OPTOELECTRONIC TRANSMITTER WITH PHASE-CONTROLLED GROUP ANTENNA WITH INTEGRATED CONTROL DEVICE

Description

TECHNICAL FIELD

[0001] The field of the invention is that of photonic circuits on a chip, and more specifically that of optoelectronic transmitters, of the phase-controlled antenna array type, implemented on a photonic chip. The invention may find application particularly in the field of LIDAR ( Light Detection and Ranging, in English). PREVIOUS STATE OF THE ART

[0002] Optoelectronic transmitters with phased array antennas (PAA, for Optical Phased Array Optoelectronic devices (LIDARs) are devices that emit a monochromatic light beam in free space in a directional manner. They find applications particularly in the field of laser distance detection and estimation (LIDAR), but also in free-space optical communications, holographic displays, and medical imaging.

[0003] There Figure 1AFigure 1 schematically illustrates the operating principle of such an optoelectronic transmitter. A laser source 2 emits an optical signal which is distributed by a power divider 3 into N arms 4 of the optoelectronic transmitter 1. Each arm 4 comprises a phase shifter 5 and an elementary emitter 6, also called an optical antenna. Each elementary emitter 6 emits an optical signal into free space, for example by diffraction; the optical signals then combine by interference to form a light beam. This beam exhibits a far-field emission pattern determined, in particular, by the relative phase Δϕ applied by the phase shifters 5 to the optical signals propagating in the arms 4.

[0004] Such optoelectronic emitters 1 can be made in integrated photonics, that is to say that its various optical components (waveguides, power divider, elementary emitters...) are made on and from a photonic chip.

[0005] In this respect, the figure 1B illustrates schematically and partially an example of such an optoelectronic emitter 1 made on a photonic chip and including an integrated control device, as described in the article by Hulme et al entitled Fully integrated hybrid silicon two dimensional beam scanner, Opt. Express 23 (5), 5861-5874 (2015).

[0006] This optoelectronic emitter 1 comprises, in addition to the laser source 2, the power divider 3, and the array of phase shifters 5 and elementary emitters 6, a control device consisting of a graded-index lens 8A coupled to the output ends of the elementary emitters 6 at its input face 8A.1, and to an array of photodetectors 10 at its output face 8A.2. The control device makes it possible to determine the relative phase Δϕ actually present between the optical signals propagating in the arms 4. For this purpose, an undiffracted portion of these optical signals is transmitted by the elementary emitters 6 to the lens 8A via input waveguides 7. The lens 8A provides an image of the far-field optical radiation emitted by the optoelectronic emitter, which is then detected by the photodetectors 10.A control module 11 allows obtaining far-field optical information (in particular the angle Φ of directivity of the light beam with respect to the Y axis in the XY plane) and thus calibrating the optoelectronic emitter 1.

[0007] However, the graded-index lens is of the photonic crystal type and is formed from a plurality of holes of varying sub-wavelength diameters fabricated in a thin silicon layer of a SOI substrate. The holes are arranged so that the effective refractive index of the optical field exhibits a parabolic profile along the width of the lens. It appears that these holes forming the photonic crystal of the lens are particularly complex to fabricate, and that imperfections in their fabrication significantly impact the performance of such an optoelectronic emitter. Publication WO 2018 / 160729 A2 discloses a three-dimensional (3D) optical sensing system for a vehicle that includes a modular architecture. Light can be transmitted to an optical signal processing module, which may include a photonic integrated circuit (PIC) capable of generating at least one signal with suitable amplitude, phase, and spectral characteristics.The multiple optical signals processed by the optical signal processing module can be sent to beam steering units distributed around the vehicle. These units can direct multiple optical beams toward targets. The returning optical signal can be detected by a PIC receiver comprising a sensor array and using either direct intensity detection or coherent detection. The sensor array can convert the returning optical signal into an electrical signal, which can then be processed by the electronic signal processing unit, allowing for the quantification of target location and velocity information. DESCRIPTION OF THE INVENTION

[0008] The invention aims to remedy at least in part the drawbacks of the prior art, and more particularly to offer a phase-controlled optoelectronic array antenna transmitter on a photonic chip in which the impact of the manufacturing process on its performance is reduced or even eliminated.

[0009] To this end, the object of the invention is a phase-controlled optoelectronic antenna array transmitter on a photonic chip, comprising: a power divider, intended to be coupled at the input to a laser source, and coupled at the output to N waveguides forming the arms of the optoelectronic emitter, with N>1; an array of phase shifters and elementary emitters, arranged in the arms, the elementary emitters being aligned laterally and extending parallel to each other; an integrated control device comprising: input waveguides, coupled to the elementary emitters; a focusing lens, having an input face coupled to the input waveguides, and an output face; output waveguides, coupled to the output face; photodiodes, coupled to the output waveguides; a control module adapted to control the phase shifters according to signals received by the photodiodes.

[0010] According to the invention, the focusing lens is an interferometric lens whose entrance and exit faces are curved and delimit a free-propagating region of homogeneous refractive index, the exit face extending along a focal line of the focusing lens.

[0011] Furthermore, the input waveguides are connected to the input face orthogonally to it.

[0012] Finally, an effective index of guided modes, in a portion called input waveguide correction, is adapted so that the optical paths of the input waveguides are identical to each other.

[0013] Some preferred, but not limiting, aspects of this optoelectronic transmitter are as follows.

[0014] The input waveguides may each exhibit, in the correction portion, a variation of at least one transverse dimension, said transverse dimension variation being predefined for each input waveguide so that the optical paths of the input waveguides are identical to each other.

[0015] Each input waveguide can exhibit, in the correction portion, an adiabatic variation in width, ranging from an identical reference value We0 for each input waveguide to a different maximum value Wec(i) from one input waveguide to another.

[0016] The optoelectronic emitter may include so-called additional waveguides made of a different refractive index than the input waveguides, extending only in the correction portion, and arranged opposite and parallel to the input waveguides so as to allow modal coupling with the input waveguides, the additional waveguides having a predefined length for each of them so that the optical paths of the input waveguides are identical to each other.

[0017] Additional phase shifters can be coupled to said input waveguides in the correction portion, and can apply a relative phase shift between the guided modes in the input waveguides predefined for each of them so that the optical paths of the input waveguides are identical to each other.

[0018] The focusing lens can be a star coupler or a ladder grating.

[0019] The focusing lens may feature a Rowland mount or a confocal configuration.

[0020] The photonic chip can be made from an SOI substrate.

[0021] The optoelectronic emitter may include the laser source. This can be assembled on a so-called upper surface of the photonic chip. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Other aspects, objectives, advantages, and features of the invention will become clearer upon reading the following detailed description of preferred embodiments thereof, given by way of non-limiting example, and made with reference to the accompanying drawings in which: there Figure 1A The already described schematic and partial view of a phase-controlled array antenna optoelectronic transmitter is shown below. figure 1BThe already described figure is a schematic and partial top view of such an optoelectronic transmitter, comprising an integrated control device, according to an example of the prior art; figure 2 is a schematic and partial top view of a phase-controlled array antenna optoelectronic transmitter according to one embodiment, in which the interferometric lens is a star coupler; the figures 3A to 3C These are schematic, partial, top views of a portion of the optoelectronic emitter illustrated on the fig.2 illustrating the focusing of the optical field in three different zones of the exit face of the interferometric lens; the figure 4A is a schematic and partial top view of a part of the optoelectronic emitter illustrated on the fig.2 illustrating the dimensioning of the interferometric lens in a Rowland setup; and the Figures 4B and 4C are detailed views of an area of ​​the entrance face ( fig.4B ) and an area of ​​the exit face ( fig.4C ) of the interferometric lens illustrated on the fig.4A ; there figure 5 is a schematic, partial, top view of a portion of the optoelectronic emitter illustrated on the fig.2 illustrating the longitudinal variation in width of the input waveguides in the correction portion; the figure 6 is a schematic and partial cutaway and exploded view of a portion of the optoelectronic emitter according to an alternative embodiment in which additional waveguides are coupled to the input waveguides in the correction portion; the figure 7 is a schematic and partial cutaway and exploded view of a portion of the optoelectronic emitter according to another embodiment, in which thermo-optical phase shifters are arranged at the level of the input waveguides in the correction portion; the figure 8is a schematic and partial top view of a phase-controlled array antenna optoelectronic transmitter according to another embodiment, in which the interferometric lens is a ladder array; the Figures 9A and 9B illustrate, in top view and in cross section, a part of the optoelectronic emitter according to two other embodiment variants, in which phase change portions are arranged at the level of the input waveguides in the correction portion. DETAILED DESCRIPTION OF SPECIFIC METHODS OF IMPLEMENTATION

[0023] In the figures and throughout the description, the same reference numerals represent identical or similar elements. Furthermore, the various elements are not drawn to scale to ensure clarity. Moreover, the different embodiments and variants are not mutually exclusive and may be combined. Unless otherwise stated, the terms "approximately," "around," and "in the order of" mean within 10%, and preferably within 5%. Furthermore, the terms "between ... and ..." and equivalents mean inclusive of the bounds, unless otherwise specified.

[0024] The invention relates to a phase-controlled array antenna optoelectronic transmitter comprising an integrated control device. Such an optoelectronic transmitter is fabricated on a photonic chip. It is adapted to emit a light beam whose intensity, in the far field, exhibits a specific angular distribution around a principal emission axis. This angular distribution of the far-field light beam emitted by the optoelectronic transmitter is called the 'far-field emission pattern'. It is thus distinguished from the near-field emission patterns of elementary transmitters (optical antennas). The far field (or Fraunhofer zone) corresponds to a distance D greater than the ratio of the square of a major dimension of the elementary transmitter (here, the length Lee along the Z-axis) to the emission wavelength λe, in other words: D > 2Lee / λe.The far-field emission pattern can be adjusted here by the integrated control device.

[0025] There figure 2 is a schematic and partial view of an optoelectronic emitter 1 according to one embodiment. In this example, the optoelectronic emitter 1 comprises a laser source 2, a power divider 3, an array of phase shifters 5 and elementary emitters 6. It also includes a control device, connected to the phase shifters 5, allowing the phase shift Δϕ applied to the optical signals circulating in the N arms 4 to be determined or even modified, in order to improve the performance of the optoelectronic emitter 1, and in particular to generate the desired emission pattern of the far-field light beam emitted by the optoelectronic emitter 1.

[0026] We define here and for the rest of the description a direct orthonormal XYZ frame, where the XZ plane is parallel to the plane of the photonic chip, and the Z axis is oriented along the axis of the elementary emitters 6, the Y axis being thus oriented towards the free space in which the light beam is emitted by the optoelectronic emitter 1.

[0027] The optoelectronic emitter 1 is integrated onto a photonic chip, for example, within the framework of so-called silicon photonics technology. The photonic chip, also called a photonic integrated circuit (PIC, for Photonic Integrated Circuit(in English), includes a support substrate (not shown) from which active photonic components (modulators, diodes, etc.) and passive photonic components (waveguides, multiplexers or demultiplexers, etc.) can be fabricated and optically coupled to each other. In silicon photonics, both the support substrate and the photonic components are made of silicon. The support substrate can therefore be of the SOI type. (Silicon On Insulator, in English).

[0028] The optoelectronic emitter 1 comprises a laser source 2 adapted to emit a pulsed or continuous monochromatic optical signal of wavelength λe. The wavelength can be, for example, 905 nm. The laser source 2 can be wavelength-tunable, notably to modify the emission angle θ formed by the far-field light beam emitted by the optoelectronic emitter with respect to the Y-axis in the YZ plane. The laser source 2 can be a hybrid source consisting of a gain medium made from a III / V compound and bonded to the surface of the photonic chip. Optical reflectors of the Bragg mirror type can thus be fabricated in an integrated waveguide coupled to the gain medium. Alternatively, the optoelectronic emitter 1 may not include the laser source 2, which can be located remotely and therefore not assembled in contact with the surface of the photonic chip.It can then be coupled to an integrated waveguide of the photonic chip, notably by a lattice coupler.

[0029] A power divider 3 is coupled to the output of the laser source 2. It thus has one input port and N output ports, with N > 1. The number N defines the number of arms 4 of the optoelectronic emitter 1. In this example, the power divider 3 is formed of several MMI-type dividers ( Multimode Interferometer (in English) arranged in cascade, but other types of optical components can be used.

[0030] The optoelectronic transmitter 1 comprises an array of N waveguides 4, preferably parallel to each other, each extending from a first end coupled to an output port of the power divider 3 and a second end coupled to an elementary transmitter 6. Each waveguide 4 is therefore adapted to receive an optical signal from the power divider 3, and to allow the propagation of this optical signal to an elementary transmitter 6.

[0031] The optoelectronic transmitter 1 also includes a network of phase shifters 5 and elementary emitters 6. More precisely, a plurality of arms 4 are equipped with at least one phase shifter 5 adapted to modify the phase of the optical signal circulating in the arm in question, and thus to generate a phase difference Δϕ, or relative phase, between the optical signals circulating in adjacent arms 4. The phase shifters 5 are arranged between the power divider 3 and the elementary emitters 6. Each arm 4 may be equipped with a phase shifter 5, or only some of the arms 4, such as every other arm. Furthermore, the reference arm may not include any phase shifters 5.

[0032] Phase shifters 5 can be electro-refractive or thermo-optical. In both cases, the phase change is achieved by modifying the refractive index of the material forming the core of the waveguide 4. This modification of the refractive index can be achieved by changing the free carrier density in the case of the electro-refractive phase shifter, or by changing the applied temperature in the case of the thermo-optical phase shifter.

[0033] The phase shifters 5 are preferably adapted to apply the same relative phase value Δϕ to the optical signals propagating in the arms 4, so as to obtain a determined non-zero angle Φ of the principal emission axis with respect to the Y-axis in the XY plane (orthogonal to the axis of the elementary emitters 6). However, the relative phase Δϕ may not be identical between the arms 4, either to obtain a different far-field pattern, or to take into account and compensate for possible phase errors. These phase errors may arise from the degradation over time of certain components of the optoelectronic emitter 1, non-uniformities during the manufacturing process, non-zero tolerances in the manufacturing process, the impact of the environment of the optoelectronic emitter 1 (e.g., possible effect of the encapsulation elements ( packaging ) covering the elementary emitters).

[0034] The phase shifters 5 are preferably connected to the control module 11 described later. Depending on the control signals sent by the control module 11, the phase shifters 5 can generate a predetermined relative phase Δϕ in the optical signals circulating in the different arms 4.

[0035] The optoelectronic transmitter 1 comprises N elementary emitters 6, or optical antennas, coupled to the arms 4, and therefore arranged downstream of the phase shifters 5. The relative phase Δϕ between the optical signals emitted by the elementary emitters 6 determines in particular the value of the angle Φ which forms the principal emission axis of the light beam in the far field with respect to the Y axis in the XY plane of the optoelectronic transmitter 1.

[0036] The elementary emitters 6 are diffraction gratings formed in waveguides of the arms 4. They extend parallel to each other along the Z-axis and are aligned along the X-axis; in other words, their first ends have the same position along the Z-axis as their second ends. They are spaced from each other by a distance preferably between λe / 2 and 2λe. For information purposes, the number N of elementary emitters can range from a few tens to tens of thousands.

[0037] Thus, the optical signals circulating in the arms 4 are partly transmitted into free space by diffraction by the elementary emitters 6. The extracted optical signals propagate in free space, recombine by interference, and thus form in the far field the light beam emitted by the optoelectronic emitter 1 whose angular distribution around the main emission axis is determined and defines the far-field emission pattern of the optoelectronic emitter 1.

[0038] The angle θ formed by the main emission axis with respect to the Y-axis in the YZ plane depends, in a known manner, on the emission wavelength λe of the laser source 2 and the period Λ of the diffraction grating formed by the elementary emitters 6. As previously stated, the angle Φ formed by the main emission axis with respect to the Y-axis in the XY plane depends on the value of the relative phase Δϕ applied by the phase shifters 5 to the optical signals circulating in the arms 4. It is generally desired that a relative phase Δϕ of zero results in a zero angle Φ, i.e., that the main emission axis of the far-field light beam is parallel to the Y-axis.

[0039] Furthermore, the elementary emitters here exhibit an extraction efficiency of less than 100%, resulting in the fact that the intensity I out of the optical signal (guided mode) at the output of the elementary emitters 6 along the Z-axis is non-zero. To achieve this, the length L ee of the elementary emitters is adjusted so that the intensity I out of the guided mode at the second end of the elementary emitters 6 has a predefined non-zero value. Alternatively or in addition, the non-zero intensity I out can be obtained by adjusting the etching depth of the teeth of the grating forming the elementary emitter, and / or by adjusting the grating fill factor ( filling factor (in English), that is, the ratio between the length of a tooth and the length of a pit, over one period (made up of a tooth and a pit) of the elementary emitter.

[0040] However, it may be necessary to control or correct the relative phase Δϕ value applied by the phase shifters 5. As mentioned previously, this may be the case for calibrating the optoelectronic emitter 1, for example, so that a zero relative phase Δϕ effectively results in a zero angle Φ of the principal emission axis, i.e., an orientation parallel to the Y-axis in the XY plane. This may also be the case to compensate for phase errors arising from the manufacturing process or the aging of certain components of the optoelectronic emitter. Finally, it may be the case to modify the value of the angle Φ in real time and with precision, so that the far-field emission pattern exhibits the desired characteristics.

[0041] For this purpose, the optoelectronic emitter 1 includes a control device consisting of a focusing interferometric lens 8, a plurality of photodetectors 10 and a control module 11. The control device also includes N input waveguides 7 which ensure the coupling between the elementary emitters 6 and the interferometric lens 8, as well as output waveguides 9 which ensure the coupling between the interferometric lens 8 and the photodetectors 10. The control device is said to be integrated insofar as the waveguides 7, 9, the interferometric lens 8 and the photodetectors 10 are made on or from the control chip.

[0042] The N input waveguides 7 connect the second end of the elementary emitters 6 to the input face 8.1 of the interferometric lens 8. They have identical optical paths from one input waveguide 7 to another, so as not to degrade the phase information (value of Δϕ) carried by the optical signals transmitted (undiffracted) by the elementary emitters 6.

[0043] As described in detail later, they each have at least two parts, namely: a first part 7.1 in which the input waveguides 7 extend parallel to each other from the elementary emitters 6, and a second part 7.2 in which the input waveguides 7 extend to the input face 8.1 of the interferometric lens 8.

[0044] In the first part 7.1, the input waveguides 7 can have the same length. In the second part 7.2, they are connected to the input face 8.1 orthogonally to it, that is to say that each input waveguide 7 is connected to an input port of the input face 8.1 orthogonally to a plane tangent to it.

[0045] However, since the entrance face 8.1 of the interferometric lens 8 is curved in the XZ plane, the lengths of the entrance waveguides 7 are therefore not identical from one entrance waveguide 7 to another. The length Le here corresponds to the distance between a first end of an entrance waveguide 7 connected to an elementary emitter 6 and a second end opposite the first and connected to the entrance face 8.1 of the interferometric lens 8.

[0046] The focusing interferometric lens 8 is based on Fourier optics, and thus provides on its output face 8.1 a Fourier transform of the optical field provided to the input face 8.1 by the input waveguides 7, and formed of the optical signals transmitted by the elementary emitters 6. The output face 8.3 forms the image plane of the interferometric lens 8, so that the diffraction pattern formed on the output face 8.3 is correlated to the far-field emission pattern of the light beam emitted by the optoelectronic emitter 1.

[0047] For this purpose, the interferometric lens 8 is defined by the entrance face 8.1, the exit face 8.3, and a free propagation region 8.2 (FPR, for Free Propagation Region,(in English) located between the two faces 8.1, 8.3. The inlet face 8.1 and the outlet face 8.3 are curved, the radius of curvature of each being located on the optical axis Δ of the interferometric lens 8 (here parallel to the Z-axis). The interferometric lens 8 is planar since it extends in the XZ plane of the photonic chip. In the FPR region 8.2, the optical field is confined along the Y-axis orthogonal to the XZ plane of the photonic chip, but it can propagate freely in the XZ plane. The FPR region 8.2 is defined by a medium with a homogeneous refractive index in the XZ plane, unlike the GRIN lens described in the article by Hulme et al. 2015 mentioned previously. In other words, the FPR 8.2 region has a refractive index that is identical at every point of this region 8.2, and preferably over its entire thickness which is then constant over the entire region 8.2.

[0048] Such an interferometric lens 8 is also called a star coupler ( star coupler (in English). It ensures the focusing at the output face 8.3 of the optical signals received at the input face 8.1. The optical signals propagate in the FPR region 8.2, interfere constructively, and converge towards a zone or point on the focal line defined by the output face 8.3 (image plane). The spatial distribution of the optical field intensity on the output face 8.3 is therefore correlated with the far-field emission pattern of the optoelectronic emitter 1. Thus, the characteristics of the far-field emission pattern are found on the output face 8.3, and from this, the angle Φ formed in the far field by the light beam of the optoelectronic emitter with respect to the Y-axis and in the XY plane can be deduced.

[0049] The control device comprises M photodetectors 10, coupled to the output face 8.3 by M output waveguides 9. The number M can be equal to or different from the number N. Together with the lens radius, it defines the resolution of the control device and the far-field observation angle of the light beam. However, the resolution of the control device is not limited by the size of the output face ports. Indeed, if the beam is located between two ports, its precise position can be extracted by calculating the ratio of the intensities of the photodiodes connected to the two adjacent ports. The output waveguides 9 thus extend from the output face ports 8.3 to the inputs of the photodetectors 10. The output ports are therefore located along the focal line of the interferometric lens 8.

[0050] The control module 11 is connected to the photodetectors 10 and, advantageously, to the phase shifters 5. The latter are controlled by the measurement signals transmitted by the photodetectors 10. The photodetectors 10 therefore receive optical signals from the ports of the output face 8.3, and transmit to the control module 11 measurement signals representative of the optical intensity of the optical field received on the output face 8.3. The control module 11 includes a calculator which, based on the measurement signals, can determine the effective phase shift value Δϕ between the optical signals circulating in the arms 4, as well as a phase shift value Δϕ to be applied to the optical signals via the phase shifters 5 in order to correct any phase errors.

[0051] It is then possible to calibrate the optoelectronic emitter 1, for example by applying a zero relative phase Δϕ and determining the angle θcr of the peak intensity of the received optical field on the output face 8.3 of the interferometric lens 8. The angle θcr is defined as the angle of inclination between the optical axis Δ of the interferometric lens 8 and a straight line passing through the peak intensity on the output face 8.3 and through an entrance point Pe of the interferometric lens 8. A compensation value for the relative phase Δϕ can thus be determined if it has been established that a zero relative phase Δϕ induces a non-zero angle θcr. It is also possible to control the optoelectronic emitter 1 in real time and with precision by inducing a predefined relative phase Δϕ and determining the induced angle θcr (and vice versa).This avoids the need for an external device to detect and measure the far-field light beam emitted by the optoelectronic emitter, such an external device comprising a matrix photodetector located opposite the elementary emitters along the Y-axis. Furthermore, the fact that the interferometric lens 8 is a lens with a homogeneous refractive index FPR 8.2 region prevents manufacturing uncertainties of this lens from significantly impacting the performance of the optoelectronic emitter 1.

[0052] In this respect, the figures 3A to 3C are schematic and partial top views of the interferometric lens 8 and the photodetectors 10. In this case, a relative phase Δϕ of zero corresponds to a zero angle Φ of the principal emission axis, the peak intensity of the optical field on the output face being aligned with the optical axis Δ of the interferometric lens 8 ( fig.3A). On the other hand, when the relative phase Δϕ has a non-zero value Δϕ 1, and induces a non-zero angle Φ 1 of the main emission axis, the intensity peak on the output face forms a non-zero angle θ cr,1 ( fig.3B ). Finally, when the relative phase Δϕ has a non-zero value -Δϕ 1, and induces a non-zero angle -Φ 1 of the main emission axis, the intensity peak on the output face forms a non-zero angle -θ cr,1 ( fig.3C ).

[0053] The interferometric lens 8 can have a Rowland mount or a confocal configuration. These two variants differ essentially in the value of the radius of curvature of the circle defining the exit face 8.3. As detailed later, it can be a star coupler, or a ladder grating, which is a special case of the star coupler.

[0054] There figure 4Ais a schematic and partial top view of a portion of the optoelectronic emitter according to an embodiment in which the focusing interferometric lens has a Rowland configuration. Figures 4B and 4C are schematic and partial top views, respectively, of the Ze zone of the entrance face 8.1, and of the Zs zone of the exit face 8.3 of the interferometric lens 8.

[0055] In Rowland setup, the entrance face 8.1 is curved with a radius of curvature denoted R cr, and the exit face 8.3 is curved with a radius of curvature R cr / 2. The dimensioning of the interferometric lens 8 depends in particular on the desired detection resolution of the optical field focused on the exit face 8.3, and on the maximum angular detection aperture defined by the angle 2×θ cr,max around the optical axis Δ.

[0056] More precisely, the optical axis Δ of the interferometric lens 8 extends here along the Z-axis between the entrance point Pe of the entrance face 8.1, and an exit point Ps of the exit face Ps. The center of the Rowland circle of radius R cr is located at the exit point Ps, and an arc of the Rowland circle of radius R cr defines the entrance face 8.1. An arc of a circle of radius R cr / 2 defines the exit face 8.3, whose center is located on the optical axis Δ.

[0057] Thus, the angle Φ of the far-field light beam depends on the relative phase Δϕ according to the following relationship: Δϕ = p ee / λ e × sinΦ, where p ee is the lateral spacing between the elementary emitters along the X-axis (generally between λ e / 2 and 2λ e). For illustration, the emission wavelength λ e can be equal to 0.905 µm, and the lateral spacing p ee can be approximately 1.6 µm.

[0058] The deviation angle θcr formed by the peak intensity of the optical field focused on the output face 8.3 depends on the far-field angle Φ by the following relation: sinθcr = p ee / (λ e ×p cr,e ×β FPR )×sinΦ, where p cr,e is the lateral pitch between the ports of the input face 8.1, and β FPR is the propagation constant in the FPR region 8.2. This relation can be deduced from the article by Smit & van Dam entitled PHASAR-Based WDM-Devices: Principles, Design and Applications, IEEE J. Sel. Top. Quantum Electron, 2 (2), 236-250 (1996). For illustration, the lateral pitch p cr,e can be approximately 1.2 µm (here for N=16), and the constant β FPR can be approximately 12.31 µm⁻¹. The propagation constant β FPR is equal to 2π × n eff,FPR / λ e, where n eff,FPR is the effective index of the optical mode in the FPR region. It can be determined by a finite element simulation.

[0059] In general, the effective refractive index (neff) associated with an optical mode supported by a waveguide is defined as the product of the propagation constant β and λ / 2π. The propagation constant β depends on the wavelength λ of the optical mode, as well as the waveguide properties (refractive index and transverse dimensions). The effective refractive index of the optical mode corresponds, in a certain way, to the refractive index of the waveguide as seen by the optical mode. It is usually between the core refractive index and the cladding refractive index of the waveguide.

[0060] We can then define a lateral output dimension dcr,s along which the output ports extend along the X-axis with a lateral pitch pcr,s. This dimension depends on the maximum angle Φmax that we wish to detect, as well as the radius Rcr. It is defined by the following relationship: dcr,s = (2 × Rcr × pee) / (λe × pcr,e × βFPR) × sin Φmax. Thus, for example, for a radius Rcr of approximately 310 µm and a maximum angle Φmax of approximately 30°, we obtain a lateral output dimension dcr,s of approximately 37 µm. Here, we can have 25 output ports arranged with a lateral pitch pcr,s of approximately 1.5 µm.

[0061] Therefore, the choice of the radius R cr depends on the desired resolution (value of p cr,s) and the desired maximum angular aperture 2 × θ cr,max of detection. This avoids having to make a gradient-index lens ( GRIN lens) as in the example of the prior art mentioned above, the obtaining of which can be technically complex and whose manufacturing tolerance can lead to a degradation of the performance of the optoelectronic emitter 1. On the contrary, the fact that the interferometric lens 8 is formed of two curved faces 8.1, 8.3 delimiting the FPR region 8.2 of homogeneous refractive index makes it possible to simplify the manufacturing process while limiting the impact of manufacturing tolerances on the performance of the optoelectronic emitter 1.

[0062] However, the inlet waveguides 7 are connected to the inlet face 8.1 in a manner locally orthogonal to it; that is, each inlet waveguide 7 opens onto the FPR region 8.2 at the inlet face 8.1, oriented towards the exit point Ps of the interferometric lens 8 (center of the circle of radius R cr). It follows that the inlet waveguides 7 do not have the same physical length Le from one waveguide 7 to another. We denote by Le(i) the physical length of an inlet waveguide 7 of rank i, with i ranging from 1 to N, ranks 1 and N being located at the edge of the array of inlet waveguides 7.

[0063] The input waveguides 7 then have a correction portion Pc of the effective index of the guided mode so that the optical paths are identical from one input waveguide to the other. In general, the optical path length of a waveguide is equal to the product of its (physical) length and the effective index of the guided mode. This correction portion Pc is preferably located in the first part 7.1 of the input waveguides 7. We denote by neff,ec(i) the effective index of the input waveguide 7 within the correction portion Pc, and by neff,e0(i) the effective index of the input waveguide 7 outside the correction portion Pc. We thus also obtain the optical path C (i) =n eff,ec(i) ×L ec +n eff,e0(i) ×(L e -L ec ) which we want to be constant from one input waveguide 7 to the other.

[0064] Several possibilities can be implemented to make the optical paths C(i) of the input waveguides 7 identical. A first possibility (illustrated on the fig.5 ) consists of modifying the lateral dimensions of the input waveguides over a length L ec(i) in the correction portion Pc, so as to correct the effective index of the guided mode in each of the input waveguides. A second possibility (illustrated in the fig.6 ) consists of performing modal coupling between each input waveguide 7 and an additional waveguide 12 located opposite it along the Y-axis and having a different refractive index than the input waveguide 7, so as to obtain a supermode locally and over a given length. A third possibility (illustrated on the fig.7) consists of actively modulating the effective index of the guided mode in each input waveguide 7, by means of additional thermo-optical phase shifters (heaters) 14.

[0065] There figure 5 is a schematic and partial top view of several input waveguides 7 at the correction portion Pc.

[0066] Apart from the correction portion Pc, the inlet waveguides 7 have the same transverse dimensions in thickness ee and width We0, over a length L e(i) - L ec(i), where L e is the total length of the i-th inlet waveguide between the downstream end of the elementary emitter 6 and the inlet face 8.1, and where L ec(i) is the length of the correction portion Pc for this same inlet waveguide. It was previously noted that the lengths L e(i) differ from one inlet waveguide 7 to another because the inlet face 8.1 is curved and these inlet waveguides 7 open onto it orthogonally. Furthermore, the length L ec(i) may be the same or different from one inlet waveguide to another.

[0067] In the correction portion Pc, the input waveguides 7 have a maximum transverse dimension, here the width We ec(i) , different from one input waveguide 7 to another, so that the optical paths C (i) of the input waveguides 7 are identical over their entire length.

[0068] More precisely, the input waveguides 7 exhibit a gradual, here linear, variation in width W, from We0 to Wec(i) and vice versa, over a length Lec(i). Thus, the first-order input waveguide exhibits a linear widening of its width from We0 to Wec(1) over a length Lec(1) / 2, followed by a linear narrowing from Wec(1) to We0 over a length Lec(1) / 2. The length Lec(i) / 2 is chosen so that the width variation is adiabatic, that is, without optical losses. It can be at least approximately 25 µm. The maximum widths W ec(i) are different from one input waveguide 7 to another, so as to induce a variation of the effective index of the guided mode for each input waveguide 7, this longitudinal variation of the effective index being determined so that the optical paths C (i) of the input waveguides 7 are identical to each other over their entire length.In other words, we adjust the longitudinal variation of width W so that: . n eff W e 0 × L e i − L ec i + ∫ 0 L ec i n eff W ec i dL = A where A is a non-zero constant that is the same for all input waveguides.

[0069] As an example, the input waveguide 7 can be formed from a silicon nitride core with a refractive index of 2 for an emission wavelength λe of 905 nm, and a constant thickness of 300 nm and a width We0 of 600 nm. The core is surrounded by a cladding of silicon oxide with a refractive index of 1.45. In this case, the effective index neff,e0 for the fundamental guided mode with polarization TE00 is 1.72. Conversely, for a width Wec(i) of 1 µm, the effective index neff,ec(i) is 1.78.

[0070] Also, the longitudinal variation of the local width W e of the input waveguides 7 in the correction portion Pc induces a variation of the effective index n eff,ec(i) of the guided mode, thus making it possible to equalize the optical paths C (i) of the input waveguides 7. It limits or eliminates any degradation of the phase information associated with the guided modes, so that the angular intensity distribution of the optical field focused on the output face 8.3 of the interferometric lens 8 is effectively correlated with the far-field emission pattern of the optoelectronic emitter 1.

[0071] This is particularly advantageous since the lateral pitch of the input waveguides is generally small, for example here on the order of 1.5µm, and depends in particular on the lateral pitch p ee between the elementary emitters 6 (between λ e / 2 and 2λ e ). A modification of the length L e(i) to correct the optical path difference of the input waveguides 7 is therefore insufficient.

[0072] There figure 6 is a schematic and partial cross-sectional view of several input waveguides 7 at the correction portion Pc according to an alternative embodiment of the optoelectronic emitter 1. The input waveguides 7 are shown in exploded view.

[0073] In this embodiment, additional waveguides 12 are arranged with respect to the inlet waveguides 7 along the Y-axis so as to allow modal coupling between the two waveguides 7 and 12. The additional waveguides 12 have a refractive index different from that of the inlet waveguides 7, and preferably higher. Preferably, the inlet waveguides 7 have transverse dimensions (thickness and width) that remain constant along their entire length Le(i). Also preferably, the upper waveguides 12 have transverse dimensions (thickness and width) that remain constant along their entire length. A sheath 13, for example made of silicon oxide, surrounds the waveguides 7 and 12.

[0074] Modal coupling refers to the fact that the guided mode circulating in the input waveguide 7 extends spatially into both the input waveguide 7 and the upper waveguide 12, thus forming a supermode. More precisely, the electric field component of the supermode has a spatial distribution that covers both the input waveguide 7 and the upper waveguide 12. Modal coupling is ensured by a suitable vertical spacing of the upper waveguides 12 relative to the input waveguides 7, for example, a few tens to hundreds of nanometers.

[0075] Thus, due to the modal coupling between the two waveguides 7 and 12, the guided mode exhibits an effective index that differs in the correction portion Pc (due to the modal coupling) from that outside the correction portion. The difference in length Les(i) between the upper waveguides 12 in the correction portion Pc induces a variation in the effective index of the guided mode for each of the inlet waveguides 7. Thus, as illustrated by the fig.6The length Les(1) of the waveguide 12 associated with the first-order input waveguide 7 has a value less than that of Les(2), which is itself less than Les(3). Thus, the guided mode circulating in the third-order waveguide 7 has an optical path length equal to that of the second-order waveguide 7 and that of the first-order waveguide 7. Therefore, the length Les(i) of the waveguides 12 is predefined for each waveguide 7 so that the optical path lengths of the waveguides 7 are identical from one waveguide 7 to another along their entire length.

[0076] There figure 7 This is a schematic and partial cross-sectional view of several input waveguides 7 at the correction portion Pc according to another embodiment of the optoelectronic emitter. The input waveguides 7 are shown in exploded view.

[0077] Here, the variation of the effective index of the guided mode for each of the input waveguides 7 is actively ensured by means of several additional phase shifters 14, preferably thermo-optical to limit optical losses, each arranged opposite each other along the Y axis of an input waveguide 7 in the correction portion Pc.

[0078] The temperature value applied by each additional phase shifter 14 is predetermined as a function of the effective index variation to be obtained in each of the input waveguides 7, so as to equalize the optical paths C (i) of the input waveguides 7. Thus, any degradation of the phase information carried by the guided modes in the input waveguides 7 is limited or avoided.

[0079] Specific embodiments have just been described. Various variations and modifications are possible without departing from the scope of the invention. In particular, the different variations described above are not mutually exclusive but can be combined.

[0080] The interferometric lens 8 may have other types of mounting, for example a confocal mounting. In this case, the interferometric lens 8 differs from the one described in the fig.2 essentially in that the exit face 8.3 is located opposite the entrance face 8.1 and is defined by the same Rowland circle of radius R cr.

[0081] There figure 8This is a schematic, partial top view of an optoelectronic emitter 1 according to another embodiment, in which the interferometric lens 8 is a ladder grating. The input face 8.1 and the output face 8.3 are regions of the same curved face defined by the Rowland circle of radius R cr. A reflecting face 8.4, opposite the input 8.1 and output 8.3 regions, delimits, together with them, the homogeneous refractive index FPR region 8.2. The reflecting face 8.4 comprises a grating of elementary reflectors 8.5 reflecting, in the direction of the focal line defining the output face 8.3, the optical signals transmitted by the input waveguides 7.

[0082] THE Figures 9A and 9BThese are schematic and partial views of one of the input waveguides 7 at the correction portion Pc according to two other embodiments of the optoelectronic emitter. Here, the variation of the effective index of the guided mode for each of the input waveguides 7 is achieved by means of a phase-change material that contributes to forming the cladding of the waveguide 7. The phase-change material can be chosen, in particular, from among the chalcogenides, especially of the GST type, i.e., formed from germanium Ge, antimony Sb, and tellurium Te. See the paper by Abdollahramezani et al. entitled "Tunable nanophotonics enabled by chalcogenide phase-change materials," Nanophotonics 2020, 9(5), 1189-1241.The portions 15 of phase-change materials apply a relative phase shift, statically or dynamically, between the modes guided in the input waveguides 7 predefined for each of them so that the optical paths of the input waveguides 7 are identical to each other.

[0083] There fig.9AThis illustration, in top and cross-sectional views, presents a first example in which the phase-change material has a predefined crystalline phase during the fabrication process of the optoelectronic emitter. Here, the waveguide 7 rests on a substrate 13 that contributes to forming the cladding. A portion 15 of thin-film phase-change material has been deposited on and around the waveguide 7, extending along it for a length, for example, between 1 µm and 100 µm. It has a uniform thickness, for example, between 5 nm and 100 nm. The crystalline phase of the phase-change material was defined during the fabrication process, for example, by means of laser pulses. The effective index of the guided mode therefore depends on the refractive index of the phase-change material (which depends on its crystalline phase).

[0084] There fig.9BThis illustrates, in top and cross-sectional views, a second example in which the phase-change material has a crystalline phase that can be dynamically (or statically) modified during the operation of the optoelectronic emitter. Here, the waveguide 7 is embedded in the substrate 13, which forms part of the sheath. A heater 16 is positioned on the surface of the substrate 13, above the waveguide 7, and extends on either side of it along the X-axis. A portion 15 of the phase-change material extends over the heater 16 and above the waveguide 7, for a length, for example, between 1 µm and 100 µm. Here, it has a uniform thickness, for example, between 5 nm and 100 nm. Two electrodes 17 are in contact with the heater 16.Thus, applying an electrical voltage between the electrodes causes the heater 16 to heat up, which induces a change in the crystalline phase of the phase-change material. The effective index of the guided mode is then modified accordingly.

Claims

1. An optoelectronic transmitter (1) with a phased array antenna on a photonic chip, comprising: ∘ a power splitter (3), intended to be coupled at its input to a laser source (2), and coupled at its output to N waveguides (4) forming the arms of the optoelectronic transmitter (1), with N>1; ∘ an array of phase shifters (5) and elementary emitters (6), arranged in the arms (4), the elementary emitters (6) being laterally aligned and extending parallel to each other; ∘ an integrated control device comprising: • input waveguides (7), coupled to the elementary emitters (6); • a focusing lens (8), comprising an input face (8.1) coupled to the input waveguides (7), and an output face (8.3); • output waveguides (9), coupled to the output face (8.3); • photodiodes (10), coupled to the output waveguides (9); • a control module (11), adapted to control the phase shifters (5) based on signals received by the photodiodes (10); ∘ characterised in that: • the focusing lens (8) is an interferometric lens whose input (8.1) and output (8.3) faces are curved and delimit a free propagation region (8.2) with a homogeneous refractive index, the output face (8.3) extending along a focal line of the focusing lens (8); • the input waveguides (7) being connected to the input face (8.1) orthogonally thereto; • an effective index of the guided modes, in a portion called correction portion (Pc) of the input waveguides (7), being adapted such that the optical paths (C(i)) of the input waveguides (7) are identical to each other.

2. The optoelectronic transmitter (1) according to claim 1, wherein the input waveguides (7) each have, in the correction portion (Pc), a variation of at least one transverse dimension, said variation of transverse dimension being predefined for each input waveguide (7) such that the optical paths (C(i)) of the input waveguides (7) are identical to each other.

3. The optoelectronic transmitter (1) according to claim 2, wherein each input waveguide (7) has, in the correction portion (Pc), an adiabatic variation of the width, ranging from an identical reference value We0 for each input waveguide (7) to a maximum value Wec(i) different from one input waveguide (7) to another.

4. The optoelectronic transmitter (1) according to any one of claims 1 to 3, comprising waveguides called additional waveguides (12) made in a refractive index different from that of the input waveguides (7), extending only in the correction portion (Pc), and arranged opposite and parallel to the input waveguides (7) so as to enable a modal coupling with the input waveguides (7), the additional waveguides (12) having a predefined length for each of them such that the optical paths (C(i)) of the input waveguides (7) are identical to each other.

5. The optoelectronic transmitter (1) according to any one of claims 1 to 4, wherein additional phase shifters (14) are coupled to said input waveguides (7) in the correction portion (Pc), and apply a relative phase shift between the guided modes in the input waveguides (7) which is predefined for each of them such that the optical paths (C(i)) of the input waveguides (7) are identical to each other.

6. The optoelectronic transmitter (1) according to any one of claims 1 to 5, wherein a portion (15) of a phase-change material, located in the correction portion (Pc), is associated with each input waveguide (7) and contributes to forming a cladding for the latter, the portions (15) of phase-change materials applying a relative phase shift, statically or dynamically, between the guided modes in the input waveguides (7) which is predefined for each of them such that the optical paths (C(i)) of the input waveguides (7) are identical to each other.

7. The optoelectronic transmitter (1) according to any one of claims 1 to 6, wherein the focusing lens (8) is a star coupler or an echelle grating.

8. The optoelectronic transmitter (1) according to any one of claims 1 to 7, wherein the focusing lens (8) has a Rowland or a confocal configuration.

9. The optoelectronic transmitter (1) according to any one of claims 1 to 8, wherein the photonic chip is produced from an SOI substrate.

10. The optoelectronic transmitter (1) according to any one of claims 1 to 9, comprising the laser source (2).