FR3020188A1 - LASER SOURCE - Google Patents

Abstract

The invention relates to a laser source (10) comprising: - a first waveguide (12) having a first extension, called the first length (L1), along the longitudinal direction (Z), - a second guide wave (22) distinct from the first waveguide (12) and extending parallel to the first waveguide (12), the second waveguide (22) having a portion (30) superimposed with the first waveguide (12) wave (12) along a stacking direction (Y) perpendicular to the longitudinal direction (Z), the portion (30) having a second extent along the longitudinal direction (Z) said length of superposition (L superposition) ), the superposition length (the superposition) being greater than 90% of the first length (L1).

Description

The present invention relates to a laser source. The invention lies in the field of the heterogeneous integration of laser sources on an integrated optical structure. Such integrations find applications in optical telecommunications using the components of integrated photonics. Integrated photonics has been developed on substrates such as glass, silicon or III-V materials. It is recalled that a "III-V" type semiconductor is a composite semiconductor manufactured from one or more elements of column III of the periodic table of the elements (boron, aluminum, gallium, indium, etc.). and one or more column V or pnictogene elements (nitrogen, phosphorus, arsenic, antimony ...). Photonic integration on silicon has many advantages such as the ability to integrate a large number of elements on the same chip, and the possibility of integrating both optical components and electronic components.

However, there is still no effective monolithic laser source on silicon. To overcome this problem, heterogeneous laser source integrations using 111-V materials on a silicon substrate have been developed. For this, it is known from the article by Hyundai Park et al. entitled "Device and Integration Technology for Silicon Photonics Transmitters" from the IEEE Journal of Quantum Electronics, Volume 17, Issue 3, May / June 2011 and the article by M. Lamponi et al. entitled "Low-threshold heteregeneously integrated InP / S01 laser with a double adiabatic type coupler" from the journal IEEE Photonics Technology Letters, volume 24, pages 76 to 78 published in 2012 laser sources of this type.

However, the energy yields, defined as the ratio between the transmitted optical power produced and the electrical power consumed, of these structures are insufficient. There is therefore a need for a laser source having a better performance. For this purpose, the subject of the invention is a laser source comprising a first waveguide extending in a longitudinal direction, the first waveguide forming a laser waveguide having a first extent, called the first length, along the longitudinal direction, and a second waveguide distinct from the first waveguide and extending parallel to the first waveguide, the second waveguide having a portion superimposed with the first waveguide along a stacking direction perpendicular to the longitudinal direction, the portion having a second extent along the longitudinal direction called superposition length, the superposition length being greater than 90% of the first length, preferably greater than 95% of the first length and preferably equal to the first length. According to particular embodiments, the laser source comprises one or more of the following characteristics, taken in isolation or in any technically possible combination: the first waveguide has a first extent, called the first width, along a transverse direction and the portion of the second waveguide has a second extent, called second width, along the transverse direction, the transverse direction being perpendicular to the stacking direction and the longitudinal direction. the second width is substantially constant along the longitudinal direction. the first waveguide comprises a layer made of p-doped semiconductor, the second width being such that the confinement factor of the light wave in the p-doped semiconductor layer is less than 5%, preferably less than 2% and / or the confinement factor of the light wave in the layer comprising multiple quantum wells is greater than 1%, preferably greater than 2%. the second width is smaller than the first width. the second width is greater than or equal to 25% of the first width and / or less than or equal to 40% of the first width. the first waveguide is made from at least one material belonging to column III of the periodic table composed of a material according to column V of the periodic table. the second waveguide is made from silicon. the second waveguide comprises a network. the laser source further comprises an electro-absorption modulator. the laser source comprises a wavelength-selective and light-reflective element selected from the group consisting of two ring resonators, an assembly formed by a ring resonator and a Mach-Zehnder interferometer and a sampled array. Other features and advantages of the invention will appear on reading the following description of embodiments of the invention, given by way of example only and with reference to the drawings which are: FIG. 1, a diagrammatic view from above an example of a laser source; FIG. 2, a schematic view of a section of the laser source along the line II-II shown in FIG. 1; FIG. 3, a schematic view from above of another example of a laser source; FIG. 4, a schematic view from above of another example of a laser source; FIG. 5, a schematic view from above of another example of a laser source, and FIGS. 6 to 8, representations of the result of simulations of examples of a laser source according to FIG. 1 in operation. For the remainder of the description, a longitudinal direction corresponding to a general direction of propagation of light is defined. It is also defined a stacking direction perpendicular to the longitudinal direction and contained in the plane of Figure 1 and a transverse direction perpendicular to the longitudinal direction and the stacking direction. The longitudinal direction, the stacking direction and the transverse direction are respectively symbolized by Z, Y and X axes in FIG.

In the context of the invention, there is provided a laser source 10. A laser source 10 (acronym for "light amplification by stimulated emission of radiation" which translates into French as "amplification of light by stimulated emission" is a device that produces spatially and temporally coherent light based on the laser effect.

Each laser source 10 is a stack of different layers also called "laminate" in the field of microelectronics. The laser source 10 comprises a first waveguide 12 extending in the longitudinal direction Z. The first waveguide 12 forms a laser waveguide.

According to the example illustrated in FIG. 1, the first waveguide 12 is monomode. By definition, a monomode laser waveguide is a waveguide capable of propagating a single laser mode, this laser mode resulting from a laser effect produced in the waveguide. In a variant, the first waveguide 12 is multimode.

Preferably, the first waveguide 12 is made from at least one material belonging to column III of the periodic table composed of a material according to column V of the periodic table. Such materials are indeed effective in producing the laser effect. By way of example, the first waveguide 12 shown in FIG. 1 comprises successively four layers: a first p-doped InGaAs layer 14, a second p-doped InP layer 16, a third quantum multiwell layer 18 and a fourth layer 20 in n-doped InP. The first InGaAs layer 14 is p-doped for electrical contact. Similarly, the fourth InP layer 20 is n-doped for electrical contact.

The third layer 18 of quantum multiwell is dedicated to the laser emission. The third layer 18 also comprises several sub-layers called "wells" and "barriers". Other structures for the first waveguide 12 are conceivable. In particular, a three-layer structure with a p-doped semiconductor layer, a quantum multiwell layer and an n doped semiconductor layer can be envisaged. The first waveguide 12 has a first extension, called the first length L1, along the longitudinal direction Z. In the particular case of FIG. 1, the first waveguide 12 has two end portions 12t and one central portion 12p located between the two end portions 12t. In this case, the first length L1 corresponds to the extent of the first waveguide 12 along the longitudinal direction Z located between the two end portions 12t. The laser source 10 also comprises a second waveguide 22 distinct from the first waveguide 12. The second waveguide 22 extends parallel to the first waveguide 12. Preferably, the second waveguide 22 22 is made from silicon. For example, the second waveguide 22 comprises three layers: a fifth layer 24 providing the connection between the second waveguide 22 and the first waveguide 12, a sixth silicon layer 26, a seventh layer 28 and an eighth layer 29 forming a substrate. Preferably, the seventh layer 28 is made of silica. Advantageously, the eighth layer 29 is made of silicon. The second waveguide 22 has a portion 30 superimposed with the first waveguide 12 along the stacking direction Y. This portion 30 is part, as in the particular case of Figure 1, of the sixth layer 26 The second waveguide 22 has an extent along the longitudinal direction Z, said second length L2. The second length L2 is strictly greater than the first length L1. This is mathematically written as: L2> L1 The portion 30 has a second extent along the longitudinal direction Z called the superposition length 1 -perp'iti '. By defining the portion 30, the superposition length The superposition is less than or equal to the first length. This is written mathematically as: L superspos. <Ll The superposition length The superposition is greater than or equal to 90% of the first length L1. This is written mathematically as: 90 T, - L superposition - - 100 L This allows a better coupling between the second waveguide 22 and the first waveguide 12. Preferably, to increase this effect, the length superposition The superposition is greater than or equal to 95% of the first length L1. This is written mathematically as: Preferably, to increase this effect even more, the superposition length The superposition is equal to the first length L1. FIG. 1 corresponds to the case where the superposition length The superposition is equal to the first length L1. The portion 30 of the second waveguide 22 has a second extent, called the second width d2, along the transverse direction X. Preferably, the portion 30 of the second waveguide 22 has a second substantially constant width d2 along the longitudinal direction Z. The term "substantially constant" means that the difference between the lowest value and the largest value for the second width is less than 1% of the second width d2. Advantageously, the portion 30 of the second waveguide 22 has a second substantially constant width d2 along the longitudinal direction Z only in the portion of the portion 30 superimposed with the central portion 12p of the first waveguide 12.

According to a preferred embodiment, the portion 30 of the second waveguide 22 has a constant second width d2 along the longitudinal direction Z. When the first waveguide 12 has a layer made of p-doped semiconductor, the second width d2 is such that the confinement factor of the light wave in the p-doped p-layer 16 is less than or equal to 5%, preferably less than or equal to 2%. For the laser source of FIG. 1, since the dimensions of the layers considered along the stacking direction Y are much smaller than the optical wavelength of the photons, it follows that the photons occupy an area which extends beyond the first waveguide 12. The confinement factor F makes it possible to measure the proportion of photons actually in the layer in question. It is defined by: y) 12dxdy F _ considered layer +00 r r + co 1E (x, y) 12dxdy -00 In this case, the p-doped semiconductor layer is the second InP layer 16. Indeed, the second p-doped layer 16 introduces significant optical losses. The reduction of the confinement factor in the second layer 16 makes it possible to reduce the losses of the optical mode during its propagation in the first waveguide 12.

Alternatively, the layer made in p-doped semiconductor considered is the first layer 14 in InGaAs. In the particular case of Figure 1, the first layer 14 of InGaAs is relatively far from the third layer 18 so that the confinement factor F is zero.

When the first waveguide 12 has a layer comprising multiple quantum wells, the second width d2 is such that the confinement factor of the light wave in the layer comprising multiple quantum wells is greater than 1%, preferably greater than 2. %. In this case, the layer comprising multiple quantum wells is the third layer 18.

Preferably, the second width d2 is not arbitrary with respect to the dimensions of the first waveguide 12. Notably, the first waveguide 12 has a first extent, called the first width d1, along the transverse direction X In the particular case illustrated in FIG. 1, the first width d1 is defined by the extent of the central portion 12p of the first waveguide 12 along the transverse direction X.

More specifically, the end portions 12t have a tapered shape, the extent along the transverse direction X decreasing as the end portion 12t is traveled away from the central portion 12p. Each end portion 12t is thus a tapered shape adaptation portion, often referred to as "taper portion". As a variant, the end portions 12t have a trapezoidal shape. According to a preferred embodiment, the second width d2 of the portion 30 is modified along the longitudinal direction Z when the portion 30 is superimposed with an end portion 12t. Typically, the second width d2 is decreased to correspond to the extent of the end portion 12t along the longitudinal direction Z. This promotes a good transfer of the guided power to the second waveguide 22. Advantageously after the portion 30 superimposed with the end portion 12t, the width of the second waveguide 22 increases again, so as to guarantee a monomode character at the second waveguide 22. In the example of FIG. , the central portion 12p has a first extent, along the transverse direction X constant. In the various possible cases, preferably, the first width is greater than or equal to the second width d2. This is written mathematically: d2 d1 Preferably, the second width d2 is greater than or equal to 25% of the first width d1. This is written mathematically: d1 d2 100 Advantageously, the second width d2 is less than or equal to 40% of the first width d1. This is written mathematically: d2 <40 dl 100 In the particular case of FIG. 1, the laser source 10 also has an anode 32 and two cathodes 34 and 36. The anode 32 is in electrical contact with the first layer 14. The two cathodes 34, 36 are in electrical contact with the fourth layer 20. The laser source 10 proposed with reference to Figures 1 and 2 has many advantages. The laser source 10 allows in particular to consider multiple applications because of its ease of implementation. By way of illustration, three particular applications with reference to FIGS. 3, 4 and 5 will be more particularly detailed in the following. According to the example of FIG. 3, the second waveguide 22 comprises a grating 50 so that the laser source 10 is a distributed feedback laser. The network 50 is superimposed on the central portion 12p and on one of the two end portions 12t of the first waveguide 22. A distributed feedback laser is often referred to as a DFB laser, DFB referring to the acronym "distributed feedback". ". In this case, for illustrative purposes, the network 50 comprises a periodic structuring. Such a network 50 is a diffractive network. In this case for the case of Figure 3, the pattern is an alternation of areas along the longitudinal direction Z with different optical indices. Such a grating 50 is easily etched on the second waveguide 22. In addition, according to a preferred embodiment, the dimensions of the second waveguide 22 are chosen so as to optimize the efficiency of the grating 50.

In addition, according to another embodiment, the network 50 is superimposed only with the central portion 12p of the first waveguide 12. According to the example of FIG. 4, the laser source 10 is similar to that of FIG. Only the differences with respect to the laser source 10 shown in Figure 3 are highlighted.

The laser source 10 further has an electro-absorption modulator 60 formed from the second waveguide 22. An electro-absorption modulator is often referred to by the acronym EAM which refers to the English terminology "electro-absorption modulator ". Such a modulator is a semiconductor device capable of modulating the intensity of a laser beam by means of an electrical voltage. This modulation is relatively effective. Its operating principle is based, for example, on the Confined Quantum Starck effect (QCSE), ie a change in the absorption spectrum of the modulator caused by an applied electric field. The presence of the electro-absorption modulator 60 therefore makes it possible to have a modulated laser emitter 10. According to a preferred embodiment, the electro-absorption modulator 60 and the second waveguide 22 are made from the same layers obtained by epitaxial growth. This makes it possible to glue vignettes or plates made of a "III-V" type material of one and the same epitaxial structure for the second waveguide 22 and the electro-absorption modulator 60. Preferably, the the extent of the second waveguide 22 along the transverse direction X at the electroabsorption modulator 60 is less than the second width d2 of the portion 30 below the central portion of the laser of the second waveguide 22. This makes it possible to increase the modulation efficiency of the electroabsorption modulator 60 since a good modulation efficiency by the electroabsorption modulator 60 assumes a strong confinement factor in the third layer 18 of quantum multiwells.

Advantageously, the extent of the second waveguide 22 along the transverse direction X at the level of the electro-absorption modulator 60 is between 5% of the second width d2 and 60% of the second width d2. Preferably, the extent of the second waveguide 22 along the transverse direction X at the level of the electroabsorption modulator 60 is equal to half of the second width d2. For example, the extent of the second waveguide 22 along the second transverse direction X at the level of the electroabsorption modulator 60 is equal to 0.4 μm while the second width d2 of the portion 30 of the second waveguide According to another variant illustrated in FIG. 5, the laser source 10 is similar to that of FIG. 4. Only the differences with respect to the laser source 10 shown in FIG. evidence. The laser source 10 further comprises a wavelength selective element 70 and reflecting light. In this case, the wavelength-selective and light-reflective element 70 comprises two ring resonators. Alternatively, the wavelength-selective, light-reflective element 70 is an assembly formed by a ring resonator and a Mach-Zehnder interferometer or sampled array. Such a laser source 10 has good tunability. The wavelength tunability corresponds to the ability to modify the wavelength emitted from a laser source. This capacitance is quantified by a wavelength tunability range defined as the wavelength range on which the laser source 10 is capable of emitting laser emission. The modification of the wavelength is obtained for example under the effect of the injection of a current or under the effect of a rise in temperature. Tunability makes it possible, in particular, to introduce flexibility in the implementation of wavelength multiplexing or to produce coherent receivers. To make it possible to improve the tunability, it is favorable for the extension of the second waveguide 22 in the transverse direction X to increase towards the electro-absorption modulator 60. Indeed, the greater the extent, the better the mode. generated in the first waveguide 12 is coupled in the second waveguide 22. It then becomes possible to realize a reflector having a wide bandwidth by the etching of a Bragg grating in the transition zone between the second waveguide 22 and the electro-absorption modulator 60. In addition, it is preferable that the portion 30 of the second waveguide 22 has an enlarged portion along the second transverse direction X. Such an enlarged portion schematically shown in Figure 5 shows a second extent along the second transverse direction X greater than the second width d2. The network 50 is included in the enlarged part. In this case the second width d2 is, for example, defined as the extension along the second transverse direction X of the portion 30 of the second waveguide 22 outside the enlarged portion. In addition, the laser source 10 proposed with reference to FIGS. 1 and 2 has a better coupling between the first waveguide 12 and the second waveguide 22. This is shown by means of the simulations illustrated in FIGS. 8. These figures show an example of intensity profile calculated for the laser mode propagating in the laser source 10 for three different values of second distance d2. FIG. 6 was calculated for a first distance d1 of 2 μm and a second distance value d2 of 0.5 μm; FIG. 7 for a first distance d1 of 2 μm and a second distance value d2 of 0.7 μm and FIG. 8 for a first distance d1 of 2 μm and a second distance value d2 of 0.8 μm. In the case of FIG. 6, the laser mode is confined in the first waveguide 12. In the case of FIG. 7, the laser mode propagates both in the first waveguide 12 and in the second waveguide 12. In the case of FIG. 8, the laser mode is mainly guided in the second waveguide 22.

These simulations show that the laser source 10 simulated in FIG. 8 with the second waveguide 22 under the first waveguide 12 allows a better coupling between the two waveguides 12, 22. Indeed, the laser mode FIG. 7 is simulated between the two waveguides 12, 22. A rather short transition zone is sufficient to couple the simulated laser mode in FIG. 7 to the mode confined in the second waveguide 22. Similarly, these simulations also show that in the case of FIG. 7 (hybrid mode with a laser mode in the first waveguide 12 and in the second waveguide 22), a decrease in losses due to a decrease in the factor confinement in the second layer 16 is observed. Thus, a gain of a factor of two in power is obtained.

Thus, the laser sources 10 according to the invention have a better performance while offering prospects for making new optoelectronic components.

Claims (11)

A laser source (10) comprising: - a first waveguide (12) extending in a longitudinal direction (Z), the first waveguide (12) forming a laser waveguide presenting a first waveguide (12) extending in a longitudinal direction (Z), the first waveguide (12) forming a laser waveguide presenting a first extension, said first length (L1), along the longitudinal direction (Z), and - a second waveguide (22) distinct from the first waveguide (12) and extending parallel to the first waveguide wave (12), the second waveguide (22) having a portion (30) superimposed with the first waveguide (12) along a stacking direction (Y) perpendicular to the longitudinal direction (Z) ), the portion (30) having a second extent along the longitudinal direction (Z) called the superposition length (the superposition), the overlap length (the superposition) being greater than 90% of the first length (L1), preferably greater than 95% of the first length (L1) and preferably equal to the first length (L1). 2. Laser source according to claim 1, wherein the first waveguide (12) has a first extension, called first width (dl), along a transverse direction (X) and the portion (30) of the second waveguide (22) has a second extent, said second width (d2), along the transverse direction (X), the transverse direction (X) being perpendicular to the stacking direction (Y) and the longitudinal direction (Z). 3. Laser source according to claim 2, wherein the second width (d2) is substantially constant along the longitudinal direction (Z). 25 4. Laser source according to claim 2 or 3, wherein the first waveguide (12) comprises a layer (16) made of p-doped semiconductor, the second width (d2) being such that the confinement factor (F) of the light wave in the layer (16) made of p-doped semiconductor is less than 5%, preferably less than 2%, and / or the confinement factor (F) of the light wave in the layer (18) comprising multiple quantum wells is greater than 1%, preferably greater than 20%. A laser source according to any one of claims 2 to 4, wherein the second width (d2) is smaller than the first width (d1). 6. Laser source according to claim 5, wherein the second width (d2) is greater than or equal to 25% of the first width (dl) and / or less than or equal to 40% of the first width (dl). 7. Laser source according to any one of claims 1 to 6, wherein the first waveguide (12) is made from at least one material belonging to column III of the periodic table composed of a material according to column V of the periodic table. 8. Laser source according to any one of claims 1 to 7, wherein the second waveguide (22) is made from silicon. 9. Laser source according to any one of claims 1 to 8, wherein the second waveguide (22) comprises a network (50). 10. Laser source according to any one of claims 1 to 9, further comprising an electro-absorption modulator (60). The laser source according to claims 9 and 10, further comprising a wavelength-selective, light-selective element (70) selected from the group consisting of two ring resonators, an assembly formed by a ring resonator. and a Mach-Zehnder interferometer and a sampled network.