the first waveguide layer 1706 from a second waveguide layer 1710. The interlayer 1708 has a lower refractive index than the waveguide layers 1706, 1710 to help confine the optical signal to the waveguides. In one embodiment, the substrate 1702 is formed of silicon, the growth layer 1704 formed of silicon dioxide, the first and second waveguide layers 1706, 1710 formed of silicon nitride and the interlayer 1708 formed of silicon dioxide. It will be appreciated that the vertically coupled elementary optical switch 1700 may be formed of other materials. For example the waveguide layers 1706, 1710 may be formed of silicon or the interlayer 1708 may be formed of another material having a lower refractive index than the waveguide layers 1706, 1710, for example a benzocyclobutene-based polymer. In another approach, the waveguide layers 1706, 1710 may be formed of doped silica, with the surrounding layers 1704, 1708 formed of a lower index material, such as un-doped silica or a polymer.

The dimensions of the different layers and waveguides may be selected to optimize performance based on the specific materials employed. In some embodiments the thickness of the interlayer 1708 is less than 1.1 pm and may be around 1.0 pm, i.e. in the range 1 pm ± 0.05 pm, in order to permit coupling of light between the two waveguides 1712, 1714. For carrying optical signals having a wavelength in the range 1250 nm to 1650 nm, the waveguides may be formed of silicon nitride with a height of e.g. 300 nm - 400 nm and have a width in the range 0.5 mih - 1.5 mih.

The different layers of the vertically coupled optical switch 1700 may be formed using standard vacuum planar growth techniques, for example chemical vapor deposition (CVD), or a related technique such as plasma-enhanced chemical vapor deposition (PECVD). The structures of the first waveguide 1712 and the second waveguide 1714 may be formed using standard photolithographic techniques, including masking and etching. If a polymer material is used for any of the layers, such as the interlayer, the polymer layer may be deposited using any suitable technique, including spin coating.

The upper surface of the second waveguide layer 1710 may be left uncoated, or provided with a thin coating, such as an anti -wetting layer, up to a few lOs of nm thick, so that the propagation conditions of light propagating along the second waveguide 1714 may be affected by the presence of a liquid droplet 1716 over the second waveguide 1714. The first waveguide 1712 is buried sufficiently far below the upper surface of the optical switch 1700 that propagation conditions are affected by the presence of the liquid droplet 1716 to a negligible degree.

A plan view of the waveguides 1712, 1714 of the vertically coupled optical switch 1700 is shown in FIG. 17B. At the ends of the switch 1700 the waveguides are laterally separated but are positioned in a substantially vertical relationship in the central coupling region 1720, where optical coupling between the waveguides 1712, 1714 takes place. A plan view of another embodiment is schematically illustrated in FIG. 17C, in which the waveguides 1712, 1714 crossover each other in the region 1720, which may be advantageous in some circuit configurations.

The dilated switch of the type shown in FIGs. 8A and 8B, can be implemented on an optical chip in a manner schematically shown in FIGs. 18A and 18B. The dilated switch 1800 comprises two elementary switches 1802, 1804, outlined with dotted lines. The first elementary switch 1802 has a first waveguide 1806 and a second waveguide 1808. The first waveguide 1806 forms a first input 1810 and a first output 1812 of the dilated switch 1800. The second waveguide 1808 couples as an input to the second elementary switch 1804. A third waveguide 1814 forms a second input of the dilated switch 1800, and a second input to the second elementary switch 1804.

In the illustrated embodiment, light does not couple between waveguides in the elementary switch when a liquid droplet 1818 is not present over the switch 1800, as shown in FIG. 18 A, where the liquid droplet 1818 has a higher refractive index than the ambient fluid. Thus, light entering the dilated switch 1800 along the first input 1810 exits via the first output 1812, while light entering the dilated switch 1800 via the third waveguide 1814 exits via the second output 1816, as shown by the arrows.

FIG. 18B shows the state of the switch 1800 when the droplet 1818 is positioned over the dilated switch 1800. In this condition, light entering the dilated switch along the first waveguide 1810 is coupled to the second waveguide 1808 in the first elementary switch 1802 and passes into the second elementary switch 1804, where it is coupled to the third waveguide 1814 and exits via the second output 1816, as shown by the arrows associated with the waveguides.

A multi-layer embodiment of dilated switch 1900 is schematically illustrated in

FIGs. 19A and 19B. The dilated switch 1900 is formed of two elementary switches 1902, 1904. A first waveguide 1906 in the first elementary switch 1902 provides a first input 1910 and first output 1912 as before. A second waveguide 1908 couples between the two elementary switches 1902, 1904. A third waveguide 1914 provides a second input to the second elementary switch 1904 and provides the second output 1916 from the dilated switch 1900.

In this embodiment, the first elementary switch 1902 is vertically coupled while the second elementary switch 1904 is horizontally coupled. Accordingly, the first waveguide 1906 (dashed line) is not in the same plane as the second and third waveguides, 1908, 1914 (solid lines). A cross-section through the waveguide structure 1920 of the dilated switch 1900 is schematically illustrated in FIG. 19B. The waveguide structure 1920 is formed on a substrate 1922 on which is a growth layer 1924 formed of a relatively low refractive index material. A first waveguide layer 1926, formed of a relatively high refractive index material is provided above the first growth layer 1924. The first waveguide layer 1926 includes the first waveguide 1906, formed as a ridge on the waveguide layer 1926. In other embodiments, waveguides may be formed as isolated strip waveguides that are not part of a continuous waveguide layer.

An interlayer 1930 separates the first waveguide layer 1926 from the second waveguide layer 1932. The refractive index of the interlayer 1930 is lower than the refractive indices of the first and second waveguide layers 1926, 1932. The second waveguide layer 1932 includes part of the second waveguide l908a provided vertically above the first waveguide 1906. Here,“vertical” means in a direction perpendicular to the substrate 1922. In a selected switch state, for example when a liquid droplet 1915 is over the optical switch, light couples vertically between the first and second waveguides 1906, l908a of the first elementary switch 1902.

The second waveguide layer 1932 also includes a second part of the second waveguide l908b and the third waveguide 1914, which form part of the second

elementary switch 1904 and permit optical coupling of light between the waveguides l908b, 1914 in a horizontal direction, parallel with the plane of the second waveguide layer 1932, i.e. in a direction parallel to the substrate 1922. The waveguide portions l908a, l908b of the first and second elementary switches 1902, 1904, respectively, are at or close to the upper surface of the waveguide structure 1920 and so the propagation properties of light passing along these waveguide portions l908a, l908b can be affected by a liquid droplet positioned in a recess 1940 of the cover layer 1936, at the liquid- waveguide coupling region. The cover layer 1936 selectively covers portions of the second waveguide layer 1932 which are desired to remain unaffected by the presence of the liquid droplet 1915. Thus, the structure 1920 contains two microfluidically activatable switches, in a dilated switch unit, one of which provides planar or horizontal optical coupling and the other of which provides vertical optical coupling.

Another embodiment of dilated switch is schematically illustrated in FIGs. 20A and 20B. In this embodiment, two dilated switches are fabricated on an optical chip close together and use the same liquid droplet for activation. To reduce the footprint of the switches, the vertically coupled elementary switches may be placed adjacent to each other. FIG. 20A shows a first dilated switch 2000 formed from first and second elementary switches 2002, 2004, and a second dilated switch 2006, formed from third and fourth elementary switches 2008, 2010.

The first elementary switch 2002 of the first dilated switch 2000 has a first waveguide 2012 and a second waveguide 2014. The first waveguide 2012 forms a first input 2016 and a first output 2018 of the first dilated switch 2000. The second waveguide 2014 couples as an input to the second elementary switch 2004. A third waveguide 2020 forms a second input to the second elementary switch 2004 and a second output 2022 from the first dilated switch 2000. In the illustrated embodiment, the first dilated switch 2000 is covered by a liquid droplet 2036, and so light is coupled between the first waveguide 2012 and a portion of the second waveguide 20l4a in the first elementary switch 2002. Thus, an optical signal entering the dilated switch 2000 along the first waveguide 2012 is coupled from the first elementary switch 2002 via the second waveguide 2014 to the second elementary switch 2004, where it is coupled horizontally into the third waveguide 2020

The first elementary switch 2008 of the second dilated switch 2006 has a fourth waveguide 2024 and a fifth waveguide 2026. The fourth waveguide 2024 forms a first input 2028 and a first output 2030 of the second dilated switch 2006. The fifth waveguide 2026 passes through the first elementary switch 2008 of the second dilated switch 2006. The fifth waveguide 2026 couples as an input to the second elementary switch 2010 of the second dilated switch 2006. A sixth waveguide 2032 forms a second input 2033 to the second elementary switch 2010 of the second dilated switch 2006 and a second output 2034 from the second dilated switch 2006. In the illustrated embodiment, the second dilated switch 2006 is covered by the liquid droplet 2036, and so light entering the dilated switch 2006 via the fourth waveguide 2024 is vertically coupled to a portion of the fifth waveguide 2026a in the first elementary switch 2008. Also, light entering the second elementary switch 2010 along the fifth waveguide 2026 is horizontally coupled from the portion of the fifth waveguide 2026b in the second elementary switch 2010 to the sixth waveguide 2032. Thus, an optical signal entering the second dilated switch 2006 along the fourth waveguide 2024 is coupled from the first elementary switch 2008 via the fifth waveguide 2026 to the second elementary switch 2010, where it is coupled vertically into the sixth waveguide 2032.

FIG. 20B schematically presents a cross-section through optical chip 2040 containing the dilated switches 2000 and 2006, showing the waveguide structure. In this embodiment the optical chip is formed on a substrate 2042 on which is a low index growth layer 2044. A first waveguide layer 2046 lies on top of the growth layer 2044 and contains the first waveguide 2012 of the first dilated switch 2000 and the fourth waveguide 2024 of the second dilated switch 2006. A low index interlayer 2048 lies between the first waveguide layer 2046 and the second waveguide layer 2050. A cover layer 2052 overlies the second waveguide layer 2050. The cover layer 2052 is provided with a recess 2054 which forms a liquid-waveguide coupling region that enables the liquid droplet 2036 to closely approach the waveguides 2014, 2026 so as to enable optical coupling among the waveguides of the dilated switches 2000, 2006.

Another multilayer embodiment of a dilated switch 2100 is schematically illustrated in FIGs. 21A and 21B. The dilated switch 2100 comprises two vertically- coupled elementary switches 2102, 2104, outlined with dotted lines. The first elementary switch 2102 has a first waveguide 2106 and a second waveguide 1208. The first waveguide 2106 (dashed line), which is on a first level, forms a first input 2110 and a first output 2112 of the dilated switch 2100. The second waveguide 2108 (solid lines), which is on a second level, couples as an input to the second elementary switch 2104. A third waveguide (dot-dashed lines) 2114, which is on a third level, forms a second input 2116 to the second elementary switch 2104 and a second output 2118 from the dilated switch 2100.

In the illustrated embodiment, light is vertically coupled between waveguides in an elementary switch when a liquid droplet 2120 is present over the switch 2100, as shown in FIG. 21 A. Thus, light entering the dilated switch 2100 along the first input 2110 is vertically coupled to the portion of the second waveguide 2108a in the first elementary switch 2102. This light propagates into the second elementary switch 2104 along the second waveguide 2108 and is coupled vertically from the portion of the second waveguide 2108b in the second elementary switch 2104 into the third waveguide 2114 in the second elementary switch 2104, as is generally shown by the arrows on the various waveguides.

A cross-section through an embodiment of an optical chip 2130 that contains the dilated switch 2100 is schematically illustrated in FIG. 21B. A substrate 2132 is provided with a low index growth layer 2134, above which is a first waveguide layer 2136 that contains the first waveguide 2106. A first interlayer 2138 separates the first waveguide layer 2136 from a second waveguide layer 2140 that contains the second waveguide portions 2108a and 2108b. A second interlayer 2142 separates the second waveguide layer 2140 from a third waveguide layer 2144 that contains the third waveguide 2114. A cover layer 2146 is provided over the third waveguide layer 2144. A recess 2148 in the cover layer 2146 provides liquid access to the third waveguide 2114 for activating the second elementary switch 2104. The recess 2148 is shown to extend to the right side of the figure, but it should be understood that the recess is finite in dimension and the cover layer would resume at some distance to the right of the figure. A further recess 2150 in the second interlayer 2142, within the recess of the cover layer 2146, provides liquid access to the second waveguide portion 2108a for activating the first elementary switch 2102. Thus, this structure permits a single liquid droplet 2120 to simultaneously activate the two vertically-coupled elementary switches 2102, 2104. The optical chip 2130 may include etch-stop regions 2152 to limit the extent of an etching process used to form the recesses 2148, 2150. Some embodiments of dilated optical switches incorporate three waveguides, for example where two of the waveguides can be used as independent inputs and the dilated optical switch is activatable to couple light from one of the input waveguides to another of the waveguides. A schematic plan view of such a dilated switch 2900 is shown in FIG. 29A. The dilated switch 2900 has three waveguides passing therethrough, 2902, 2904,

2906. When a liquid droplet 2908 is present over the switch 2900, light can be coupled between neighboring waveguides. For example, in a first switch state an optical signal entering the switch 2900 on the first waveguide 2902 may be coupled to the third waveguide 2906, via the second waveguide 2904, as indicated by the arrows on the waveguides 2902, 2904 and 2906. In a second switch state, for example where there is no liquid droplet over the switch 2900, an optical signal entering the switch 2900 on the first waveguide 2902 remains in the first waveguide 2902 and is not coupled to the other waveguides 2204 and 2206.

The cross-section through a first exemplary embodiment of a three- waveguide dilated switch 2920 is schematically illustrated in FIG. 29B. The switch 2920 includes a single waveguide layer 2922 of relatively high refractive index that is contained between lower and upper layers 2924, 2926 of relatively lower refractive index. The upper layer 2926 of relatively low refractive index material is provided with two recesses 2928 exposing the first and third waveguides 2902, 2906, that form liquid-waveguide coupling regions where the liquid droplet may affect the propagation constant of an optical signal. The second waveguide 2904 is buried beneath a ridge 2930 of the upper layer 2926 and, therefore, its propagation constant is affected by the presence of the liquid droplet 2908 to a lesser degree than the first and third waveguides 2902, 2906. Some of the structures shown herein, such as those shown in FIGs. 29B-29D do not include substrates, but it should be appreciated that such structures are normally formed on substrates, typically silicon, for example silicon on insulator (SOI) or silica.

A second exemplary embodiment of a cross-section for a three- waveguide dilated switch 2940 is schematically illustrated in FIG. 29C. The switch 2940 includes a single waveguide layer 2942 of relatively high refractive index that is contained between lower and upper layers 2944, 2946 of relatively lower refractive index. The upper layer 2946 of relatively low refractive index material is provided with a recess 2948 exposing the second waveguide 2904 that forms a liquid-waveguide coupling region where the liquid droplet may affect the propagation constant of an optical signal in the second waveguide 2904.

The first and third waveguides 2902, 2906 are buried beneath the upper layer 2946 and, therefore, their propagation constants are affected by the presence of the liquid droplet to a lesser degree than that of the second waveguide 2904.

A third exemplary embodiment of a cross-section for a three-waveguide dilated switch 2960 is schematically illustrated in FIG. 29D, which includes both planar and vertical coupling. The switch 2960 includes a lower layer 2962 of relatively low refractive index material below a first waveguide layer 2964, which contains the first waveguide 2902. An interlayer 2966 lies between the first waveguide layer 2964 and a second waveguide layer 2968. The second waveguide layer 2968 includes the second waveguide 2904 and the third waveguide 2906. A cover layer 2970, having a relatively low refractive index, lies over the second waveguide layer. A recess 2972 in the cover layer 2970 exposes the second waveguide 2904 to form a liquid-waveguide coupling region where the liquid droplet may affect the propagation constant of an optical signal in the second waveguide 2904. Thus, in this embodiment, light may couple vertically between the first and second waveguides 2902, 2904 and horizontally between the second and third waveguides 2904, 2906.

In each of the embodiments of dilated switch 2920, 2940, 2960 illustrated in FIGs. 29B-29D, light propagating along the first waveguide 2902 can be switched to the third waveguide 2906 by way of the second waveguide 2904. Likewise, light propagating along the third waveguide 2906 can be switched to the first waveguide 2902 by way of the second waveguide 2904.

FIG. 30A schematically presents an embodiment of a dilated switch 3000 in which light signals can be coupled from one waveguide to another waveguide via two

intermediate waveguides. The switch 3000 includes a first waveguide 3002, a second waveguide 3004, a third waveguide 3006 and a fourth waveguide 3008. In the illustrated embodiment, light is coupled between waveguides when the liquid droplet 3010 is present. Thus, a light signal entering the switch 3000 on the first waveguide 3002 is coupled to the fourth waveguide 3008, as is indicated by the dashed arrow 3012. Likewise, a light signal entering the switch 3000 on the fourth waveguide 3008 is coupled to the first waveguide 3002, as is indicated by the dashed arrow 3014.

A first exemplary cross-section of an optical chip 3020 that may be used to implement the dilated switch 3000 is schematically illustrated in FIG. 30B. The chip 3020 includes the first, second, third and fourth waveguides 3002, 3004, 3006, 3008 as part of a single waveguide layer 3024. The waveguide layer 3024 is formed of relatively high refractive index material between a lower growth layer 3022 and an upper cover layer 3026, both formed of a relatively low index material. Recesses 3028 in the cover layer 3026 expose portions of the first and fourth waveguides 3002, 3008 to be at or near the surface of the cover layer 3026 so as to form a liquid-waveguide coupling region where the liquid droplet can affect the propagation constants of optical signals in the first and fourth waveguides 3002, 3008.

Another exemplary cross-section of an optical chip 3040 that may be used to implement the dilated switch 3000 is schematically illustrated in FIG. 30C. The chip 3040 includes the first, second, third and fourth waveguides 3002, 3004, 3006, 3008 as part of a waveguide layer 3044. The waveguide layer 3044 is formed of relatively high refractive index material between a lower growth layer 3042 and an upper cover layer 3046, both formed of a relatively low index material. A recess 3048 in the cover layer 3046 exposes portions of the second and third waveguides 3004, 3006 to be at or near the surface of the cover layer 3046 so as to form a liquid- waveguide coupling region where the liquid droplet can affect the propagation constants of optical signals in the second and third waveguides 3004, 3006.

Another exemplary cross-section of an optical chip 3060 that may be used to implement the dilated switch 3000 is schematically illustrated in FIG. 30D. This approach includes both horizontal coupling and vertical coupling between waveguides. The chip 3060 includes the first waveguide 3002 in a first waveguide layer 3064. The waveguide layer 3064 is formed of relatively high refractive index material between a lower growth layer 3062 and an interlayer 3066, both formed of a relatively low index material. A second waveguide layer 3068 is positioned above the interlayer 3066 and below an upper cover layer 3070, which is also formed of a material having a refractive index less than that of the second waveguide layer 3068. The second, third and fourth waveguides 3004, 4006, 3008 are formed in the second waveguide layer 3068, with at least a portion of the second waveguide 3004 being positioned above the first waveguide 3002. A recess 3072 in the cover layer 3070 exposes portions of the second and third waveguides 3004, 3006 to be at, or near, the surface of the cover layer 3070 so as to form a liquid waveguide coupling region where the liquid droplet can affect the propagation constants of optical signals in the second and third waveguides 3004, 3006.

In another embodiment of a dilated switch, light signals can be coupled from one waveguide to another waveguide via more than two intermediate waveguides. Thus, a switch, similar to the illustrated embodiment shown in FIG. 30 A, can have more than two waveguides between the outside waveguides 3002 and 3008. The collection of waveguides situated between the two outer waveguides 3002 and 3008 may be referred to as a waveguide array and can be obtained by inserting additional waveguides between the inner waveguides 3004 and 3006. In a manner like that described above for the embodiment illustrated in FIG. 30 A, light can be coupled between neighboring

waveguides when a liquid droplet is present. A light signal entering the switch 3000 on waveguide 3002 can be coupled to waveguide 3008, as is indicated by the dashed arrow 3012, by traversing the waveguide array. Likewise, a light signal entering the switch 3000 on waveguide 3008 can be coupled to waveguide 3002, as is indicated by the dashed arrow 3014, by traversing the same waveguide array. Exemplary cross-sections of an optical chip are similar to those shown in FIGs. 30B-30D, but where the waveguides 3004 and 3006 are replaced by a waveguide array, comprising more than two waveguides. The waveguide array may be shielded from the liquid droplet, in a manner similar to that shown in FIG. 30B, or at least a portion of the waveguide array may be exposed to the liquid droplet, in a manner like that shown in FIGs. 30C and 30D. The waveguide array for a similar configuration as shown in FIG. 30D is formed in the second waveguide layer 3068, with one outer waveguide of the waveguide array adjacent to waveguide 3008 and at least a portion of the outer waveguide at the other side of the waveguide array positioned above waveguide 3002.

Examples

A. Comparison of Intersecting and Vertically-Separated Crossovers

In-plane crossovers were formed in SiN waveguides with a width of 1.50 pm intersecting at 90°. Transmission losses were measured for light propagating through the in-plane crossovers at various wavelengths, 1260 nm, 1360 nm, 1500 nm and 1630 nm.

The transmission losses were found to lie in the range of about 0.22 dB to 0.27 dB over the wavelength range, as shown in FIG. 22.

FIG. 23 shows the transmission loss of vertically-separated crossovers, without the presence of a coupling fluid, over the same wavelength range as the in-plane

measurements shown in FIG. 22. The vertically-separated crossovers included two waveguides crossing at 90°, but with one waveguide displaced vertically relative to the other waveguide. The layer separating the two waveguides was 1 pm. The waveguides each had a width of 1.5 pm and height of 350 nm. The transmission losses per crossover were in the range of about 0.037 dB to 0.060 dB. The vertically-separated cross-overs were fabricated using a process like that discussed below for vertically coupled switches. Thus, the use of an optical chip having at least two waveguide layers forming vertically separated crossovers can significantly reduce the chip transmission losses compared to a chip having crossovers formed in a single waveguide layer.

B. Vertically-Coupled Switches

Vertically-coupled, microfluidically controlled switches were fabricated and characterized. The fabrication steps are described with reference to FIGs. 28A-28G.

FIG. 28A shows a thermally oxidized Si wafer 2802 with a 4.5 pm thick S1O2 growth layer 2804, on which a 350 nm thick SiN first waveguide layer 2806 was deposited using PECVD. The waveguide 2808 in the first waveguide layer 2806 was defined by a 175 nm partial etch. Patterning of the waveguide 2808 was performed using standard lithographic techniques. Etching was performed using RIE, to produce the structure illustrated in FIG. 28B.

Next, a 1.0 pm S1O2 first oxide layer 2810 was deposited over the first waveguide layer 2806 using PECVD and planarized by chemico-mechanical polishing (CMP), to produce the structure shown in FIG. 28D. The thickness of oxide above the first waveguide layer 2806 was controlled by measuring the remaining oxide thickness of the layer 2810 after planarization and then depositing a S1O2 second oxide layer 2812 to achieve a total oxide layer 2814 having a desired thickness, in this case 1.0 pm, to produce the structure shown in FIG. 28E. The layer of oxide 2814, comprising the first oxide layer 2810 and the second oxide layer 2812, is referred to as the oxide interlayer 2814.

A second 350 nm thick SiN waveguide layer 2816 was deposited by PECVD over the oxide interlayer 2814, to produce the structure illustrated in FIG. 28F. The second waveguide layer 2816 was patterned using techniques similar to those used to form the first waveguide 2808 in the first waveguide layer 2806, thus forming a second waveguide 2818 in the second waveguide layer 2816. The width of the waveguides 2808, 2818 was between 0.9 pm and 2.0 pm over the region where the waveguides 2808, 2818 were in vertical relationship. The height of the waveguides 2808, 2818 was that of the unetched original waveguide layer 2806, 2816, i.e. 350 nm. The thickness of the oxide interlayer 2814 between the waveguides 2806, 2816 was 1.0 pm. This procedure resulted in a structure like that shown in FIG. 28G. The length of the region over which the waveguides 2806, 2816 were in a vertical, overlapping relationship was 900 pm.

The vertical distance between the waveguides can affect the performance of a switch network. An increased separation can result in reduced insertion loss and crosstalk of the vertically separated crossovers, but an increased insertion loss of the elementary switches, and so selection of the interlayer thickness is important. The simulated transmission of a 600 um long device in the cross state is shown in FIG. 24 for different interlayer thicknesses. The insertion loss and crosstalk in the bar state are calculated to be better than 0.1 dB and -48 dB, respectively. Lines 2402 and 2404 respectively show the cross-state insertion loss and crosstalk for an interlayer thickness of 0.90 pm. Lines 2406 and 2408 respectively show the cross-state insertion loss and crosstalk for an interlayer thickness of 1.0 pm. Lines 2410 and 2412 respectively show the cross-state insertion loss and crosstalk for an interlayer thickness of 1.1 pm.

The vertically-coupled switches, having a 900 pm long coupling length, whose fabrication is described in FIGs. 28A-28G, were characterized in the cross and bar state. The interlayer oxide thickness was 1 pm and the liquids used to activate the switch had a refractive index of 1.37 (bar state) and 1.55 (cross-state) at 1550 nm. The results are presented using a notation described with reference to FIG. 25, which shows a generic switch 2500 having first input L0 and a first output R0 on a first waveguide 2502, and a second input Ll and second output Rl on a second waveguide 2504. An optical signal may couple intentionally between the waveguides 2502, 2504, i.e. L0-R1 and L1-R0, when the switch 2500 is in the cross state and unintentionally when it is in the bar state. The latter is referred to as bar state crosstalk. Also, an optical signal may pass along the same waveguide 2502, 2504, i.e. L0-R0, Ll-Rl, intentionally when the switch 2500 is in the bar state and unintentionally when it is in the cross state. The latter is referred to as cross state crosstalk.

FIG. 26 shows the wavelength dependent performance of the switch in the bar state. Line 2600 shows the transmission between the bar ports L0-R0 and line 2802 shows the transmission between the bar ports Ll-Rl . In this example, the lower, buried waveguide was waveguide 2504 and the upper waveguide, exposed to the liquid droplet, as waveguide 2502. Line 2604 shows the crosstalk between L0-R1 and line 2606 shows the crosstalk between L1-R0.

FIG. 27 shows the wavelength dependent performance of the switch in the cross state. Line 2706 shows the transmission between the cross ports L1-R0 and line 2704 shows the transmission between the cross ports L0-R1. Line 2700 shows the crosstalk between L0-R0 and line 2702 shows the crosstalk between Ll-Rl.

The switch insertion loss is less than 4 dB for the bar and cross states. The crosstalk is less than -20 dB for the bar state. The crosstalk in the cross state is relatively high compared to the bar state, particularly for shorter wavelengths. At wavelengths of the more than about 1350 nm, the cross state crosstalk is better than about -12 dB.

Various modifications, equivalent processes, as well as numerous structures to which the present invention may be applicable will be readily apparent to those of skill in the art to which the present invention is directed upon review of the present specification. The claims are intended to cover such modifications and devices. For example, optical circuits that incorporated the present invention may use different numbers of switch units from those illustrated here, for example optical circuits may be configured as 8 x 8, 16 x 16 or even with different numbers of rows and columns. Additionally, it is not required that a switch unit include only a configuration of elementary switches as shown here, but may include other configurations of elementary switches so as to achieve various optical functions, as may be desired for a particular application. Furthermore, the term“droplet” as used herein refers to a volume of liquid as may be present, for example, in a fluid microchannel, a capillary, and which is used to switch one or more elementary switches.

As noted above, the present invention is applicable to optical communication and data transmission systems, including active optical switch systems. Accordingly, the present invention should not be considered limited to the particular examples described above, but rather should be understood to cover all aspects of the invention as fairly set out in the attached claims.

Claims

Claims What we claim as the invention is:

1. An optical switch unit, comprising:

a first elementary optical switch configured to be activatable by a first liquid droplet;

a second elementary optical switch configured to be activatable by the first liquid droplet;

a first input optically coupled to at least one of the first and second elementary optical switches; and

a first output optically coupled to at least one of the first and second elementary optical switches.

2. An optical switch unit as recited in claim 1, wherein a first output of the first elementary optical switch is optically coupled to a first input of the second elementary optical switch.

3. An optical switch unit as recited in claim 2, wherein a second output of the first elementary optical switch is optically coupled to a first input of a third elementary optical switch.

4. An optical switch unit as recited in claim 2, further comprising a third elementary optical switch having a first output optically coupled to a first input of a fourth elementary optical switch, and wherein a second output of the third elementary optical switch is optically coupled to a second input of the second elementary optical switch and a second output of the first elementary optical switch is optically coupled to a second input of the fourth elementary switch.

5. An optical switch unit as recited in claim 2, wherein a first input to the optical switch unit is optically coupled to a first input of the first elementary optical switch, a second input of the optical switch unit is optically coupled to a second input of the second elementary optical switch, a first output of the second elementary optical switch is coupled to a first output of the optical switch unit and a second output of the first elementary optical switch is optically coupled to a second output of the optical switch unit.

6. An optical switch unit as recited in claim 1, wherein the first elementary optical switch lies on a first optical path between a first input and a first output of the optical switch unit and the second elementary optical switch lies on a second optical path between a second input and a second output of the optical switch unit.

7. An optical switch unit as recited in claim 6, further comprising a third elementary optical switch on the first optical path between the first elementary optical switch and the first output of the optical switch unit.

8. An optical switch unit as recited in claim 1, in an equilibrium state, when the first liquid droplet is in a position to activate the first elementary optical switch, the first liquid droplet is also in a position to activate the second elementary optical switch.

9. An optical switch unit as recited in claim 1, in an equilibrium state, when the first liquid droplet is in a position to activate the first elementary optical switch, the first liquid droplet is also in a position that does not activate the second elementary optical switch, and when the first liquid droplet is in a position to activate the second elementary optical switch, the first liquid droplet is also in a position that does not activate the first elementary optical switch.

10. An optical switch unit as recited in claim 1, wherein the first and second elementary switches are configured in series so as to permit an optical signal to pass from the first elementary optical switch to the second elementary optical switch.

11. An optical switch unit as recited in claim 1, wherein the first and second elementary switches are configured in parallel so that no optical signal can pass between the first and second elementary optical switches.

12. An optical circuit, comprising:

a plurality of inputs;

a plurality of outputs; and a plurality of optical switch devices disposed between the plurality of inputs and the plurality of outputs, the optical switch devices being configurable to selectively connect at least one input of the plurality of inputs to at least one output of the plurality of outputs;

wherein at least one of the optical switch devices comprises an optical switch unit having a first elementary optical switch configured to be activatable by a first liquid droplet, a second elementary optical switch configured to be activatable by the first liquid droplet; at least one of the first and second

elementary optical switches lying on a selectable path between a first input of the optical switch unit and a first output of the optical switch unit.

13. An optical circuit as recited in claim 12, wherein the optical switch devices are configured in a cross/bar configuration between the plurality of inputs and the plurality of outputs.

14. An optical circuit as recited in claim 12, wherein the optical switch devices are configured in a path-independent loss network between the plurality of inputs and the plurality of outputs.

15. An optical circuit as recited in claim 12, further comprising an input polarization splitting and rotating (PSR) unit at an input of the optical circuit and an output PSR at an output of the optical circuit, first and second polarized light signals from the input PSR unit passing to the switch unit, the first and second elementary optical switches in the switch unit being arranged in parallel to respectively receive the first and second polarized light signals.

16. An optical circuit as recited in claim 12, further comprising respective input polarization splitting and rotating (PSR) units at the inputs to the optical circuit and respective output PSR units at outputs from the optical circuit, each optical switch device comprising a switch unit having respective first elementary optical switches configured to be activatable by respective first liquid droplets, and respective second elementary optical switches configured to be activatable by the respective first liquid droplets, the first and second elementary optical switches in each switch unit being arranged in parallel to respectively receive polarized light signals.

17. A method of operating an optical switch unit, comprising: changing a state of a first elementary switch of the optical switch unit by moving a liquid droplet;

simultaneously changing a state of a second elementary switch of the optical switch unit by moving the liquid droplet;

wherein changing the state of the first elementary switch and changing the state of the second elementary switch results in changing a switch state of the optical switch unit.

18. A method as recited in claim 17, wherein moving the liquid droplet comprises moving the liquid droplet to a position over a first waveguide of the first elementary switch and moving the liquid droplet to a position over a first waveguide of the second elementary switch.

19. A method as recited in claim 17, wherein moving the liquid droplet comprises moving the liquid droplet to a position over a first waveguide of the first elementary switch and moving the liquid droplet from a position over a first waveguide of the second elementary switch.

20. A method as recited in claim 18, further comprising simultaneously changing a state of a third elementary switch by moving the liquid droplet to a position over a first waveguide of the third elementary switch.

21. A method as recited in claim 19, further comprising simultaneously changing a state of a third elementary switch by moving the liquid droplet from a position over a first waveguide of the third elementary switch.

22. An optical switch unit, comprising:

a first waveguide disposed at a first distance from a substrate; a second waveguide over the first waveguide, the second waveguide being disposed at a second distance from the substrate, the second distance being greater than the first distance;

at least a third waveguide disposed at the second distance from the substrate; a first low-index layer disposed at a third distance from the substrate greater than the second distance, the first low index layer having a refractive index less than the refractive index of the second waveguide, a recess formed in the first low index layer over at least the second waveguide to form a liquid-waveguide coupling region over the second waveguide; and

a liquid droplet movable into the recess of the first low index layer to affect coupling of optical signals between the first and second waveguides.

23. An optical switch unit as recited in claim 22, wherein the first waveguide is formed in a first waveguide layer.

24. An optical switch unit as recited in claim 22, wherein the second waveguide is formed in a second waveguide layer and the third waveguide is formed in the second waveguide layer.

25. An optical switch unit as recited in claim 22, further comprising a fourth waveguide disposed at the second distance from the substrate, wherein the recess in the first low index layer is formed over the third waveguide but not over the fourth

waveguide, the liquid droplet being movable into the recess of the low index layer to affect coupling of optical signals between the third and fourth waveguides.

26. An optical switch as recited in claim 25, further comprising a fifth waveguide disposed at the first distance from the substrate, a sixth waveguide disposed over the fifth waveguide at the second distance from the substrate, and seventh and eighth waveguides disposed at the second distance from the substrate, the recess of the first low- index layer being formed over the sixth and seventh waveguides, the liquid droplet being movable into the recess of the first low index layer to affect coupling of optical signals between the fifth and sixth waveguide and coupling of optical signals between the seventh and eighth waveguides.

27. An optical switch unit as recited in claim 22, further comprising a fourth waveguide disposed at a fourth distance from the substrate, the fourth distance being greater than the third distance, and a second low index layer disposed at a fifth distance from the substrate, the fifth distance being greater than the fourth distance, the second low index layer being formed with a recess over the recess of the first low index layer and over the fourth waveguide to form a liquid-waveguide coupling region over the fourth waveguide, the liquid droplet being moveable into the recess of the second low index layer to affect coupling of optical signals between the third and fourth waveguides.

28. An optical switch unit as recited in claim 22, wherein the first and second waveguides comprise at least one of silicon and silicon nitride.

29. An optical switch unit as recited in claim 22, wherein the first and second waveguides comprise silica.

30. An optical switch unit as recited in claim 22, wherein the first low index layer comprises silica and has a thickness of no more than 1.1 pm.

31. An optical switch as recited in claim 30, wherein the first low index layer has a thickness of around 1.0 pm

32. An optical switch circuit, comprising:

a plurality of optical switch units arranged on a substrate;

a plurality of waveguides connecting between selected pairs of optical switch units;

wherein at least one of the optical switch units comprises a first

microfluidically controlled elementary switch having a first waveguide and a second waveguide disposed horizontally relative to the first waveguide and a second microfluidically controlled elementary switch having a third waveguide and a fourth waveguide disposed vertically relative to the third waveguide.

33. An optical switch circuit as recited in claim 32, wherein, in the second microfluidically controlled elementary switch of the at least one of the optical switch units the third waveguide and fourth waveguide form a waveguide crossover.

34. An optical switch circuit as recited in claim 32, wherein the optical switch units are arranged in a Pi-loss network.

35. An optical switch circuit as recited in claim 32, wherein the optical switch units are arranged in a crossbar network.

36. An optical switch circuit as recited in claim 32, further comprising circuit inputs coupling to certain switch units of the plurality of switch units, wherein a first of the at least one optical switch units is arranged to switch an optical signal that entered one of the circuit inputs in a first polarization state and a second of the at least one optical switch units is arranged to switch an optical signal that entered the one of the circuit inputs in a second polarization state orthogonal to the first polarization state.

37. An optical switch unit, comprising:

a first waveguide;

a second waveguide proximate the first waveguide;

a third waveguide proximate the second waveguide, the second waveguide being positioned so that light coupling from the first to the third waveguide couples first from the first waveguide to the second waveguide and then from the second waveguide to the third waveguide;

a low index layer provided above the first, second and third waveguides, the low index layer having a recess over at least one of the first, second and third waveguides;

a liquid droplet movable into the recess of the low index layer to affect coupling of optical signals between i) the first and second waveguide and ii) the second and third waveguide.

38. An optical switch unit as recited in claim 37, wherein the recess extends above the first and third waveguides but not the second waveguide.

39. An optical switch unit as recited in claim 37, wherein the recess extends above the second waveguide, but not the first or third waveguide.

40. An optical switch unit as recited in claim 37, wherein the first waveguide is positioned between the second waveguide and a substrate, light coupling between the first and second waveguides coupling in a direction generally perpendicular to the substrate, light coupling between the second and third waveguides coupling in a direction generally parallel to the substrate.

41. An optical switch unit as recited in claim 37, wherein the first, second and third waveguides are substantially coplanar.

42. An optical switch unit as recited in claim 37, wherein the liquid droplet is movable into the recess of the low index layer to affect coupling of optical signals between the first and third waveguides via the second waveguide.

43. An optical switch unit as recited in claim 42, comprising at least a fourth waveguide between the second and third waveguides.

44. An optical switch unit, comprising:

a first waveguide;

a second waveguide proximate the first waveguide;

a fourth waveguide proximate the third waveguide, the third waveguide being positioned between the second and third waveguides;

a low index layer provided above the first, second, third and fourth waveguides, the low index layer having a recess over at least two of the first, second, third and fourth waveguides;

a liquid droplet movable into the recess of the low index layer to affect coupling of optical signals between at least one of i) the first and second waveguides and ii) the third and fourth waveguides.

45. An optical switch unit as recited in claim 44, wherein the first, second, third and fourth waveguides are coplanar.

46. An optical switch unit as recited in claim 45, wherein the first waveguide is disposed on one side of the second waveguide, the third and fourth waveguides are disposed on the other side of the second waveguide, the fourth waveguide is disposed on one side of the third waveguide and the first and second waveguides being disposed on the other side of the third waveguide.

47. An optical switch unit as recited in claim 46, wherein the recess extends over the first and fourth waveguides but not over the second and third waveguides.

48. An optical switch unit as recited in claim 46, wherein the recess extends over the second and third waveguides but not over the first and fourth waveguides.

49. An optical switch unit as recited in claim 44, wherein the second, third and fourth waveguides are coplanar and the first waveguide is positioned between the second waveguide and a substrate.

50. An optical switch unit as recited in claim 44, wherein the liquid droplet is movable into the recess to affect coupling of light between the first and fourth waveguides.

51. A method of making an optical chip, comprising:

providing a first layer of relatively high index material over a first layer of relatively low index material;

forming a first waveguide in the first layer of relatively high index material; providing a second layer of relatively low index material over the first layer of relatively high index material having the first waveguide;

planarizing the second layer of relatively low index material; providing a third layer of relatively low index material on the second layer of relatively low index material, the third layer of relatively low index material having a thickness so that the second layer of relatively low index material and third layer of relatively low index material together have a desired separation thickness;

providing a second layer of relatively high index material over the third layer of relatively low index material; and

forming a second waveguide in the second layer of relatively high index material above the first waveguide; wherein a separation distance between the first waveguide and the second waveguide is the desired separation thickness.

52. A method as recited in claim 51, wherein forming the first waveguide comprises etching a portion of the first layer of relatively high index material, the first waveguide comprising at least a portion of the first layer of relatively high index material not etched in the step of etching.

53. A method as recited in claim 51, wherein planarizing the second layer of relatively low index material, comprises polishing the second layer of relatively low index material using a chemico-mechanical polish.

54. A method as recited in claim 51, wherein the first layer of relatively high index material and the second layer of relatively high index material are both formed of substantially the same material.

55. A method as recited in claim 51, wherein the first layer of relatively low index material and the second layer of relatively low index material are both formed of substantially the same material.

56. An optical device, comprising:

a substrate;

a first low index layer on the substrate;

a first waveguide, the first low index layer being positioned between the first waveguide and the substrate;

a second low index layer, the first waveguide being positioned between the second and first low index layers; and

a second waveguide substantially above the first waveguide, at least a portion of the second waveguide above the first waveguide being parallel to the first waveguide, the second low index layer being positioned between the second waveguide and the first waveguide.

57. An optical device as recited in claim 56, further comprising a third low index layer, the second waveguide being positioned between the third and second low index layers, the third low index layer comprising a recess over the second waveguide and comprising a liquid droplet movable into the recess so as to affect coupling of light signals between the first and second waveguides.

58. An optical device as recited in claim 57, wherein the first waveguide is formed in a first waveguide layer between the first and second low index layers.

59. An optical device as recited in claim 56, wherein the second waveguide is formed in a second waveguide layer between the second and third low index layers.

60. An optical device as recited in claim 56, wherein the second low index layer comprises a first low index sub-layer and a second low index sub-layer, the first and second low index sub-layers together having a desired thickness of low index material separating the first and second waveguides.

61. An optical waveguide device, comprising:

a substrate;

a first optical waveguide in a first waveguide layer above the substrate; and a second optical waveguide in a second waveguide layer between the first waveguide layer and the substrate, at least part of the second waveguide being disposed below the first waveguide and parallel to the first waveguide so as to permit optical coupling between the first and second optical waveguides.