WO2022002578A1

WO2022002578A1 - Integrated optical circuit with waveguide lens

Info

Publication number: WO2022002578A1
Application number: PCT/EP2021/066064
Authority: WO
Inventors: Manuel DECKER; Sören SCHMIDT
Original assignee: Carl Zeiss Ag
Priority date: 2020-07-02
Filing date: 2021-06-15
Publication date: 2022-01-06
Also published as: DE102020117491A1; DE102020117491B4

Abstract

The invention relates to an integrated optical circuit (10) comprising an optical planar waveguide (12), which has a core layer (14) extending on a first plane, and a waveguide lens (24; 24'). The waveguide lens extends on a second plane (22) parallel to the first plane and collects or scatters light guided in the planar waveguide (12). The waveguide lens (24; 24') has a constant thickness (d_L) in one direction (x) perpendicular to the second plane (22). Multiple sub-wavelength structures (341, 342, 343; 34; 34'; 34''; 34''') are arranged on the second plane (22) in a transition region (32) which extends at least over a part of the circumference of the waveguide lens (24; 24'), said structures having the same thickness perpendicularly to the second plane (22). The sub-wavelength structures are distributed with a filling factor (F) which decreases continuously or incrementally as the distance to the waveguide lens increases. In this manner, losses based on a mode mismatch are reduced.

Description

Integrated optical circuit with waveguide lens

BACKGROUND OF THE INVENTION

1. Field of the invention

The invention relates to the field of integrated optical circuits (English. Photonic Integrated Circuits, PIC). The invention relates in particular to such circuits with a layer waveguide and a waveguide lens which extends in a plane parallel to the plane of the layer waveguide.

2. Description of the prior art In integrated optical circuits, a large number (sometimes hundreds) of optical functions, e.g. light generation and transmission, beam splitting, intensity or phase modulation, filtering or switching, are integrated on a chip. The material platform most frequently used commercially for integrated optical circuits has so far been indium phosphide (InP). In contrast to electronic integration, in which silicon is the predominant material, numerous other material systems are also used in integrated optical circuits, including electro-optical crystals such as lithium niobate, silicon dioxide on silicon, silicon on insulator, silicon nitride, various Polymers and semiconductor materials such as GaAs and InP.

The functions that can be integrated also include the focusing and scattering of light that propagates in an optical layer waveguide. In aschichtwellenlei ter the light is only guided in one direction; in the other two directions the light spreads like in free space (but with a different refractive index than 1). Since there are no infinitely wide layers in practice, one speaks of layer waveguides even if the wave-guiding area, also known as the core layer, has a width that is much greater than its thickness. A »1 then applies to the aspect ratio A of the waveguide cross-section. Various types of waveguide lenses have been proposed for focusing and dispersing, including so-called mode index lenses, geodesic lenses and Fresnel lenses. Geodetic lenses and other waveguide lenses with continuous thickness profiles can, however, only be produced photolithographically with great effort or with insufficient accuracy.

From a manufacturing point of view, waveguide lenses that do not have a continuous thickness profile but a constant thickness are more favorable. Such waveguide lenses are arranged in a plane which runs parallel to the sheet waveguide. In order to have a spatially inhomogeneous effect on the light passing through, waveguide lenses with constant thickness - similar to a conventional lens - have at least one curved surface. Curved surfaces can be dispensed with if the waveguide lens has an inhomogeneous refractive index distribution, as is known, for example, from EP 0 261 849 A1.

So far, however, the problems with such waveguide lenses due to reflection, decoupling and scattering losses at the interface between the sheet waveguide and the waveguide lens or - if the sheet waveguide and the waveguide lens do not touch each other directly - in the transition area between the sheets have not been solved lenleiter and the waveguide lens arise. These losses depend, among other things, on the materials used, their refractive indices and the layer thickness. These losses are caused by insufficient overlap of the modes in the sheet waveguide and in the waveguide lens in the area of the interface of the waveguide lens (so-called mode mismatch).

In connection with strip waveguides, which restrict the propagation of light in two orthogonal directions, it has been proposed to create sub-wavelength structures by structuring the strip waveguide and thereby influence the mode index in such a way that losses caused by mode mismatch can be reduced. An overview of the main features and various applications of this approach can be found in an article by R. Halir et al. entitled "Waveguide sub-waveiength structures: a review of principles and appäcations", Laser & Photonics Reviews, 9 (1), 25-49. US 2012/0183250 A1 deals with a further development of this approach.

From US 2008/0193080 A1 it is known to provide the core of a strip waveguide with subwel lenlängenstructures in order to couple light from the strip waveguide in a second strip waveguide or in an optical fiber. No. 6,980,720 B2 describes a mode transformation from a strip or sheet waveguide into a strip waveguide arranged above it. For this purpose, the latter tapers in a wedge-shaped manner at the beginning and end, it also being possible for the wedges to be approximated by means of stepped structures.

SUMMARY OF THE INVENTION

The object of the invention is to provide an integrated optical circuit with a sheet waveguide and a waveguide lens of constant thickness, which is easy to manufacture and in which the optical losses caused by the waveguide lens are low.

According to the invention, this object is achieved by an integrated optical circuit which has an optical layer waveguide which has a core layer extending in a first plane. The integrated optical circuit further comprises a waveguide lens which extends in a second plane parallel to the first plane and is set up to collect or to scatter light guided in the layer waveguide. In a direction perpendicular to the second plane, the waveguide lens has a constant thickness. According to the invention, a plurality of sub-waveguide structures, which have the same thickness perpendicular to the second plane, are arranged in the second plane in a transition region which extends at least over part of the circumference of the waveguide lens. The sub-waveguide structures are distributed with a fill factor that decreases continuously or gradually as the distance between the waveguide lens increases.

The invention is based on the knowledge that by arranging subwave length structures in a transition region adjoining the waveguide lens, the mode overlap can be improved so that the reflection and scattering losses decrease at the interface of the waveguide lens. That way it is It is possible to reduce the losses caused by mode mismatches, which are typically in the order of magnitude of around 15%, for common materials and layer thicknesses to values below 10%. Since not only the waveguide lens, but also the sub-wavelength structures each have the same thickness, the integrated optical circuit according to the invention can be produced with high accuracy in a simple manner using the known photographic method.

The material density of a material is called the fill factor F. For a material i, the fill factor F, is thus defined by

Fi = dVi / dV, where dV is the sum per volume unit

the fill factor is equal to 1. If the sub-wavelength structures consist of several materials with different refractive indices, this can be taken into account by assuming a uniform material that has an effective mean refractive index.

The concept of the fill factor can also be applied to configurations in which both the waveguide lens and the subwavelength structures are formed as cavities in a surrounding material.

In principle, the thickness of the subwavelength structures can differ from the thickness of the waveguide lens by a certain amount. In terms of the manufacturing process, however, it is simpler if the thickness of the subwavelength structures is equal to the thickness of the waveguide lens.

For the formation of the subwavelength structures, it is only important that the fill factor decreases continuously or gradually with increasing distance from the waveguide lens. In the prior art, various possibilities are known as to how sub-wavelength structures can be configured that meet this requirement. In one embodiment, at least some subwavelength structures are designed as webs with an at least substantially rectangular cross section. Everyone The web is at a constant distance from the waveguide lens, the distances between different webs and the waveguide lens being different. In the case of a circular disk-shaped waveguide lens, this means, for example, that the webs are circular rings or circular ring sections which are arranged concentrically to the geometric center of the waveguide lens. The webs can completely or only partially surround the waveguide lens. In general, it is sufficient to provide the webs only where they are exposed to the optical mode guided in the layer waveguide.

In another embodiment, at least some sub-wavelength structures are formed integrally with the waveguide lens. One-piece sub-waveguide structures or sub-waveguide structures that are separate from the waveguide lens can have the shape of a polygon, in particular a triangle or a trapezoid, in a section parallel to the second plane. In this way, by defining the angles of the triangle or the trapezoid, the decrease in the fill factor with increasing distance from the waveguide lens can be set. The polygon can in particular have the shape of an (e.g. isosceles) triangle. These triangular sub-wavelength structures are evenly distributed over at least part of the circumference of the waveguide lens, the base of the triangles running at least substantially parallel to the circumference of the waveguide lens. These sub-wavelength structures can also be designed in one piece with the waveguide lens, so that the base of the triangles merges into the circumference of the waveguide lens.

In principle, it comes into consideration that the material from which the waveguide lens is made is in direct contact with the core of the sheet waveguide. However, it is often more favorable if the waveguide lens and preferably also the subwavelength structures are separated from the core layer of the sheet waveguide by at least one (thin) further layer. With certain materials such as silicon, such a layer is formed without any special action by oxidation in the surrounding air. The thickness of the intermediate layer represents an additional degree of freedom with which the mode overlap can be set. The wave-guiding regions, that is to say both the core layer of the waveguide layer and the waveguide lens, are preferably embedded in a surrounding material (ciadding), as is known per se in the prior art. Such an environmental mate rial protects the structures enclosed by it from oxidation and contamination. If the waveguide lens consists of a material with a homogeneous refractive index and has a constant height, it must have at least one curved interface which is at least substantially perpendicular to the second plane.

If, on the other hand, the waveguide lens consists of a material that has an inhomogeneous refractive index distribution, a curved interface can be dispensed with. In this case, the waveguide lens can have a rectangular shape, the circumference of which is adjoined by the transition region which contains these subwavelength structures.

If the integrated optical circuit has a light source which is adapted to

To generate light with a central wavelength l, the dimensions of the sub-wavelength structures perpendicular to a direction of propagation of the light should be smaller than k · l, where k = 1 and preferably k = 0.5 and more preferably k = 0.1.

If it is mentioned above that structures are located in a plane, this means that the structures adjoin this plane or the plane penetrates these structures.

BRIEF DESCRIPTION OF THE DRAWINGS

Exemplary embodiments of the invention are explained in more detail below with reference to the drawings. In these show:

FIG. 1: a perspective view of a section from an integrated optical circuit according to the invention with a waveguide lens which is arranged over a layer waveguide; FIG. 2: a cross section through the detail shown in FIG. 1 along the line II-II;

FIG. 3: the integrated optical circuit shown in FIGS. 1 and 2 in a sectional plan view in which the waveguide lens and the transition region can be seen;

FIG. 4: a cross section similar to FIG. 2, but supplemented with schematic representations of the optical modes guided in the sheet waveguide and in the waveguide lens;

FIG. 5: a graph in which the fill factor F is shown schematically for different variants as a function of the distance z from the waveguide lens;

FIGS. 6 and 7 different variants for the formation of the subwavelength structures in the transition area along a curved interface of the waveguide lens;

FIGS. 8 and 9 further variants for the formation of the subwavelength structures in the transition area along a straight boundary surface of a rectangular waveguide lens which has a Brecht number profile.

DESCRIPTION OF PREFERRED EMBODIMENTS

1. First embodiment

Figures 1, 2 and 3 show an integrated optical circuit denoted overall by 10 according to a first embodiment of the invention in a perspective view, a cross-section along the line II-II or in a detail in a plan view.

The integrated optical circuit 10 comprises a layer waveguide 12, the wave-guiding region of which is formed by a core layer 14. This extends with a constant thickness de in a first plane and consists of a material with a refractive index ni. The core layer 14 is arranged between a lower cladding layer 16 and an upper man layer 18, which in the illustrated embodiment consist of the same material that has a refractive index r2 <. The lower cladding layer 16 can be the substrate of the integrated optical circuit 10 or be applied to an additional substrate not shown in the figures.

Light that is coupled into the core layer 14 of the layered waveguide 12 in the direction indicated by an arrow 20 is guided in the core layer 14 by total reflection at the upper and lower interfaces to the cladding layers 16, 18 and passes through the layered waveguide 12 along the z -Direction. In some cases it is expedient to also surround the core layer 14 laterally (i.e. in the y direction) by a jacket material, as is known per se in the prior art, but is not shown in the figures.

A second plane 22, which is indicated in FIGS. 1 and 2 with dashed lines, runs parallel to a first plane in which the core layer 14 of the layered waveguide 12 extends. A waveguide lens 24 extends in the second plane, the refractive index n3 of which in the illustrated embodiment is greater than the refractive index r2 of the surrounding upper cladding layer 18. As can be seen in FIG 22 across a constant thickness di_. In order to achieve a lens effect, the waveguide lens 24 has a curved lateral boundary surface 26 aligned perpendicular to the second plane 22, the contour of which is oval in the exemplary embodiment shown, as can best be seen in the top view of FIG. The contour of the lateral interface 26 of the waveguide lens 24 defines its optical properties. In the illustrated embodiment, the contour is convexly curved. As a result of the small distance along the x direction between the waveguide lens 24 and the core layer 14, the optical mode guided in the layered waveguide 12 partially couples into the waveguide lens 24 and experiences a collecting effect there due to the convexly curved lateral interface 26, as shown in FIG FIG. 1 is indicated schematically by two converging light beams 28. The contour of the lateral interface can of course also be in the shape of a circular arc, which results in the effect of a spherical lens.

It can only be seen in FIG. 2 that the waveguide lens 24 is separated from the core layer 14 by an optional intermediate layer 30. The abrupt transition between the layered waveguide 12 and the waveguide lens 24 results in significant decoupling, scattering and reflection losses, which can be attributed to very different optical modes in the two wave-guiding areas. These losses can be calculated by determining the mode overlap integral, the amount of which is equal to 1 for an ideal mode overlap. The size of the mode overlap integral depends on the specific dimensions of the layer waveguide 12, the intermediate layer 30 and the waveguide lens 24 and the materials used. In the event that the core layer 14 consists of silicon nitrite and has a thickness de of 200 nm and the waveguide lens 24 consists of crystalline silicon and has a thickness of 40 nm, the mode overlap integral is approximately 85%, which corresponds to a waveguide loss of 15% is equivalent to.

In order to convert the optical modes in the layer waveguide 12 and in the waveguide lens 24 continuously or gradually into one another and in this way to reduce the light losses, in the integrated optical circuit 10 in the second level 22 in a transition region 32 which extends at least over one Part of the circumference of the waveguide lens 24, a plurality of subwavelength structures 341, 342, 343 are arranged, which have the same thickness di _. like the waveguide lens 24. These sub-wavelength structures 341 to 343 are distributed over the transition region 32 with a fill factor F which decreases continuously or in steps with increasing distance from the waveguide lens 24. In the exemplary embodiment shown, the subwavelength structures 341 to 343 are designed as webs which have an at least substantially rectangular cross section. As can best be seen in FIG. 3, the subwavelength structure 341 has a NEN distance di from the waveguide lens 24, which is the same over the entire circumference of the waveguide lens 24 away. The same applies to the distances d2 and d3 of the sub-wave length structures 342 and 343, where di <d2 <d3 applies. The distances di, d2, d3 increase more and more towards the outside. This results in a concentric arrangement of rings in the plan view, the distance between which from the waveguide lens 24 is becoming larger and larger. In this way, the fill factor F decreases quasi-continuously as the distance from the waveguide lens 24 increases.

The widths of the ring-shaped webs, of which the subwavelength structures 341 to 343 consist, are significantly smaller than the wavelength 1 of the light which passes through the layer waveguide 12. As a result, undesired diffraction effects do not occur at the subwavelength structures 341 to 343. The exact dimensions of the sub wave length structures 341 to 343 must be matched to the actual dimensioning of the layer thicknesses and the materials used in order to minimize the losses. Calculations have shown that with suitably designed subwavelength structures, the amount of the mode overlap integral can be increased to values of over 90%, which corresponds to losses of less than 10%. This corresponds to a reduction of more than 30% compared to waveguide lenses without surrounding sub-waveguide structures.

In the exemplary embodiment shown, the intermediate layer 30 is also located (and specifically likewise structured) below the subwavelength structures 341 to 343, as can be seen in FIG. As a result, the subwavelength structures 341 to 343 also extend in the second plane 22. In other exemplary embodiments, the intermediate layer 30 is not structured, so that the subwavelength structures 341 to 343 do not extend into the second plane 22. FIG. 4 shows the cross section of FIG. 2 in a somewhat enlarged representation. In addition, profiles 36 of optical modes are shown schematically at several locations. The profiles 36 illustrate how the optical mode in the core layer 14 in the transition region 32 is converted into the optical mode in the waveguide lens 24. FIG. 5 shows schematically in a graph how the fill factor F in the transition area 32 between distance coordinates Zi and Z2 continuously or step by step between values of 0 (beginning of transition area 32) and 1 (end of transition area 32 = beginning of the waveguide lens 24) can increase. A linear increase in the fill factor F is shown with a solid line. However, non-linear increases (double-dash-dotted line) or gradual increases (dashed line) are also possible. Which type of increase in the fill factor F leads to the lowest losses due to mode mismatch in the individual case depends on the specific dimensions and materials of the sheet waveguide 12 and the waveguide lens 24. The above explanations only related to the transition area 32 when the light enters the waveguide lens 24. The same considerations apply accordingly to the exit of the light from the waveguide lens 24. The subwavelength structures 341, 342, 343 in the rear section of the transition area 32 also effectively reduce losses due to mode mismatch there. 2. Further exemplary embodiments

Figures 6 to 9 show different variants for the formation of the subwave conductor structures.

In the exemplary embodiment shown in FIG. 6, the sub-wavelength structures 34 have the shape of an isosceles triangle in a section parallel to the second plane 22, the base of each triangle running at least substantially parallel to the circumference of the waveguide lens 24 and being formed in one piece therewith, so that the base of the triangles merges into the circumference of the waveguide lens 24. As a result, a linear increase in the fill factor within the transition region 32 is achieved.

In the exemplary embodiment shown in FIG. 7, the subwave length structures 34 'do not have the shape of isosceles triangles, but of trapezoids, the base of which points towards the waveguide lens 24. In contrast to the exemplary embodiment shown in FIG. 6, several rows of subwavelength structures 34 'are provided, with within each row, all sub-wavelength structures 34 'have the same dimensions. The spacing of the rows is constant, but the size of the subwavelength structures 34 ′ increases as the distance from the waveguide lens 24 decreases. The increase in the fill factor F towards the waveguide lens 24 is thus achieved here by the increasing size of the sub-wavelength structures 34 '.

In the exemplary embodiment shown in FIG. 8, the waveguide lens 24 'does not consist of a material with a homogeneous refractive index. Instead, a refractive index profile was generated by ion implantation, which is indicated by a gray shading in FIG. The refractive index never increases continuously from the edges along the y direction and is greatest on the axis of symmetry of the waveguide lens 24. In this way, a collecting effect can be achieved without the lateral boundary surface 26 of the waveguide lens 24 'having to be curved.

In this exemplary embodiment, the subwavelength structures 34 ″, similar to the exemplary embodiment shown with reference to FIGS. 1 to 3, have the shape of bars with rectangular cross-sections the width of the ridges increases with decreasing distance from the waveguide lens 24 '. The refractive indices of the ridges should increase from the edges to the symmetry axis just like the refractive index in the waveguide lens 24', which is not shown in FIG. 8. ridges as in the figure 8 can of course also be used with waveguide lenses with a homogeneous refractive index and curved interfaces. The webs are then not straight, but rather curved, preferably in such a way that the webs run parallel to the curved interface individual webs decrease in width from the inside to the outside leave. FIG. 9 shows a variant of the exemplary embodiment shown in FIG. The waveguide lens 24 'there also has a rectangular circumference and a refractive index profile. The ridge-shaped subwavelength structures 534 have in this embodiment For example, however, they all have the same width, and the distance between adjacent sub-wavelength structures 534 is also constant. In this exemplary embodiment, an increase in the fill factor F is achieved in that the ridge-shaped subwavelength structures 534 are interrupted, the ratio of ridge to interruption increasing as the distance from the waveguide lens 24 'decreases.

Here too, the refractive indices of the webs should increase from the edges to the axis of symmetry, which is not shown in FIG. As an alternative to this, the interruptions between the webs can become longer as the distance from the axis of symmetry increases, and / or the width of the webs decreases in this direction.

Claims

PATENT CLAIMS

1. Integrated optical circuit (10), with an optical layer waveguide (12) which has a core layer (14) extending in a first plane, and a waveguide lens (24; 24 ') which extends in one of the first plane parallel second plane (22), is set up to collect or scatter light guided in the layer waveguide (12) and has a constant thickness (d _L ) in a direction (x) perpendicular to the second plane (22), thereby characterized in that in the second plane (22) in a transition region (32) which extends at least over part of the circumference of the waveguide lens (24; 24 '), a plurality of sub-wavelength structures (341, 342, 343; 34; 34') ; 34 "; 34 '") which have the same thickness perpendicular to the second plane (22), the subwavelength structures being distributed with a fill factor (F) that increases with increasing distance from the waveguide lens (24; 24 ') decreases continuously or gradually.

2. Integrated optical circuit according to claim 1, characterized in that the thickness of the subwavelength structures is equal to the thickness (di _. ) Of the waveguide lens (24; 24 ').

3. Integrated optical circuit according to one of claims 1 or 2, characterized in that at least some subwavelength structures (341, 342, 343; 34 "; 34 '") are designed as webs with an at least substantially rectangular cross section, each The web has a constant distance from the waveguide lens (24; 24 ') and the distances between different webs to the waveguide lens (24; 24') are different.

4. Integrated optical circuit according to one of claims 1 or 2, characterized in that at least some subwavelength structures (34) are formed in one piece with the waveguide lens (24).

5. Integrated optical circuit according to one of the preceding claims, characterized in that at least some subwavelength structures (34; 34 '; 34 "; 34'") have the shape of a polygon in a section parallel to the second plane (22).

6. Integrated optical circuit according to claim 5, characterized in that the polygon is a triangle, the subwavelength structures (34) are evenly distributed at least over part of the circumference of the waveguide lens (24) and the base of the triangles is at least substantially parallel to the circumference of the Wel lenleitlinse (24) runs.

7. Integrated optical circuit according to one of the preceding claims, characterized in that the waveguide lens (24; 24 ') is separated from the core layer (14) of the sheet waveguide (12) by at least one layer (30).

8. Integrated optical circuit according to claim 7, characterized in that the subwavelength structures (341, 342, 343; 34; 34 '; 34 "; 34'") from the core layer (14) of the sheet waveguide (12) by at least one layer ( 30) are separated.

9. Integrated optical circuit according to one of the preceding claims, characterized in that the waveguide lens (24) has at least one curved Grenzflä surface (26) which is directed at least substantially perpendicular to the second plane (22).