Structure for a photonic integrated circuit

Abstract

A structure for a photonic integrated circuit is provided. The structure comprises a first waveguide configured to guide light along a first light propagation axis; and a second waveguide on the first waveguide. The second waveguide is configured to couple light out of the first waveguide. The second waveguide comprises: a first portion configured to guide light along a second light propagation axis at an oblique angle to the first light propagation axis, and a second portion configured to receive light from the first portion and to guide light along a third light propagation axis, the third light propagation axis parallel to, or at a less oblique angle than the second light propagation axis to, the first light propagation axis. A method of forming a part of a photonic integrated circuit is also provided.

Description

[0001] STRUCTURE FOR A PHOTONIC INTEGRATED CIRCUIT

[0002]

[0003] Photonic integrated circuits comprising waveguides for controlling the propagation of light are known. Efficient optical coupling between waveguides can reduce losses in optical signals through a photonic integrated circuit, and can help facilitate heterogeneous integration of photonic integrated circuit components formed of different materials. It is desirable to provide improved optical coupling between waveguides of photonic integrated circuits.

[0004]

[0005] of the

[0006] Figure 1 shows schematically a cross-sectional view of a structure for a PIC according to examples;

[0007] Figures 2, 3, and 4 show schematically a plan view of the structure of Figure 1; Figures 5 and 6 illustrate further structures for PICs, according to examples; Figure 7 illustrates a method of manufacture of the structure of Figure 1;

[0008] Figures 8a-8c depict simulation data of the optical performance of structures according to examples; and

[0009] Figures 9 and 10 illustrates further examples of second waveguides.

[0010] Detailed

[0011]

[0012] In photonic integrated circuits (PICs), light is guided by, and propagates through, waveguides. It can be useful to couple light from a first waveguide to a second waveguide, for example to distribute light to different components or areas of the PIC. However, it can be important to reduce or prevent potentially detrimental optical behaviours such as back-reflection which can introduce unwanted noise into the circuit that can reduce component performance, for example.

[0013] Furthermore, there is a desire for heterogeneous integration, wherein components of a PIC are formed of different materials. Each component can be manufactured separately, and then brought together and integrated into a higher-level assembly. Components of the PIC can therefore be formed of materials which exhibit desired optical, electronic, structural, and / or chemical properties for the respectivecomponent, for example. However, forming components in separate manufacturing steps can present challenges when it comes to integrating them together, as heterogeneous integration methods such as transfer printing of a first PIC region onto a second PIC region can be less accurate than those used in homogeneous integration approaches. Therefore, it is desirable for PIC components in heterogeneous integration to have a performance which is robust, or less sensitive, to potential manufacturing inaccuracies.

[0014] In particular, existing approaches to optically coupling heterogeneous components involve fabricating structures with small structure sizes, such as waveguides which are tapered to below 500 nanometres in width. However, manufacturing such structures can introduce unwanted features such as sloped side walls or other undesired geometries, which can be detrimental to optical performance. It is desirable to develop structures which are not reliant upon the manufacture of features with very small sizes.

[0015] Examples described herein relate to a structure in which optical coupling between a first waveguide and a second waveguide is facilitated. This can allow for distribution of light through a PIC. The structure can facilitate lossless, or low loss, coupling of light from the first waveguide to the second waveguide, and can reduce or remove back-scattering of light back along the first waveguide. In some examples, the structure is a heterogeneous integration of the first waveguide and the second waveguide, with the first waveguide being formed of a different material than the second waveguide. The structure can exhibit the aforementioned optical performance, including lossless or low loss coupling between the first and second waveguide, whilst being relatively insensitive to manufacturing inaccuracies which can manifest as misalignment between the first and second waveguide, for example. This can allow for more reliable manufacture of the PIC, which can reduce wastage of materials during manufacture, and / or reduce time and / or energy taken to manufacture the PIC, for example.

[0016] To address these problems, in examples described herein, there is provided a structure for a photonic integrated circuit. The structure comprises a first waveguide configured to guide light along a first light propagation axis, and a second waveguide on the first waveguide. The second waveguide is configured to couple light out of thefirst waveguide. The second waveguide comprises a first portion configured to guide light along a second light propagation axis at an oblique angle to the first light propagation axis, and a second portion configured to receive light from the first portion and to guide light along a third light propagation axis, the third light propagation axis parallel to, or at a less oblique angle than the second light propagation axis to, the first light propagation axis.

[0017] The structure, in being for a photonic integrated circuit, can be formed of materials which are suitable for use in integrated circuit manufacture, such as semiconductor materials or dielectric materials, for example. It will be appreciated that the structure according to the present disclosure is not limited to any particular material composition.

[0018] An oblique angle is, for example, an angle between two axes which are not parallel or perpendicular to each other. For example, an angle of between 0.1-1 degree can be considered an oblique angle, or between 1-10 degrees, or between 10-45 degrees, or between 45-89.5 degrees. The third light propagation axis being parallel to, or at a less oblique angle than the second light propagation axis to, the first light propagation axis, for example means that the second light propagation axis has some angle X degrees with respect to the first light propagation axis, where 0 < |X| < 90 degrees, and the third light propagation axis can be parallel to the first light propagation axis, hence having an angle of zero degrees with respect to the first light propagation axis, or can have an angle of Y degrees with respect to the first light propagation axis, such that 0 < |V| < |X| degrees.

[0019] Light propagating through the first waveguide is coupled into the second waveguide. The first portion of the second waveguide having a second light propagation axis being at an oblique angle to the first light propagation axis means that the coupling strength can be gradually increased along the first and second light propagation axes. A gradual increase in coupling strength can reduce optical loss in the structure, such as by reducing or removing the back-scattering of light back down the first waveguide. The second portion of the second waveguide having a third light propagation axis being parallel to, or at a less oblique angle than the second light propagation axis to, the first light propagation axis, can mean that the second waveguide guides the light along a direction which can be substantially aligned with the directionof light propagation in the first waveguide. This can be useful for facilitating parallel geometries of an overall layout of the PIC, for example. Furthermore, having the third and first light propagation axes substantially aligned in this way can facilitate complete, or substantially complete, coupling of light from the first waveguide to the second waveguide. “Substantially aligned” can mean the respective directions are the same or are within manufacturing tolerances, for example, or aligned within 0.05 degrees for example, and / or within relatively minor variations which do not prevent the structure as functioning as described, for example.

[0020] In some examples, the first portion has a width perpendicular to the second light propagation axis and the width of the first portion is substantially uniform along the second light propagation axis. The first portion, in having a substantially uniform width, can be straightforwardly manufactured and can be less subject to structural inaccuracies which might arise when attempting to manufacture more complex geometries. “Substantially uniform” can mean uniform within manufacturing or measurement tolerances, for example, or within relatively minor variations such as variations within 5%, for example, and / or within relatively minor variations which do not prevent the structure as functioning as described, for example.

[0021] In some examples, the first portion has a width perpendicular to the second light propagation axis and the width of the first portion is greater than or equal to 500 nanometres. When attempting to manufacture structures with lower structure sizes, such as less than 500 nanometres, common manufacturing methods can introduce unwanted structural properties such as slanted sidewalls. This can impact the optical performance, for example. It is desirable for a structure to have sufficient size to avoid such unwanted structural properties. Accordingly, by having a width in excess of 500 nanometres, the first portion can be manufactured at a scale which does not exhibit, or exhibits reduced, unwanted structural properties. In some such examples, the width is greater than or equal to 1 micrometre, which can be reliably manufactured without introducing unwanted structural features which might be detrimental to optical performance.

[0022] In some examples, the second light propagation axis is substantially straight. The first portion of the second waveguide can be formed as a substantially straight structure, which means it can be straightforwardly manufactured. “Substantiallystraight” can mean straight within manufacturing or measurement tolerances, for example, for example, and / or within relatively minor variations which do not prevent the structure as functioning as described, for example. In other words, in some examples, the second light propagation axis is non-curved.

[0023] In some examples, the second portion is tapered, having a width, perpendicular to the third light propagation axis, which increases in a direction away from the first portion along the third light propagation axis. This can further facilitate low-loss coupling of light from the first waveguide to the second waveguide, and can modify a mode field profile of light propagating in the second portion to a desired mode field profile for onwards propagation through the PIC. In some such examples, the width of the second portion changes adiabatically along the third light propagation axis. The width of the second portion changing “adiabatically” for example means that a rate of change of width per unit length, and hence a rate of change in mode field profile of light per unit length, is small relative to the scale of the wavelength of light propagating through the waveguide. This can allow a mode field profile of light propagating in the second portion to be changed to a desired mode field profile in a low-loss or lossless manner.

[0024] In some examples, the first waveguide comprises silicon (such as being Silicon-on-Insulator, SOI, based) and the second waveguide comprises a compound semiconductor. That is, the structure may be a heterogeneous integration of the first waveguide and the second waveguide, which can facilitate coupling from a first portion of the PIC being silicon-based to a second portion of the PIC being based on compound semiconductor(s).

[0025] According to another aspect of the present disclosure, in examples described herein, there is provided a method of forming a part of a photonic integrated circuit. The part of the photonic integrated circuit can be the structure described previously. The method comprises forming a first waveguide configured to guide light along a first light propagation axis; and forming a second waveguide on the first waveguide, the second waveguide configured to couple light out of the first waveguide. The second waveguide comprises a first portion configured to guide light along a second light propagation axis at an oblique angle to the first light propagation axis, and a second portion configured to receive light from the first portion and to guide light along a thirdlight propagation axis, the third light propagation axis parallel to, or at a less oblique angle than the second light propagation axis to, the first light propagation axis. In some examples, forming the second waveguide on the first waveguide comprises transfer printing the second waveguide onto the first waveguide. Transfer printing can facilitate heterogeneous integration, for example. In some such examples, the first waveguide comprises silicon and the second waveguide comprises a compound semiconductor. The second waveguide may be transfer printed onto an intermediate layer of Benzocyclobutene. It will be appreciated that discussion of devices according to the present disclosure can be equally applied to methods, where applicable, and vice versa.

[0026] Examples of the present disclosure will now be described in view of the figures.

[0027] Figures 1 and 2 illustrate, schematically, part of a photonic integrated circuit, PIC, 10 which comprises a structure 100 according to examples. Figure 1 is in a cross-sectional view, and Figure 2 is in a plan-view. A set of orthogonal reference axes x, y, and z are provided to help explain the relative arrangements of the constituent features of the structure 100. Reference axis x can be considered a length direction, reference axis y can be considered a width direction, and reference axis z can be considered a height direction. It will be appreciated that these define a local coordinate system in which the PIC 10 and structure 100 can be described, and that references to e.g. “vertical” or “horizontal” positions are relative to this local coordinate system. In other words, for example, a first component described as being vertically above a second component refers herein to these components being displaced relative to the z reference axis, and can apply regardless of the PIC’s orientation to external objects, for example. When the structure is considered in “plan view”, this can mean the structure is described in reference to the plane defined by the y and x reference axes, or the width-length plane, which is a plane parallel to, or approximately parallel to, an upper surface of the substrate. For features which are vertically displaced from one another, description of their relative positions or alignment in “plan view” can mean considering their respective projections onto the aforementioned plane. Similarly, in “cross-sectional view” can mean in reference to the plane formed by the z and x axes, or the heightlength plane, or the z and y axes, or the height-width plane, as appropriate. Again, therelative positions or alignments of features described in reference to the cross-sectional view can mean the respective projections onto these planes.

[0028] The structure 100 comprises a first waveguide 110 and a second waveguide 120. The first waveguide 110 is on a substrate 130, such that the first waveguide 110 overlies the substrate 130. Similarly, the second waveguide 120 is on the first waveguide 110, in that it is displaced vertically from but overlies the first waveguide 110. In this example, an intermediate layer 140 extends between and separates the second waveguide 120 from the first waveguide 110. A layer being “on” another layer, as described herein, such as the first waveguide 110 being on the substrate 130, or the second waveguide 120 being on the first waveguide 110, can also mean that, in some examples, the respective layers are in contact with each other, with one overlying the other. A first layer being “on” a second layer can include the first layer only partially overlapping the second layer. For example, with respect to the v-axis. which can be considered a horizontal or lateral direction, the second waveguide can entirely overlap the first waveguide, or only partially overlap the first waveguide, whilst still being “on” the first waveguide.

[0029] The first waveguide 110 is configured to guide light along a first propagation axis 101. In isolation, that is, absent coupling with any other waveguides, light propagates through the first waveguide 110 in a first mode, and can propagate in a direction from a first end 110a of the first waveguide 110 to a second end 110b of the first waveguide 110.

[0030] The second waveguide 120 comprises a first portion 122 and a second portion 124. The second portion is configured to receive light from the first portion 122. The first portion 122 and second portion 124 can be considered to be in series. The first portion 122 being “in series” with the second portion 124 refers to these portions being arranged along a propagation direction of light through the second waveguide 120, in that light can propagate through the first portion 122 followed by the second portion 124It should be understood that the first portion 122 and the second portion 124 being “in series” does not preclude the existence of intermediate portions between the first portion 122 and the second portion 124. That is, the first portion 122 may be followed immediately by the second portion 124, such as in the example of Figures 1 and 2, but in other examples there may be intermediate portion(s) between the first portion 122and the second portion 124, which could still be considered in series since light propagates from the first portion 122 to the second portion 124, via the intermediate portion(s). The first portion 122 of the second waveguide 120 is configured to guide light along a second light propagation axis 102. In isolation, that is, absent coupling with any other waveguides, the light propagates through the first portion 122 in a second mode. The second portion 124 of the second waveguide 120 is configured to guide light along a third light propagation axis 103. In isolation, that is, absent coupling with any other waveguides, the light propagates through the second portion 122 in a third mode.

[0031] The first waveguide 110 and second waveguide 110 have, respectively, a constituent waveguide core layer and one or more waveguide cladding layers which provide a refractive index contrast with the core layer such that the first 110 and second waveguide 120 can guide light. The waveguide core and one or more waveguide cladding layers are not labelled here for clarity in the figures. It will be appreciated that the precise configuration of respective core and cladding layers of the first 110 and second waveguide 120 can vary between examples.

[0032] Figures 3 and 4 illustrate, in further detail, structural and geometric properties of the first 110 and second waveguide 120, according to examples. In particular, Figure 3 illustrates relative angles between the first waveguide 110 and the portion of the second waveguide 120, with the substrate 130 and the intermediate layer 140 omitted for clarity. Figure 4 illustrates the dimensions of the portions 122, 124 of the second waveguide 120, with the first waveguide 110, the substrate 130, and the intermediate layer 140 omitted for clarity.

[0033] The first portion 122 of the second waveguide 120 is configured such that the second light propagation axis 102 is at an oblique angle, identified as a in Figure 3, to the first light propagation axis 101. The oblique angle is typically a relatively shallow angle, such as an angle less than 10 degrees, less than 5 degrees, and in some examples between 0.2 and 1 degrees. Other, larger angles are envisaged, however. In the examples of Figures 1-4, the first portion 122 of the second waveguide 120 is elongate, extending from a first end 122a to a second end 122b of the first portion 122 of the second waveguide 120 along a length laand is substantially straight. That is, in this example, the second portion 122 is not curved. The first end 122a of the first portion122 is closer to the first end 110a of the first waveguide 110 than the second end 122b of the first portion 122 is to the first end 110a of the first waveguide 110. Considering the arrangement of the first portion 122 of the second waveguide 120 and the first waveguide 110 in plan view, the second end 122b of the first portion 122 is arranged closer to the first light propagation axis 101 than the first end 122a of the first portion 122 is to the first light propagation axis 101, such that the second light propagation axis 102 extends at the oblique angle relative to the first light propagation axis 101. In other words, in a direction from the first end 110a of the first waveguide 110 to the second end 110b of the first waveguide 110, which is also generally in a direction from the first end 122a of the first portion 122 to the second end 122b of the first portion 122, a separation between the first light propagation axis 101 and second light propagation axis 102 decreases, when considered in plan view (such as that depicted in Figure 2).

[0034] As depicted in Figure 2, the first end 122a of the first portion 122 is positioned away from the second portion 124, compared with the second end 122b of the first portion 122, and the first end 122a overhangs the first waveguide 110 such that a section of the first portion 122 is partially on the first waveguide 110 and partially on a surrounding portion of the PIC 10. In other words, some of the first portion 122 is on the first waveguide and some of the first portion 122 overhangs the first waveguide 110. The first portion 122 of the second waveguide 120 has a width waperpendicular to the length la. The width wais substantially uniform along the length la. in so far as it has the same width within manufacturing tolerances.

[0035] The second portion 124 of the second waveguide 120 is configured such that the third light propagation axis 103 is parallel to, or at a less oblique angle than the second light propagation axis 102 to, the first light propagation axis 101. In the example of Figures 1-4, the third light propagation axis 103 is substantially parallel to the first light propagation axis 101, in that they are parallel within manufacturing or measurement tolerances and / or within relatively minor variations which do not prevent the structure as functioning as described, for example. In other words, the third light propagation axis 103 overlies the first light propagation axis 101. When viewed in plan view, such as depicted in Figure 2, the first light propagation axis 101 appears colinear with the third light propagation axis 103. An angle , depicted in Figure 3, is defined between the second light propagation axis 102 and the third light propagation axis 103which is less than 180 degrees. The second portion 124 of the second waveguide 120 extends from a first end 124a to a second end 124b along a length lb. The second portion 124 of the second waveguide 120 has a width wb, wcperpendicular to the length lb. The second portion 124 is tapered, in that it has a first width wbat the first end 124a and a second, larger width wcat the second end 124b, and a gradual transition of width between the two. The width of the second portion 124 may gradually change at a consistent rate per unit length of the second portion 124, or may gradually change at a varying rate per unit length, for example.

[0036] The second end 122b of the first portion 122 of the second waveguide 120 abuts the first end 124a of the second portion 124 of the second waveguide 120. That is, the second end 122b of the first portion 122 of the second waveguide 120 has a first end facet, an exit facet through which light, propagating along the second light propagation axis 102, would exit the first portion 122. The first end 124a of the second portion 124 of the second waveguide 124 has a second end facet, an entry facet through which light would enter the second portion 124 for propagation along the third light propagation axis 103. The first end facet faces the second end facet, and, in this example, the first and second end facets abut (i.e. are in contact with each other) to form an interface 123. In other examples, a separation between the exit facet and the entry facet may exist, but nevertheless light can be coupled from the first portion 122 to the second portion 124.

[0037] In this example, the first width wbat the first end 124a of the second portion 124 matches, within manufacturing or measurement tolerances, the width waof the first portion 122. That is, the first end facet has a first width generally perpendicular to the second light propagation axis 102, and the second end facet has a second width generally perpendicular to the third light propagation axis 103, with the first width being the same as the second width. In other examples, the first end facet and second end facet may have different widths.

[0038] In use, light can be propagated from the first end 110a of the first waveguide 110 towards the second end 110b of the first waveguide 110 along the first light propagation axis 101. Due to the proximity of the second waveguide 120 to the first waveguide 110, light couples from the first waveguide 110 into the second waveguide 120 by coupling evanescently, as indicated by arrow L in Figure 1. Since a horizontaldistance (i.e. with respect to reference axis y) between the first light propagation axis 101 and the second light propagation axis 102 gradually decreases with propagation distance from the first end 110a to the second end 110b of the first waveguide 110, the rate of coupling of light from the first waveguide 110 to the second waveguide 120 gradually increases with propagation distance from the first end 110a to the second end 110b of the first waveguide 110 in a manner which can reduce or prevent backscattering of light, which can otherwise arise at abrupt optical interfaces, for example. Additionally, the second portion 124 of the second waveguide 120, in having an increasing width along the third propagation axis 103, can expand a mode field profile of the third mode from the first end 124a of the second portion 124 to the second end 124b of the second portion 124 and also facilitate gradual coupling of light from the first waveguide 110. It will be appreciated that a portion of light may couple directly from the first waveguide 110 into the second portion 124 of the second waveguide 120, rather than into the first portion 122 of the second waveguide 120.

[0039] Figure 5 illustrates an example of a structure 100-2 which has comparable constituent components and overall function to the structure 100 of Figures 1-4, in that the structure 100-2 comprises a first waveguide 110-2 and a second waveguide 120-2 which is on the first waveguide 110-2 and configured to couple light out of the first waveguide 110-2. However, unlike the examples of Figures 1-4, in which the third light propagation axis 103 overlies the first light propagation axis 101, and can be considered to be laterally aligned, the structure 100-2 exhibits a lateral misalignment. In Figure 5, the first waveguide 110-2 has a first light propagation axis 101a, and the second waveguide 120-2 has a third light propagation axis 103a (the second light propagation axis is not depicted here). However, in plan view, the third light propagation axis 103a does not overlie the first light propagation axis 101a, and is instead parallel to, but offset from, the first light propagation axis 101a by a lateral misalignment distance 6. As per the examples of Figures 1-4, the third light propagation axis 103a is located in a higher plane, in the vertical direction, than the first light propagation axis 101a. An end portion 122a-2 of the second waveguide 120, located further from the first light propagation axis 101a compared with other portions of the second waveguide 120 therefore overhangs the first waveguide 110-2 by an increased distance compared with the first end 122a of the first portion 122 of the secondwaveguide 120 in Figures 1-4. Figure 6 illustrates a second example of such a lateral misalignment. In this example, a second waveguide 120-3 is laterally misaligned relative to a first waveguide 110-3 by a lateral misalignment distance 6’. In this example, an end portion 122a-3 of the second waveguide 120 does not overhang the first waveguide 110-2. In either of Figures 5 or 6, the structures 100-2, 100-3, being in accordance with the present disclosure, can still exhibit low-loss coupling despite the presence of lateral misalignment.

[0040] In some examples, the first portion 122 of the second waveguide 120 extends by a length laof between 100 and 200 micrometres, and has a width waof no less than 500 nanometres, and in some such examples has a width waof 1 micrometre or greater. The angle a between the first light propagation axis 101 and the second light propagation axis 102 has a magnitude of between 0.2 degrees and 1 degree in these examples. This can give rise to a tip end displacement, T, of between 2 and 3 micrometres. The second portion of the second waveguide increases in width from the width wbof no less than 500 nanometres to a width wcof 1.5 micrometres or greater, along a length lbof more than 100 micrometres. The first waveguide 110 is formed of silicon and the second waveguide 120 is formed of a compound semiconductor. The intermediate layer 130 is formed of Benzocyclobutene and provides a separation distance of 30 nanometres between the first waveguide 110 and the second waveguide 120. It is an insight of the inventors that a structure of such dimensions and materials can be particularly straightforward to manufacture whilst facilitating low loss transmission.

[0041] Figure 7 depicts a method of manufacture of the structures 100, 100-2, 100-3 described previously. In particular, Figure 7 depicts transfer printing of the second waveguide 120, attached to a further substrate 220, onto the first waveguide 110, and in particular onto the intermediate layer 140, which in this example comprises Benzocyclobutene. The skilled person will be familiar with transfer printing, but, in general, transfer printing comprises first forming a structure, which is to be transfer printed, onto a sacrificial layer on a source wafer. The structure is freed from the source wafer by removing the sacrificial layer, and the structure is picked up using an elastomeric stamp, e.g. corresponding to the further substrate 220 of Figure 7. Theelastomeric stamp transfers the structure to a target substrate, e.g. the intermediate layer 140 of Figure 7, whereupon it is integrated into e.g. a PIC. In this example, the first waveguide 110 is formed of silicon and the second waveguide 120 is formed of a compound semiconductor such as InGaAs. Figure 7 therefore depicts a heterogeneous integration method such that the resultant PIC 10 comprises both silicon-based and compound semiconductor-based components. The structure 100 is tolerant to lateral misalignment which might occur between the second waveguide 120 and the first waveguide 110 during the transfer printing process.

[0042] Figures 8a-8c present simulation data of optical performance of structures 100, 100-2, 100-3 according to the present disclosure. Each plot depicts transmission as a function of length of the first portion of the second waveguide, and tip end displacement. Tip end displacement refers to a distance, T, between: the first end of the first portion of the second waveguide; and the first light propagation axis. This distance, T, is depicted in Figure 3. Contained within a dashed contour in Figures 8a-8c is a region of low loss transmission, such that light couples from the first waveguide to the second waveguide with a transmission loss of less than 0.5 dB. As the skilled person will appreciate, a transmission loss measurement compares the amount of input power Pinto output power Poutaccording to a function such as an insertion loss function Loss = 101og10(— The lower the loss value, the less power is lost ’out

[0043] through the structure. 0.5 dB of loss is approximately a 10-15% reduction in optical power through the structure. In some examples, therefore, a transmission loss of the structure, based on light propagated through the first waveguide and subsequently coupled into the second waveguide, is less than 0.5 dB. In other examples, greater losses may be tolerable and so the transmission loss of the structure may be higher.

[0044] Figure 8a depicts simulation data for an example with negative lateral misalignment, corresponding to Figure 6, for example, with lateral misalignment 8’ = -0.5 micrometres. Figure 8b depicts simulation data for an example with no lateral misalignment, such as that depicted in Figures 1-4. Figure 8c depicts simulation data for an example with positive lateral misalignment, corresponding to Figure 5, for example, with lateral misalignment 8 = 0.5 micrometres. It should be understood that the use of terms “positive” and “negative” with respect to lateral misalignment aremerely conventions arising from the direction of the y reference axis; in a different reference system, the terms may be used differently. The simulation data of Figures 8a-8c demonstrate the relative robustness of the structures 100, 100-2, 100-3 to misalignment of the first waveguide with the second waveguide, and the low loss transmission which the structures 100, 100-2, 100-3 can facilitate.

[0045] The above examples are to be understood as illustrative examples. Further examples are envisaged.

[0046] For example, whilst the structure may be formed by transfer printing of the second waveguide onto the first waveguide, or vice versa, the structure is not limited to this method of fabrication. Other manufacture methods can be used such as methods employing photolithography, etching, and deposition. In some examples, a flip-chip technique may be used to position the second waveguide onto the first waveguide, or vice versa. A flip-chip technique is a method for interconnecting semiconductor devices, wherein a semiconductor die is flipped and directly bonded to a substrate, e.g. using conductive bumps. In some examples, a pick-and-place technique may be used to position the second waveguide onto the first waveguide, or vice versa. A pick-and-place technique is a method for interconnecting semiconductor devices, wherein a component is removed from a donor substrate and placed onto a target substrate. The component is removed using a pick-up mechanism such as a robotic arm.

[0047] The skilled person will appreciate that the description of orientation of the constituent parts of the structure with respect to a set of reference axes is for explanatory clarity. In the example of Figure 7, the second waveguide is depicted as being placed onto the first waveguide. That is, during manufacture, the second waveguide is moved relative to the first waveguide and the first waveguide remains static, and the first waveguide is below the second waveguide. In some examples, however, during manufacture the first waveguide may be moved relative to the second waveguide, with the second waveguide remaining static. In yet further examples, during manufacture both waveguides may be moved. Furthermore, the second waveguide is depicted as being above the first waveguide. In some examples, however, the first waveguide may be above the second waveguide in some reference frame. The structure may undergo further rotation or translation as part of the manufacturing process, for example.Similarly, whilst the structure is beneficial in the context of heterogeneous integration, the structure may also be used in the context of homogeneous architectures. That is, the first waveguide and the second waveguide can be formed of the same material as each other, or similar materials, and need not be formed of disparate materials.

[0048] In the examples of Figures 1-8, an intermediate layer of BCB is provided. This can e.g. prevent or reduce interaction between materials of the first waveguide with materials of the second waveguide, for example, or help produce a planar surface on top of the first waveguide for improved transfer printing of the second waveguide, for example. However, in other examples, there may be no intermediate layer, and the second waveguide may be formed directly on the first waveguide. Similarly, in other examples, the intermediate layer may be formed of two or more sublayers, and so an intermediate layer need not refer to a single layer of one material, for example.

[0049] In the examples of Figures 1-8, the second waveguide comprises two distinct portions which are optically coupled at an interface. It will be appreciated that the first portion performs a function of slowly introducing and increasing coupling strength with a simple-to-manufacture structure. The second portion can allow the mode field profile to be increased adiabatically to a desired size for use in the PIC. In other examples, the second waveguide may be formed of a single monolithic structure which nevertheless performs the functions of the two portions described previously. As a result, there may not be a specific interface between a first distinct portion and a second distinct portion, but rather, for example, the geometry of the second waveguide may give rise to an effective first portion and an effective second portion.

[0050] In an example, at least a section of the first portion of the second waveguide may be curved, rather than straight as depicted in Figures 1-8. Figure 9 depicts an example of a first portion 122-2 of a second waveguide 120-2 which comprises a second light propagation axis 102-2 that follows a curved path. The second waveguide 120-2 is on a first waveguide 110 according to examples described previously herein. An angle between the second light propagation axis 102-2 and the first light propagation axis 101 of the first waveguide 110 therefore has different values depending upon the point along the second light propagation axis 102-2 from which the angle is considered,but may nevertheless be oblique for all points along the second light propagation axis 102-2, or a majority of points along the second light propagation axis 102-2.

[0051] In an example, the first portion has non-uniform width along its length. For example, the first portion may have a tapering width along its length, e.g. as shown in Figure 10, which depicts an example of a first portion 122-3 of a second waveguide 120-3. The second waveguide 120-3 is on a first waveguide 110 according to examples described previously herein. The first portion 122-3 of the second waveguide 120-3 has a first width wa2at a first end and a second, larger width wa3at a second end, the first portion 122-3 of the second waveguide 120-3 gradually increasing in width from the first end to the second end, and the second end closer to a second portion 124 of the second waveguide 120-3 than the first end is to the second portion 124 of the second waveguide 120-3. That is, as a second light propagation axis 102-3 of the second portion 122-3 approaches the first light propagation axis 101 of the first waveguide 110, the first portion 122-3 of the second waveguide 120-3 increases in width. The change in width of the first portion 122-3 may be relatively small, such as an increase in less than 5% over the length of the first portion 122-3.

[0052] In examples, the first portion may have both linear and curved sections, and may have sections of uniform width and sections of non-uniform width, whilst having a second light propagation axis which is obliquely angled relative to the first light propagation axis. In some examples, the first portion may have two or more linear sections.

[0053] It will be appreciated that a mode supported by a waveguide may have a mode field profile which corresponds to light being concentrated around a geometric centroid of the waveguide structure, in a cross-sectional plane perpendicular to the light propagation axis. In other words, light may typically be guided down the central longitudinal axis of a waveguide. In some examples, however, the mode field profile may correspond to light being guided in an off-centre region of the waveguide crosssection. Accordingly, the function of the structure described herein can be considered in terms of arrangement of the modes and light propagation axes the waveguides respectively support, and whilst this may correlate with the structural geometries of the waveguides, the two are not necessarily linked. In particular, e.g. misalignment of twowaveguide structures may not mean that the light propagation axes of modes supported by these waveguide structures are themselves misaligned.

[0054] A description of some terms and features used previously is now given, to elaborate on features of examples described herein.

[0055] A PIC herein integrates a plurality of photonic functions, for example any of an optical amplifier, an electro-optical modulator, an interferometer, a Mach-Zehnder interferometer, a grating, a laser or a photodiode, though other photonic functions are envisaged. In some examples, a PIC is configured for use with at least one of ultraviolet light, visible light, or infrared light. Optical radiation e.g. includes at least one of ultraviolet light, visible light, or infrared light. In some examples, a PIC comprises an electrical circuit. PICs may be used for communications devices, biomedical devices, and photonic computing, but other applications are envisaged.

[0056] A component of a PIC can be considered a structure, building block or assembly designed to perform a particular optical and / or electrical function in the PIC, such as coupling, waveguiding, amplification, modulation, interfere or another optical function. A component can be electrically active or electrically passive depending, e.g., on whether the component comprises electrodes for application of a voltage and / or electrical current for the component to perform the optical function.

[0057] A waveguide herein is for guiding light; when a waveguide is in use light propagates along the waveguide. A waveguide comprises a core and cladding at least partly in contact with the core. Properties of a waveguide including, for example: a boundary of the waveguide, a boundary between the core and the cladding, the refractive index of the core, the refractive index of the cladding, and / or the structure of the waveguide at least partly confine light propagating along the waveguide to within the waveguide. For example, light propagating along the waveguide might be predominantly within the core. In some examples, a refractive index difference at the boundary between the core and the cladding confines light to propagate substantially within the core, in that a majority of optical power is confined to the core. An evanescent field may exist in the cladding when light is guided by the waveguide. The cladding may comprise a solid structure; however, in some examples the cladding comprises gas, liquid and / or a vacuum in contact with the core. The core may have agreater refractive index than the cladding for the wavelengths of light guided by the waveguide. In some examples, the cladding comprises a plurality of portions, e.g., with different refractive indices. Examples of such cladding include step-index cladding and graded-index cladding. In some examples the waveguide comprises a plurality of cores; such waveguides may be referred to as multi-core waveguides. In some examples, the waveguide is at least one of: a ridge waveguide, a semiconductor junction or an electro-refractive modulator.

[0058] Light within a waveguide is guided by modes. A mode supported by a waveguide can be considered to have an effective refractive index which arises from a combination of the structural geometry of the waveguide and the material refractive index, or indices, of the constituent core and cladding layers. The effective refractive index of a mode determines the rate of propagation of light through the waveguide in that particular mode. Modification of the effective refractive index changes the optical path length of the waveguide. As those skilled in the art will appreciate, the optical path length OPL for a path of geometrical length I in a medium of constant refractive index n is given by:

[0059] OPL = nl

[0060] and, accordingly, a change in the refractive index will alter the optical path length of a waveguide even when the geometrical length I remains unchanged.

[0061] A mode has a mode field profile which describes the distribution of optical field intensity in a cross-section perpendicular to the light propagation axis. A mode field diameter is a metric which quantifies a width of the mode field profile. An adiabatic transition between mode field profiles can occur when structural changes in the waveguide, such as a change in width, height, or cross-sectional area, occur gradually enough that the guided mode changes without significant scattering or loss.

[0062] An optical coupling is such that light propagates between the optically connected elements. In other words, light energy is transferred from a first component to a second component. The optically coupled optical elements are, for example, configured such that light may propagate through free space between the optically coupled optical elements and / or the optically coupled optical elements are connected by an intermediate waveguide such that light may propagate through the waveguide between the optically coupled optical elements. As the skilled person will appreciate,optical as used herein refers to at least one of ultraviolet, visible, mid-infrared, infrared C-band, or infrared light.

[0063] Generally speaking, one arrangement for optical coupling between waveguides, for example, may occur by a first waveguide having an exit facet which faces an entry facet of a second waveguide, such that light propagates from the exit facet to the entry facet. Coupling may be enhanced by ensuring that the exit facet is proximate or directly in contact with the entry facet. Furthermore, coupling may be enhanced by ensuring a mode field profile of the first waveguide substantially overlaps with a mode field profile of the second waveguide, wherein substantially overlaps can mean to overlap a majority of the mode field profile. Coupling being “enhanced” refers to increasing a proportion of light being coupled from the first waveguide into the second waveguide compared with a proportion being lost, such as by being reflected by the entry facet of the second waveguide. This can describe how light propagates from the first portion of the second waveguide to the second portion of the second waveguide as per the examples described herein, for example.

[0064] Optical coupling between waveguides, for example, may also and / or alternatively occur by evanescent field coupling. A first waveguide is brought into proximity with a second waveguide such that a first evanescent field of a first mode of the first waveguide overlaps with a second evanescent field of a second mode of the second waveguide. This can describe how light couples from the first waveguide into the second waveguide as per the examples described herein, for example. The strength of coupling, or in other words the rate at which light transfers from the first waveguide to the second waveguide per unit length propagated, can be dependent upon properties such as the degree of spatial overlap of the evanescent fields and the similarity between effective refractive indices of the first mode and the second mode. Light can “beat” between the first and second waveguide in that, given sufficient propagation distance, light will transfer initially from the first waveguide to the second waveguide before transferring back from the second waveguide to the first waveguide. Selecting the distance along which the first and second waveguides are coupled can therefore allow light to transfer from the first waveguide to the second waveguide but prevent the return transfer to facilitate total optical coupling, for example.A substrate may also be referred to as a chip, a slice, a wafer, or a layer. A substrate is, e.g., a generally planar or relatively thin portion of material, and in some examples is crystalline. A substrate may be a disc or part of a disc of crystalline Si for use in a semiconductor fabrication plant, and in some such examples is a 125 gram, 300 millimetre diameter disc. A substrate may alternatively be a disc or part of a disc of crystalline InP for use in a semiconductor fabrication plant, and in some such examples is a 25 millimetre, 51 millimetre, 76 millimetre, 100 millimetre, 200 millimetre or 300 millimetre diameter disc. A substrate referred to herein is, for example, a single layer of the same homogenous material, though it is envisaged for other examples that a substrate instead comprises one or more layers or portions each deposited or formed independently of each other (for example one after another during a manufacture process to form a stack of sub-layers which together could be considered a substrate). In some examples, a substrate comprises portions of different materials, for example, for fabrication.

[0065] In some examples, the substrate herein is a semiconductor, a III-V semiconductor, a polymer, and / or a dielectric. In some examples, the substrate comprises at least one of: silicon (Si), gallium (Ga), germanium (Gr), lithium niobate (LiNbCh), graphene (C), indium (In), or an alloy, oxide, nitride, or phosphide of at least one of such. In some examples, the substrate comprises an electrical-insulator. In some examples, integrated photonic circuit elements are formed on a silicon-on-insulator substrate or a silicon nitride substrate, for example. In some examples, an electrical conductivity of the substrate at 20 Celsius (293 Kelvin) is less than at least one of 0.00001 Siemens per metre, 0.0001 Siemens per metre, 0.001 Siemens per metre, 0.01 Siemens per metre, 0.1 Siemens per metre, 1 Siemens per metre, or 10 Siemens per metre. In some examples, an electrical resistivity of the substrate at 20 Celsius (293 Kelvin) is more than at least one of 0.1 Ohm metres, 1 Ohm metres, 10 Ohm metres, 100 Ohm metres, 1000 Ohm metres, 10000 Ohm metres, or 100000 Ohm metres.

[0066] In some examples, the layer or portion herein is a single layer of the same homogeneous material, though it is envisaged for other examples that a layer instead comprises one or more sub-layers or portions each deposited or formed independently of each other (e.g., one after another during a fabrication process to form a stack of sublayers which together could be considered a layer). A layer or portion may have sub-portions of different materials, for example, for fabrication. Sub-portions of a layer or portion may have different dopant concentrations.

[0067] In some examples at least one of: the first and / or second waveguide comprises indium gallium arsenide phosphide (InGaAsP).

[0068] In some examples, any of the portions, layers, materials, or electrodes described herein, comprises at least one of a semiconductor, a dielectric, or a polymer.

[0069] In various examples, a semiconductor described herein, comprises at least one of Si, InP, gallium arsenide (GaAs), gallium antimonide (GaSb), gallium nitride (GaN), indium gallium arsenide (InGaAs), indium gallium arsenide phosphide (InGaAsP), indium aluminium arsenide (InAlAs), indium aluminium gallium arsenide (InAlGaAs), AlGaAs, or InGaAsP. Other materials are envisaged in further examples. A compound semiconductor, as used herein, means a semiconductor formed of two or more elements. These elements may be from different groups on the periodic table. A compound semiconductor may be formed of binary (two element), ternary (three element) and quaternary (four element) alloys. A compound semiconductor can be contrasted with elemental semiconductors such as silicon and germanium. The compound semiconductor may be a III-V semiconductor, for example. In various examples, other materials used in integrated circuit manufacturing such as SiN, silicon oxide (SiO2), aluminium oxide (A12O3, or alumina), tantalum pentoxide (Ta2O5 or tantala), aluminium nitride (AIN) or lithium niobate (LiNbO3) may be used. Waveguides according to the present disclosure may be a solid-state waveguide. Waveguides according to the present disclosure may be formed of a semiconductor or dielectric material, for example. For example, waveguides according to the present disclosure may be formed of SiN in combination with SiO2, or may be formed using thin-film lithium niobate, for example.

[0070] It is to be understood that any feature described in relation to any one example may be used alone, or in combination with other features described, and may also be used in combination with one or more features of any other of the examples, or any combination of any other of the examples. Furthermore, equivalents and modifications not described above may also be employed without departing from the scope of the accompanying claims.

Claims

22CLAIMS1. A structure for a photonic integrated circuit, the structure comprising:a first waveguide configured to guide light along a first light propagation axis; anda second waveguide on the first waveguide, the second waveguide configured to couple light out of the first waveguide, and the second waveguide comprising:a first portion configured to guide light along a second light propagation axis at an oblique angle to the first light propagation axis, anda second portion configured to receive light from the first portion and to guide light along a third light propagation axis, the third light propagation axis parallel to, or at a less oblique angle than the second light propagation axis to, the first light propagation axis.

2. The structure of claim 1, wherein a section of the first portion is partially on the first waveguide.

3. The structure according to claim 1 or 2, wherein the first portion has a width perpendicular to the second light propagation axis and the width of the first portion is substantially uniform along the second light propagation axis.

4. The structure according to any previous claim, wherein the first portion has a width perpendicular to the second light propagation axis and the width of the first portion is greater than or equal to 500 nanometres.

5. The structure according to any previous claim, wherein the second light propagation axis is substantially straight.

6. The structure according to any one of claims 1 to 5, wherein the second light propagation axis is curved.

7. The structure according to any previous claim, wherein the oblique angle between the second light propagation axis and the first light propagation axis has a magnitude of between 0.2 degrees and 10 degrees.

8. The structure according to claim 7, wherein the oblique angle has a magnitude of between 0.2 degrees and 2 degrees.

9. The structure according to claim 7 or 8, wherein the first portion has a length, in a direction parallel to the second light propagation axis, of between 100 and 200 microns.

10. The structure according to any previous claim, wherein the second portion is tapered, having a width, perpendicular to the third light propagation axis, which increases in a direction away from the first portion along the third light propagation axis.

11. The structure according to claim 10, wherein the width of the second portion changes adiabatically along the third light propagation axis.

12. The structure according to any previous claim, wherein the third light propagation axis substantially overlies the first light propagation axis.

13. The structure according to any previous claim, wherein the first portion comprises a first end facet and the second portion comprises a second end facet, the first end facet facing the second end facet, the first portion and second portion configured such that light guided by the first portion propagates through the first end facet and the second end facet into the second portion.

14. The structure of claim 13, wherein the first end facet is in contact with the second end facet.

15. The structure according to claim 13 or 14, wherein the first end facet has a first width generally perpendicular to the second light propagation axis, the second end facet has a second width generally perpendicular to the third light propagation axis, and the first width the same as the second width.

16. The structure according to any previous claim, comprising an intermediate layer disposed between the first waveguide and the second waveguide.

17. The structure according to any previous claim, wherein the first waveguide comprises silicon and the second waveguide comprises a compound semiconductor.

18. The structure of claim 17 when dependent on claim 16, wherein the intermediate layer comprises Benzocyclobutene (BCB).

19. A photonic integrated circuit (PIC) comprising the structure of any previous claim.

20. A method of forming a part of a photonic integrated circuit, the method comprising:forming a first waveguide configured to guide light along a first light propagation axis; andforming a second waveguide on the first waveguide, the second waveguide configured to couple light out of the first waveguide, and the second waveguide comprising:a first portion configured to guide light along a second light propagation axis at an oblique angle to the first light propagation axis, anda second portion configured to receive light from the first portion and to guide light along a third light propagation axis, the third light propagation axis parallel to, or at a less oblique angle than the second light propagation axis to, the first light propagation axis.

21. The method of claim 20, wherein forming the second waveguide on the first waveguide comprises transfer printing the second waveguide onto the first waveguide.

22. The method of claim 21, wherein the first waveguide comprises silicon, the second waveguide comprises a compound semiconductor, and transfer printing the second waveguide onto the first waveguide comprises transfer printing the second waveguide onto an intermediate layer of Benzocyclobutene (BCB).

23. The method of claim 20, wherein forming the second waveguide on the first waveguide comprises using a flip-chip and / or pick-and-place technique.

24. A part of a photonic integrated circuit manufactured by any one of claims 20 to

Images