Technical Field

The invention relates to the manufacturing of semiconductor optical conversion layers for use in detectors and emitters for applications in the infrared electromagnetic domain. Particularly, the invention provides a solution to achieve direct bandgap crystalline layers having a high conversion efficiency of light into electrical charges and vice-versa of electrical charges into light. More precisely the invention relates to GeSn crystalline layers and to efficient short- and mid-wave infrared photodetectors and light emitters that incorporate these GeSn layers. The invention also relates to the field of Lidars that comprise infrared detectors and emitters.

Background of the art

Germanium-tin (GeSn) alloys with their ability of bandgap engineering by varying Sn mole fraction, along with their compatibility with the complementary metal-oxide- semiconductor (CMOS) process have led to widespread efforts to achieve Si-based short-wave infrared (SWIR) and mid-infrared (MIR) optoelectronics devices such as modulators, photodetectors or light emitters that could lead to monolithic integrated products at low costs, which are essential components needed for future high-volume applications. Since the demonstration of direct bandgap Gei- _x Sn _x alloys, the technology has been developed in different flavors, e.g. epitaxy, chemical vapor deposition (CVD) or other deposition methods, without having reached its maturity yet.

Growth methods have been demonstrated in which molecular beam epitaxy (MBE) and CVD have obtained device quality material and high Sn incorporation. Up to now, GeSn alloy films have been prepared with Sn content of approximately 25% by molecular beam epitaxy (MBE) at about 120°C. Polycrystalline GeSn with a 25% Sn content has also been prepared by Sn-induced crystallization of amorphous Ge at 70°C. These techniques are extremely time-consuming and cannot deliver the required high deposit uniformity. In recent works, using a commercial reduced pressure-CVD (RP- CVD) equipment with tin tetrachloride (SnCL) and low-cost germane (GeH ₄ ) as precursors, successfully led to Gei- _x Sn _x growth directly on Si or on Ge-buffered Si substrates [Dou, 2018-1 : Tran, 2018; Pham, 2018].

The GeSn growth is extremely challenging due to the large lattice mismatch between GeSn and Si (>4.2%), and the low solid solubility (<0.5%) of Sn in Ge. Therefore, Ge buffer layers and non-equilibrium growth conditions at low temperature are now used. Currently, regarding the optical conversion layers - also defined as active layers enabling the absorption or the emission of light - higher order Ge hydrides, such as Ge2H6 and GesHs, are preferred due to their higher decomposition rate at lower temperatures. From an industrial manufacturing perspective GeFU is a better choice due to its much lower cost and commercial availability. Lately, a new growth approach with a RP-CVD equipment led to the demonstration of GeSn lasers with Sn content of 22.3% [Dou, 2018-1]. The realization of such lasers require multiple buffer layers of GeSn and relatively high process temperature. In theory, for emitters, high Sn content increases the bandgap directness and thus enhances the light emission efficiency whereas for detectors, high Sn content would extend the spectral response cutoff wavelength. However, in order to increase the Sn content - and at the same time allow for a monolithic integration - a substantially lower growth temperature is required to avoid Sn precipitation due to the low solubility of Sn in Ge. On the other hand, this results in lower growth rate, making the epitaxy process impractical. Therefore, increasing the Sn incorporation using GeH ₄ as the gas precursor faces a bottleneck with conventional CVD techniques.

So, there is a need for another scalable and low cost technique allowing for GeSn layers having a higher Sn content. The applicant has described in the document WO2018153503A1 a low energy plasma-enhanced chemical-vapor-deposition (LEPECVD) equipment to grow GeSn alloys on Ge-buffered Si at low temperature using GeH ₄ and SnCL as precursors. This technique has also been commented in the reference [Dou, 2018-2; Isella, 2004 and 2007] showing a significant enhancement of the Sn incorporation thanks to the rapid growth rate of GeSn epitaxy and far beyond the normal equilibrium of conventional CVD. A ten times higher growth rate of 51.4 nm/min and a Sn composition of up to 6% could be achieved confirming the potential of the technique towards enhanced Sn content GeSn films. However, the GeSn growth still remains an issue as epitaxy non-uniformities were observed due to complicated controllability of plasma density. Furthermore, the GeSn crystal layer quality deteriorates quickly as the growth temperature decreases.

In recent years, another deposition technique, i.e. sputtering epitaxy, has been developed for the growth of GeSn optical conversion layers. In particular, magnetron sputtering is a low-cost technique that provides an independent and easily controlled growth rate and growth temperature, a simple control over the alloy composition. Most significantly, the Ge and Sn targets are much safer than the precursor gases used in PECVD or CVD. By this technique, crystalline GeSn thin films with Sn content up to 28% could be deposited on Sn graded GeSn buffer on a Ge substrate at low temperatures [Zheng, 2018]. The thin films were found to be stable after annealing at temperatures below 400°C, which theoretically meets the needs of thermal budget for future photonic devices fabrication, indicating as well that high-Sn content GeSn-based photonic device fabrication should be compatible with the thermal treatments of further oxide deposition (CMOS process) and contact formation. However, in order to get high quality GeSn alloys, the threading defects density in the underlying layer is known to be critical [Zheng, 2016]. Although the physics of the process is not well understood as of today it is recognized that annealing an optical conversion layer may lead to improvements in the crystalline quality and optical characteristics of the GeSn layers [Mahmodi, 2018]

As the technological difficulty caused by the difference in lattice constants and dilatation coefficients between Si substrate and semiconductor films such as Ge or GeSn is eliminated for thin-film transistors fabricated on amorphous insulating substrates, advanced crystal growth of high-quality Ge and Ge-based crystals at temperature below 400°C led to significant enhancements in carrier mobility and electroluminescence efficiency. Among the various techniques investigated - such as laser annealing, chemical vapor deposition, flash lamp annealing and metal-induced crystallization - solid-phase crystallization (SPC) has many advantages. It is a simple and cost effective method, which generates no metal contamination and no melting- induced surface-ripples. However, the hole mobility of SPC-Ge deposited on glass has been limited to 140 cm ² V- ¹ s- ¹ by small grains. Only recently, by Sn-doped polycrystalline-Ge films on insulators, the grain boundary scattering could be lowered in order to achieve an improved hole mobility up to 320 cm ² V- ¹ s- ¹ [Sadoh, 2016]. It was also demonstrated that the atomic density of the deposited amorphous Ge influences subsequent SPC [Toko, 2017] and more recently, as heterogeneous nucleation at interfaces is dominant, the crystallinity of Ge thin films could be further improved with a record hole mobility of 620 cm ² V- ¹ s- ¹ by using Ge0 ₂ as most suitable underlying material during SPC [Imajo, 2018]. These significant enhancements in carrier mobility by advanced low temperature growth of high quality Ge and GeSn crystals on amorphous insulating films allow to eliminate the difficulty caused by the difference in lattice constants between Si substrate and semiconductor films.

Therefore, there is an improvement potential to achieve higher performance photonic devices, such as photodetectors and emitters. The present application addresses such improvements by providing a new and innovative optical conversion layer.

Further alloying of GeSn thin films with Si forming GeSnSi provides an additional way for bandgap engineering and thus the formation of hetero structures that may be useful for light emission applications [Sun, 2010, Stange, 2016, 2017, 2018; von den Driesch, 2017] Another benefit of adding Si is the release of the strain and thus the reduced degree of Sn segregation for a higher thermal stability [Kouvetakis, 2006].

Summary of the invention

The present invention solves the limitations of optical conversion layers of prior art and teaches how to provide high performance optical conversion GeSn-based layers to be used for SWIR and MIR photodetectors and light emitters. The range covering SWIR and MIR is defined as a wavelength range comprised between T000 nm and 5Ό00 nm. The fabrication process of the invention allows to achieve a crystal optical conversion layer having a very low threading dislocation density. It make use of a removable van der Waals film that has a weak interfacial energy with the underlying substrate and allows for further transfer on a crystalline substrate.

The GeSn optical conversion layer of the invention is realized before its incorporation or adaptation to emitter and detector parts, bases or supports. The optical conversion layer of the invention allows to provide a wide range of processing possibilities such as processes implicating high power lasers and thermal treatments that may not be compatible with certain types of detector and emitter parts or layers, such as layers comprising CMOS circuits or other elements that are very sensitive to heat and/or light such as high power layer light, or that may not stand radiation comprising high energy wavelengths such as UV light.

More precisely the invention provides a method of fabrication of a short-/mid-wave infrared (SWIR/MIR) optical conversion layer comprising at least germanium (Ge) and tin (Sn), and comprises the steps (A-F) of:

A. providing a substrate having a first surface and a second surface opposite to said first surface ;

B. depositing a van der Waals layer on a predetermined deposition area of said first surface;

C. defining at least one growth area on or in said van der Waals layer, said at least one growth area being defined in a virtual illumination plane (V) that is defined substantially parallel to said van der Waals layer;

D. depositing an optical conversion layer, made of a GeSn alloy, on said van der Waals layer.

In an advantageous embodiment, after step D a step E and a step F is executed, consisting of:

E: initiating a phase transition of said optical conversion layer, said phase transition starting at said predetermined growth areas;

F: crystallize at least partially said optical conversion layer by thermal annealing.

In an embodiment after said phase transition step E, a plurality of crystallographic and/or geometrical distinct portions are formed in said optical conversion layer.

In an embodiment, at the end of said crystallization step F substantially all of said distinct portions are fully crystallized. In an advantageous embodiment said step E is performed by the following sub steps E1 and E2

E1 : providing a light source;

E2: transmitting light provided by said light source, on said predetermined growth areas.

In an embodiment said light source is a laser. In an advantageous embodiment said van der Waals layer comprises graphene. In another embodiment said van der Waals layer is one of: a single graphene layer, a multi-layer graphene layer, a single boron nitride (BN) layer, a BN multilayer, a transition metal dichalcogenide (TMD) multilayer, or a combination thereof.

In an embodiment said van der Waals layer is a substantially uniform layer and wherein said growth areas are defined in close proximity, or in contact, to said van der Waals layer, to the side away from said substrate, and wherein said light comprises a plurality of focused light beams being substantially focused on said predetermined growth areas. In an embodiment said van der Waals layer is a structured layer comprising a predetermined arrangement of through-holes wherein said growth areas are defined in said through-holes.

In an embodiment said growth areas are located onto or near to said first surface.

In an embodiment wherein said optical conversion layer is made by a direct crystalline deposition technique. In an embodiment at least one mask layer is arranged between said substrate and said van der Waals layer. In an embodiment said mask layer is a structured layer, and may have substantially the same structure as said van der Waals.

In an embodiment the method comprises a removal step G consisting in removing said optical conversion layer from said substrate and so as to provide an optical conversion layer.

In an embodiment said removal step consists in applying a mechanical force that may be applied by a mechanical part that is adapted, before said removal step G, to said optical conversion layer, to the side away from said substrate. In an embodiment an additional step H is performed consisting of removing, at least partially, said substrate. In an embodiment said additional step H is realized at least partially by an etching technique. In an embodiment said additional step H is made by a focused ion beam (FIB)-cutting.

In an embodiment said GeSn is a layer made of a Gei- _x-y Sn _x Siy alloy, x and y being numbers lower than 1 , and 1-x-y being lower than 1.

In an embodiment said substrate is a crystalline substrate. In an embodiment said substrate comprises at least a portion made of Si or Ge. In an embodiment said substrate comprises at least a portion made of a polymer. In a variant said substrate is a flexible substrate. The invention is also achieved by a method to realize a short-/m id-wave infrared photodetector comprising the steps of:

- providing an optical conversion layer made by the described method;

- adapting said optical conversion layer to a photodetector substrate comprising at least one electrical charge converting portion and which is configured to collect electrical charges produced by said optical conversion layer when the photodetector) is in operation.

In an embodiment said at least one electrical charge converting portion comprises at least a CMOS read-out circuit. In an embodiment said photodetector is configured as an avalanche photodetector and comprises at least one avalanche portion. In an embodiment said photodetector is configured as an array of photodetectors. In an embodiment said optical conversion layer is bonded to said photodetector substrate by a bonding technique implying bonding temperatures lower than 350°C.

The invention is also achieved by a method to realize a short-/m id-wave infrared emitter comprising the steps of:

- providing an optical conversion layer made by the described method;.

- arranging said optical conversion layer onto or into a photo-emitter layer stack configured to generate electrical charges and convert said electrical charges into emitted photons when the photo- emitter is in operation.

In an embodiment said photo-emitter is configured as an array of light emitters and at least one of said light emitters may be configured as a micro laser. In an embodiment said optical conversion layer is bonded to an emitter substrate by a bonding technique implying bonding temperatures lower than 350°C.

The invention is achieved also by providing an optical conversion layer, realized by the described method said optical conversion layer being substantially flat and defines, parallel to its plane, two opposite surfaces, said optical conversion layer being made of a GeSn alloy and comprising a plurality of distinct crystalline portions.

In an embodiment at least one of said crystalline portions extends from one of said opposite surfaces to the other of said opposite surfaces.

In an embodiment said GeSn alloy is made of Gei- _x-y Sn _x Siy _, x and y being numbers lower than 1 , and 1-x-y being lower than 1. The invention provides also a photodetector comprising the described optical conversion layer. In an embodiment the photodetector comprises at least one CMOS readout circuit. In an embodiment the photodetector comprises at least two electrical contact layers and is configured as an avalanche photodetector.

In an embodiment the photodetector comprises an array of photodetectors.

The invention provides also a light emitter comprising the optical conversion layer as described. In an embodiment the light-emitter, comprises an array of light emitters. In an embodiment wherein at least one of said light emitters is a micro laser.

The invention is also achieved by providing short-/mid-wave infrared Lidar comprising at least one of said photodetector and/or at least one of said light emitter as described.

Brief description of the drawings

Embodiments of the invention are now described, by way of example only, with reference to the drawings in which:

- Figures 1a-1f illustrate steps of the method of fabrication of the optical conversion layer of the invention;

- Figure 2 illustrates an optical conversion layer of the invention;

- Figure 3 illustrates a variant of one of the fabrication steps of the method of the invention, implementing a structured van de Waals layer;

- Figure 4 illustrates a variant of the method wherein a mask layer is arranged between a substrate and a van der Waals layer;

- Figure 5 illustrates a variant of the method wherein a mask layer is arranged between a substrate and a van der Waals layer and wherein the first surface of the substrate has been structured prior before the deposition a GeSn layer;

- Figure 6 illustrates a method step consisting in removing by mechanical forces an optical conversion layer from a substrate;

- Figure 7 illustrates a schematically a method step to transfer and adapt, by applying a force, an optical conversion layer to a detector or an emitter base; - Figure 8 illustrates a SWIR detector comprising an optical conversion layer of the invention and an array of electrical charge converters;

- Figure 9 illustrates an exemplary avalanche photodetector comprising the optical conversion layer of the invention;

- Figure 10 illustrates a schematic cross section of a light emitter comprising the optical conversion layer of the invention.

Detailed description and embodiments of the invention

The present invention will be described with respect to particular embodiments and with reference to certain drawings but the invention is not limited thereto. The drawings described are only schematic and are non-limiting. In the drawings, the size of some of the elements may be exaggerated and not drawn on scale for illustrative purposes. The dimensions and the relative dimensions do not correspond to actual reductions to the practice of the invention.

It is to be noticed that the term“comprising” in the description and the claims should not be interpreted as being restricted to the means listed thereafter, i.e. it does not exclude other elements.

Reference throughout the specification to“an embodiment” means that a particular feature, structure or characteristic described in relation with the embodiment is included in at least one embodiment of the invention. Thus appearances of the wording“in an embodiment” or,“in a variant”, in various places throughout the description, are not necessarily all referring to the same embodiment, but several. Furthermore, the particular features, structures or characteristics may be combined in any suitable manner, as would be apparent to a skilled person from this disclosure, in one or more embodiments. Similarly, various features of the invention are sometimes grouped together in a single embodiment, figure or description, for the purpose of making the disclosure easier to read and improving the understanding of one or more of the various inventive aspects. Furthermore, while some embodiments described hereafter include some but not other features included in other embodiments, combinations of features if different embodiments are meant to be within the scope of the invention, and from different embodiments. For example, any of the claimed embodiments can be used in any combination. It is also understood that the invention may be practiced without some of the numerous specific details set forth. In other instances, not all structures are shown in detail in order not to obscure an understanding of the description and/or the figures.

The term van der Waals layer 20 which use is an essential part of the invention is to be understood broadly an encompasses any layer or structure that has the property of a van der Waals layer.

Van de Waals layers are well described in the literature and is not further described here in detail [Bakti, 2013; Koma, 1999; Xia, 2017]

It is also understood that the term van der Waals layer in this invention encompasses as well ionic layered materials and van der Waals layered materials. Also it is understood here that the van der Waals layer in the invention may comprise one or more ionic layers [McKinney, 2018].

The invention includes the following methods and embodiments of realization.

The invention relates to a short-/m id-wave infrared (SWIR/MIR) optical conversion layer 1 , its fabrication methods and its application in detectors, emitters and preferably Lidars but other systems in the field of SWIR/MIR applications are not excluded from the present application. The term short-/mid-wave infrared (SWIR/MIR) is defined here broadly and comprises wavelengths typically situated between 1Ό00 nm and 5000 nm.

As outlined in the prior art section there is first a need to provide high efficiency SWIR optical conversion layer. Additionally, in order to be possibly used in detectors and emitter comprising heat-sensitive components or layer such as layers comprising CMOS circuits there is an additional need to provide optical conversion layers that may be realized during a different process than the fabrication of said emitters and detectors. One of the reasons is that the fabrication process of a SWIR optical conversion layer may not be compatible with the fabrication process of the detectors and emitters or vice- versa.

None of the optical conversion layers of prior art allow to provide a high quality optical conversion layer that has a high optical conversion efficiency and that can be arranged on or into said detectors and emitters. The invention solves the problem of a long-felt need of optical conversion layers in the SWIR/MIR wavelength range for detectors and emitters, especially optical conversion layers that need to have very high photon/electrical charge conversion rate, that are stable under harsh conditions and that may be realized with available technologies at low cost and at low temperatures.

The invention uses therefor a van der Waals layer 20 in the proposed fabrication processes of the optical conversion layer 1 of the invention. One of the properties of the van der Waals layer is that it allows to realize an optical conversion layer 1 during a fabrication process that is separate of the fabrication process of said detectors and emitters.

The van der Waals layer 20 implemented in the method of the invention has the following properties and functions in the fabrication process of the optical conversion layer:

- the van der Waals layer 20, having a predetermined width W2, allows to provide a layer on which a GeSn optical conversion layer is deposited without being in full contact with a substrate 100, i.e. the GeSn optical conversion layer 1 is only partially, or not at all, connected to a substrate 100 during its growth;

- the van der Waals layer 20 allows to initiate on growth areas 1 _a -1 _n - into or on the van der Waals layer - the crystallization of the deposited GeSn layer;

- the van der Waals layer 20 allows to remove easily the optical conversion layer 1 from the substrate 100 on which it is realized.

More precisely, the invention is realized by a method of fabrication of a short-/mid- wave infrared optical conversion layer 1 , having a predetermined width wi, comprising at least germanium (Ge) and tin (Sn), and comprises the essential steps (A-F) of :

A. providing a substrate 100 having a first surface 102 and a second surface 104 opposite to said first surface 102;

B. depositing a van der Waals layer 20 on a predetermined deposition area defined on said first surface 102;

C. defining at least one growth area 1 _a -1 _n which is defined, as illustrated in Figs. 1c, 1 d, 3 in a virtual illumination plane V defined substantially parallel to said van der Waals layer 20;

D. depositing an optical conversion layer 1 , made of a GeSn alloy, on said van der Waals layer 20. In an embodiment a step E - following the deposition step D - is executed consisting in the initiation of a phase transition of said optical conversion layer (1), said phase transition starting at said predetermined growth areas 1a-1 n. In an advantageous embodiment the phase transition step E is followed by a further step F consisting in crystallizing said optical conversion layer 1 by thermal annealing. In other embodiments described further the optical conversion layer may be deposited by a direct crystallization deposition process that does not require said phase transition step E.

It is understood that said virtual plane V may coincide with said first surface 102, or with any plane in or near the van de Waals layer 20. Said virtual plane V may be in proximity above said van de Waal layer 20, and preferably defined at a distance of maximal 200 pm, preferably maximal 20 pm.

In an embodiment, a plurality of crystallographic and/or geometrical distinct portions 10a-10h are formed in said optical conversion layer, as illustrated schematically in Fig.1 e.

In embodiments the plurality of crystallographic and/or geometrical distinct portions 10a-10h are formed after said phase transition step E but may also be realized by a direct crystallization deposition process in the case of a van der Waals layer that comprises growth areas as further commented.

The van der Waals layer 20 has the following preferred properties:

- its material is preferably selected from a single graphene layer, a multi-layer graphene layer, a single hexagonal boron nitride (h-BN) layer, a h-BN multilayer, a transition metal dichalcogenide (TMD) multilayer, or a combination thereof;

- the van der Waals layer may comprise graphene or may be made entirely of graphene (single or multi-layer) and/or pyrolytic graphite

- its thickness is typically between a few Angstrom (one atomic monolayer) and can be greater than several hundred of micrometers

- the bond strength between the van der Waals layer and the optical conversion layer should be lower than typical values for ionic, metallic or covalent bond, allowing for an easy transfer to another substrate In an embodiment, at the end of said crystallization step F substantially all of said distinct portions 10a-10h are substantially fully crystallized.

The term “fully crystallized” has to be understood broadly and is commented further in detail. Some variants of fully crystallized optical conversion layers are described and commented further in the section related to the optical conversion layer 1.

It is understood that in variants, more than one van der Waals layer 20 may be used. For example a stack of different van der Waals layers may be implemented in the process.

In embodiments, illustrated in Figs 1a-1e, the van der Waals layer 20 is a substantially uniform and non-structured layer. In such embodiments, illustrated in Fig. 1c, said growth areas 1a-1 n are defined in close proximity, or in contact, to said van der Waals layer 20, to the side away from said substrate 100. In embodiments using a non-structured van der Waals layer, said step E is performed preferably by the following sub-steps E1 and E2:

Step E1 : providing a light source;

Step E2: transmitting light provided by said light source, on said predetermined growth areas 1a-1 n. Said light is preferably, but not necessarily, provided by a laser. The light source may also be a flash lamp or the like.

In an embodiment wherein a light source is used, the formation of a crystallized optical conversion layer 1 is realized by the steps A-F illustrated in Figs.1a-1e, as described now:

o in first steps A and B (Fig. 1a) a portion of the first surface 102 of a provided (step A) substrate 100 is covered with a van der Waals layer 20; o then, a GeSn layer is deposited (step C, Fig. 1 b) on the van der Waals layer 20 (step C);

o then (step D, Fig.1c), focused light beams 200a-200m, separated by a predetermined distance e, are provided to illuminate predetermined grow areas 1a-1 n. By focusing the light on said growth areas a phase transition is initiated at the growth areas or in their proximity, as illustrated in Fig. 1c and in an enlarged view of Fig.1 d; o the phase transition of the optical conversion layer 20 is made during a predetermined time and predetermined illumination conditions; o after a predetermined time, the optical conversion layer 1 is fully crystallized by thermal annealing, as illustrated in Fig. 1f;

o after the end of the crystallization process the optical conversion layer 1 may be eventually detached from the substrate and a single optical conversion layer 1 may be provided for further use, as illustrated in Figs. 6 and 7).

Said phase transition and its initiation, as well as the crystallization of the GeSn optical conversion layer 20 may be realized preferably by first illuminating with the lasers focusing the light on precise spots, then switching to another thermal treatment (such as RTA, FLA, or furnace) to completely crystallize the sample. Said phase transition and its initiation, as well as the crystallization of the GeSn optical conversion layer 20 may be realized by a huge range of process step variants and sequences. For example it is possible that a first heating step is realized before the illumination with laser light. It is also possible that alternating illumination and heating steps may be realized before the final crystallization of the optical conversion layer 1. As described further, and possibly for all embodiments, the top surface 1” of the optical conversion layer 1 may undergo a laser or heat treatment , and may be possibly be structured or smoothed , for example by using laser, heat, mechanical or chemical treatments, or a combination of said treatments. It is also understood that additional, structured or non- structured layers may be deposited on the top surface 1” of the optical conversion layer 1. Additional layers may be for example layers that improves or make easier and more reliable the separation, handling and transfer of the optical conversion layer 1 as further described. For example, said additional layers may be an oxide coating or polymer layer. In a variant said additional layer may be an ionic layer or a van der Waals layer that may be used in a transfer process of the optical conversion layer 1 to a detector or emitter base as further described.

The physical process of transforming said amorphous GeSn layer to a crystallized layer in the case of a uniform van der Waals layer 20 is now briefly described: When the sample is illuminated by a light beam, light is absorbed, and converted to heat (thermal energy). When thermal energy is provided to the sample, the atomic vibrations increase, weakening the rigidity of the atomic bonds in the material. If sufficient thermal energy is provided, atoms can move into a more favorable configuration, and will thus rearrange into a crystalline phase. Thus, only those regions where enough thermal energy is provided will crystallize.

In preferred embodiments a high power laser is used and the light beams 200a- 200m may be focused as circular or ellipsoidal light spots or may also be rectangular shaped spots, possible linear shaped focal spots. In embodiments different types of light spots may be applied. For example some light spots may have a rectangular cross section defined in said virtual plane V, and other spots may have different cross section in that plane V.

In embodiments, as illustrated in Fig. 3, said van der Waals layer 20 is a structured layer comprising a predetermined arrangement of through-holes 2a-2d wherein said growth areas 1a-1 n are defined in said through-holes 2a-2d. Fig. 3 illustrates the areas 3a-3d illustrating the onset of the crystallization process or the direct crystalline deposition. Said through-holes 2a-2d may have any shape 100a-100c, such as illustrated in Fig. 5, and have any cross section defined in any plane parallel to said virtual plane V.

In an embodiment said optical conversion layer is achieved by direct crystalline deposition without any phase transition step on said growth areas

The surface of the growth areas should be the least rough as possible, but it’s not a strict requirement.

In an embodiment, said growth areas 1a-1 n are located onto said first surface 102 or near to said first surface, for example at a distance of less than 1pm.

In an embodiment, at least one mask layer 30 is arranged between said substrate 100 and said van der Waals layer 20.

In an embodiment, said mask layer 30 is a structured layer. In variants, said mask layer 30 has substantially the same structure as said separation layer 20. The mask 30 can be made of any material, preferably a material that can be deposited because in any case the film has been transferred at the end of the fabrication process of the optical conversion layer 1. The mask 30 should act as a separation between the van der Waals layer 20 and the substrate 100, and ideally should not allow on its surface any crystal nucleation of the optical conversion layer.

The physical process of transforming said amorphous GeSn optical conversion layer 1 to a crystallized layer in the case of a structured van der Waals layer 20 may be made by applying thermal energy through the optical conversion layer 1 or through direct epitaxial deposition, with crystal seeding at the first surface 102, or near the first surface 102 if an overetch is made into the substrate, as shown in Fig 5. The overetch may lead to different shapes in the substrate such as a V-shape as illustrated in Fig. 5.

In an embodiment, the method of fabrication comprises a removal step G, illustrated in Fig. 6, so that an optical conversion layer 1 may be transferred to another support as described further in detail and illustrated in Fig. 7. For example in embodiments said removal step G may be performed after said crystallization step F as described above.

In an embodiment, said removal step consists in applying a mechanical removal force F1. The mechanical force F1 may be applied by a mechanical part 500 or mechanical holder that is adapted, before said removal step G, to said optical conversion layer 1 , to the side away from said substrate 100, as illustrated in Fig. 6.

In an embodiment, an additional step H, not illustrated in the figures, is performed consisting of removing, at least partially, said substrate 100. This step H may be realized at least partially by an etching technique. In an embodiment, said additional step H is made by a focused ion beam (FIB)-cutting.

In an embodiment, said GeSn layer is a layer made of a Gei- _x-y Sn _x Si _y alloy, x and y being numbers lower than 1 and 1-x-y being lower than 1.

It is also understood that the optical conversion layer 1 may comprise portions having different alloy compositions. For example at the center of the optical conversion layer the GeSn may be made of Gei- _x-y Sn _x Si _y , while at the border areas the optical conversion layer may be made of another GeSn alloy or any other type of alloy. The variant of having different alloy compositions across the optical conversion layer 1 is to provide the possibility to have different optical conversion efficiencies at different locations in said optical conversion layer 1 which can be useful in arrays of detectors and emitters which require different conversion efficiencies for different portions of detector and/or emitter arrays. Indeed, one may provide detector arrays of which at least two portions are configured to address different SWIR wavelengths. It is understood that different GeSn alloy compositions may be realized according to different geometries, for example concentric geometries or geometries of parallel strips having different compositions.

It is generally understood that before, during or after the fabrication process of the optical conversion layer 1 any other additional layer or structure may also be realized , such as for example integrating a portion, such as a rim, of isolating material or metallic or polymer structures. For example such any other additional layer or structure may be a rim realized on top of said optical conversion layer 1 so that a mechanical part may be clamped or attached to said optical conversion layer 1 so as to make its separation and removal from said substrate 100 more easily and more efficient, i.e. reducing the risk of damaging the optical conversion layer 1.

It is also understood that the top surface of the optical conversion layer 1 must not be necessarily a flat surface. Means may be provided during the fabrication process so that the top surface 1”, i.e. the surface away from said substrate 100, of the optical conversion layer 1 is structured with the intention to separate and remove it from said substrate 100 more easily and more efficient. For example the GeSn optical conversion layer may comprise apertures and through holes, or may comprise U- or V-shaped structures on its surface, preferably realized directly into the top surface of the GeSn optical conversion layer 1.

It is understood that the substrate 100 may be made in different materials or may be made from different parts or may comprise different portions made of different materials. In an embodiment, said substrate 100 is a crystalline substrate 100.

In an advantageous embodiment, said substrate 100 comprises at least a portion made of Si. Preferably the substrate is a silicon (Si) substrate 100. In an embodiment, said substrate 100 comprises at least a portion made of a polymer such as PMMA, PDMS, SU-8.

In an embodiment, said substrate 100 is a flexible substrate, allowing to realize optical conversion layers 1 onto a flexible substrate. It has to be noted here that the end product must not be necessarily be a single optical conversion layer 1 but may also be a flexible substrate comprising the optical conversion layer 1 as described.

The invention is also realized by a method to realize a short-/mid-wave infrared photodetector 40 comprising the steps of:

- providing an optical conversion layer 1 made by the method described here;

- adapting said optical conversion layer 1 to a photodetector substrate 50 comprising at least one electrical charge converting portion 54 and which is configured to collect electrical charges produced by said optical conversion layer 1 when the photodetector 40 is in operation.

The term photodetector substrate 50 is to be understood broadly and may be a base comprising a plurality of layers, as illustrated in 8. For example a base layer may comprise doped layers such n- or p-doped layers.

In variants a plurality optical conversion layer 1 may be arranged in a photodetector 40. At least two of said plurality optical conversion layer 1 may be different optical conversion layers or may have different geometries.

In an embodiment, said at least one electrical charge converting portion 54 comprises at least a CMOS read-out circuit.

The method according to claim 25 or claim 26 wherein said photodetector 40 is configured as an avalanche photodetector and comprises at least one avalanche photodetector, as illustrated in Fig. 9. In an embodiment the avalanche detector comprises a p layer 54 and a p+ layer 56 and may comprise additional layers such as a front layer 58 and an intermediate layer 52 which may be a layer to improve the adherence of the optical conversion layer 1.

It is understood that in all embodiments the optical conversion layer 1 may comprise at least one doped or post-treated portion or an additional layer 42, 44, such as illustrated in Fig. 8. In an embodiment, photodetector 1 is configured as an array 52 of photodetectors 50i-50 _n .

In an embodiment, said optical conversion layer 1 is bonded to said photodetector substrate 50 by a bonding technique implying bonding temperatures lower than 350°C.

The invention is also achieved by a method to realize a short-/mid-wave infrared emitter 60 comprising the steps of:

- providing an optical conversion layer 1 made by the described fabrication method;

- arranging said optical conversion layer 1 into or onto a photo emitter layer stack configured to generate electrical charges and convert said electrical charges into emitted photons when the photo emitter 60 is in operation.

The term photo emitter layer stack is to be understood broadly and may be a photo-emitter base comprising a plurality of layers 62-68, as illustrated in the example of Fig. 10. For example a base layer may comprise doped layers such n- or p-doped contacts.

In variants a plurality optical conversion layers 1 may be arranged in a photo emitter 60. At least two of said plurality optical conversion layers 1 may be different optical conversion layers or may have different geometries.

In an embodiment, said photo-emitter 60 is configured as an array of light emitters. In an embodiment, at least one of said light emitters 60 is a micro laser.

In an embodiment, said optical conversion layer 1 is bonded to an emitter substrate 62 by a bonding technique implying bonding temperatures lower than 350°C.

It is also understood that, in a variant of the methods of the invention a hybrid emitter and detector platform may be realized. The interest of this is to provide a compact emitter/detector assembly useful in Lidars. To realize such hybrid platform an emitter/detector substrate may be provided on which at least a first optical conversion layer is arranged so as to provide an emitter part configured to emit light, and on which at least second optical conversion layer is arranged so as to provide a detector configured to detect light. The invention relates also to an optical conversion layer 1 , realized by the method of the invention as described. In a preferred embodiment, said optical conversion layer 1 is substantially flat and defines, parallel to its plane, two opposite surfaces 22, 24, said optical conversion layer 1 being made of a GeSn alloy and comprising a plurality of distinct crystalline portions 12a-12n.

In an advantageous realization the optical conversion layer may be realized on a curved substrate and may be a curved optical conversion layer. In an embodiment said two opposite surfaces 22, 24 must not necessarily have the same shape. It is also understood that the thickness of the optical conversion layer 1 must not be necessarily uniform.

In an embodiment, at least one of said crystalline portions, defined as“grains”, extends from one of said opposite surfaces 22, 24 to the other of said opposite surfaces 22, 24.

The size of the grains should be as large as possible, i.e. preferably larger than 20 micron in length, as the interface states at the grain boundaries are detrimental in terms of dark current, carrier mobility and Sn segregation.

There are no well-defined contact surface shape of the plurality of said crystalline portions. Large grains are preferable, defined as micrometer-sized grains. Ideally the grain sizes of the optical conversion layer 1 used in a SWIR/MIR detector as described should be larger than the size of a pixel of a detector

In an embodiment, the optical conversion layer 1 is made of at least partially Gei- _x-y Sn _x Si _y alloy, x and y being numbers lower than 1 and 1-x-y being lower than 1 , x and y being numbers smaller than 1.

In the case that the GeSnSi is realized by a method as illustrated in Figs. 3 or 4, the substrate should be a crystalline substrate made of Si, Ge or a combination of them, including buffer layers such as Ge (in the case of a Si substrate), or graded SiGeSn the term“graded” refers to a gradual variation in the composition with increasing mismatch with the substrate, and increasing lattice match with the optical conversion layer), or a combination thereof. The addition of silicon in the optical conversion layer 1 helps to release the strain of the GeSnSi film and also stabilize the Sn in the structure, avoiding segregation. The invention is also achieved by a photodetector 40 comprising the optical conversion layer 1 as described above and illustrated in Figs. 7-9.

In an embodiment, the photodetector 40 comprising a substrate 50 that comprises at least one CMOS readout circuit.

In an embodiment, the photodetector 40 comprises at least two electrical contact layers 42, 44, 56, 54 and is configured as an avalanche photodetector.

In an embodiment, the photodetector 40 comprises an array 52 of photodetectors 50i-50 _n that may be configured as an array of avalanche photodetector.

Fig. 7 and Fig. 9 illustrate respectively an example of an array of photodetectors comprising an array of pixels 50i-50 _n and an avalanche photodetector configured to detect SWIR light. In embodiments, such as illustrated in Fig. 9 the photodetetector 40 may comprise a bonding layer 50’ and or a buffer layer 52 that may be for example a Ge buffer layer.

The invention is also achieved by a light emitter 60 comprising the optical conversion layer 1 as described above, and illustrated in Fig. 7 and Fig. 10. Said light emitter 60 is configured to emit from its output surface 60 _a at least one light beam 300 when it is in operation. It is understood that the light emitter may be configured to emit at least one light beam that is not perpendicular to said top layer 1”.

In an embodiment, the light-emitter 60 comprises an array of light emitters, and in a variant at least one of said light emitters is a micro laser. In embodiments, such as illustrated in Fig. 10 the light-emitter 60 may comprises an adaptation or bonding layer 60’ between said optical conversion layer 1 and a substrate 62. The emitter 60 may comprise typically p-doped contact layers 64-68.

The invention is also achieved by providing a short-/m id-wave infrared Lidar comprising at least one of said photodetector 40 of the invention as described above and/or at least one of said light emitter 60 according to the invention as described above. The way how said optical conversion layer 1 can be adapted to one of the surfaces 2a, 2b of a detector basis and/or an emitter basis has been de described above in the method section and is illustrated schematically in Fig. 7.

Exemplary Applications The optical conversion layer 1 of the invention is mainly intended for applications in the field of detectors and emitters and Lidars, but may also be used for other applications that require a crystallized optical conversion layer 1 made of a GeSn alloy.

References

The following additional Publications are incorporated herein by reference thereto and relied upon:

Publications

1. Bakti U. et al.: Recent developments and future directions in the growth of nanostructures by van der Waals epitaxy. Nanoscale, 5(9), 3570, 2013

2. Dou W. et al.: Investigation of GeSn Srain Relaxation and Spontaneous Composition Gradient for Low-Defect and High-Sn Alloay Growth, Scientific Reports, 8:5640 04 April 2018

3. Dou W. et al.: Crystalline GeSn growth by plasma enhanced chemical vapor deposition, Optical Materials Express 3220, Vol. 8, No 10, 1 Oct 2018

4. von den Driesch N. et al.: SiGeSn Ternaries for Efficient Group IV Heterostructure Light Emitters. Small. 13(16), Wiley, 3 Feb 2017

5. Imajo T. et al.: High hole mobility (>500 cm ² V ¹ S ¹ ) polycrystalline Ge films on Ge0 ₂ - coated glass and plastic substrates, The Japan Society of Applied Physics, 30 Nov 2018

6. Isella G. et al.: Low-energy plasma enhanced chemical vapor deposition for strained Si and Ge heterostructures and devices, Solid-State Electron. 48(8), 1317-1323, 2004

7. Isella G. et al.: Heterojunction photodiodes fabricated from Ge/Si (100) layers grown by low-energy plasma-enhanced CVD, Semicond. Sci. Techol. 22(1), S26-S28, 2007

8. Koma, A.: Van der Waals epitaxy for highly lattice-mismatched systems. Journal of Crystal Growth, 201-202, 236-241 , 1999

9. Kouvetakis J. et al.: Tin-based Group IV Semiconductors: New Platforms fo Opto- and Microelectronics on Silicon, Annual Review of Materials Research, vol. 36:497-554, 4 Aug 2006

10. Mahmodi H. et al.: Synthesis of Gei- _x Sn _x Alloy Thin Films by Rapid Thermal Annealing of Sputtered Ge/Sn/Ge Layers on Si Substrates. Materials, 11 , 2248, 12 Nov 2018

11. McKinney R.W. et al.: Ionic vs. van der Waals layered materials: identification and comparison of elastic anisotropy, Journal of Materials Chemistry A, issue 32, 2018. 12. Pham T.: Si-based Germanium Tin Photodetectors for Short-Wave and Mid-Wave infrared Detections, University of Arkansas,, Fayetteville, Dissertation 2838, 8-2018

13. Sadoh T. et al.: Hugh carrier mobility of Sn-doped polycrystalline -Ge films on insulators by thickness-dependent low-temperature solid-phase crystallization, Appl. Phys. Lett.109, 232106, 6 Dec 2016

14. Stange D. et al.: Study of GeSn based heterostructure: towards optimized group IV MQW LEDs. Optics Express, 1358, Vol. 24, No. 2, 15 Jan 2016

15. Stange D. et al.: Short-wave infrared LEDs from GeSn/SiGeSn multiple quantum wells.

Optica, Vol. 4, No. 2, Feb 2017

16. Stange D. et al.: GeSn/SiGeSn Heterostructure and Multi Quantum Well Lasers. ACS Photonics, 5, 4628-4636, 19 Oct 2018

17. Sun G. et al.: Design of a Si-based Lattice-Matched Room-Temperature GeSn/GeSiSn Multi-Quantum-Well Mid-Infrared Laser Diode, Opt. Express 18, 19957-19965, 2010

18. Toko K. et al.: High-hole mobility polycrystalline Ge on an insulator formed by controlling precursor atomics density for solid-phase crystallisation, Scientific Reports, 7:16981 , 5 Dec 2017

19. Tran H. et al.: High performance Ge0.89Sn0.11 photodiodes for low-cost shortwave infrared imaging, Journal of Applied Physics 124, 013101 , 2 July 2018

20. Xia, W. et al.: Recent progress in van der Waals heterojunctions. Nanoscale, 9(13), 4324-4365, 2017

21. Zheng J. et al.: GeSn photodetectors with GeSn layer grown by magnetron sputtering epitaxy, Appl. Phys. Lett. 108, 033503, 20 Jan 2016

22. Zheng J. et al.: Growth of high-Sn content (28%) GeSn alloy films by sputtering epitaxy, Journal of Crystal Growth, 492, 29-34, 1 1 April 2018