WO2014053386A1 - Electronic device including filament nanostructures and method for making such devices - Google Patents

Abstract

The invention relates to an electronic device including a first electrode (9) having a first portion (10) including a metal material and a second porous portion (11) including an array of substantially aligned pores (12), the first electrode (9) being deposited onto a substrate made of an insulating material (14), a second electrode (13) including a metal material, a plurality of semi-conductor filaments (15), wherein each filament (15) includes a root (16) and an end (17), the root (16) is arranged in a pore (12) and is in electric contact with the first portion (10), the end (17) of a filament (15) is in electric contact with at least one nanoparticle (18) including a metal material, the nanoparticle (18) is in electric contact with the second electrode (13), the electric contact between the end (17) of the filament (15) and the nanoparticle (18) is of the rectifier type so that there is an energy barrier for the passage of carriers between the filament (15) and at least one particle (18), and the filament (15) has no direct electric contact with the second electrode (13).

Description

 Electronic device comprising nanostructures in filaments and method of manufacturing such devices

FIELD OF THE INVENTION

The present invention relates to the field of electronic and optoelectronic devices and more particularly devices based on nanostructures in semiconductor filaments, such as carbon nanotubes ("Carbon Nanotube" or CNT in English) or nanowires ("nano wire" or nW in English). The present invention also relates to the methods of manufacturing such devices.

The present invention applies in particular to electronic components, or to components that transform light radiation into an electrical signal, for example to generate an electric current from solar radiation or to control the switching of a signal. electric using a light source.

STATE OF THE ART

The production of electronic components based on different types of semiconductors is known.

 Regarding the transformation of light into electricity, one possibility is to use the photovoltaic effect. The photovoltaic effect is obtained by absorption of the photons in a semiconductor material which then generates electron-hole pairs (excitation of an electron from the valence band towards the conduction band) creating, when certain conditions are fulfilled, a voltage or an electric current.

The absorption of a photon in a solid having a band, or "gap" in English, is very probable because this electronic transition is only associated with two particles: electron - photon. The photon ensures the conservation of energy during the transition between the valence band and the conduction band. This direct band structure corresponds to materials such as GaAs, CdTe, etc. When the material is at indirect gap, a third particle comes into play. It is associated with the vibrations of the crystal lattice and is referred to as the "phonon" of energy. Its intervention in the absorption process leads to a reduced probability of absorption near the threshold, and therefore to an "absorption front" which is less steep than that which characterizes the direct transitions. This indirect band structure applies to materials such as silicon, germanium, GaP, etc.

When a photon of energy greater than a value Eg (semiconductors having a direct gap) or Eg + Ω (semiconductors having an indirect gap), is absorbed by the semiconductor, an electron-hole pair is created. The electric charges (electron and hole) remain bound by a coulombic force and form a set called "exciton". To generate an electric current the excitons must dissociate into electrons and free holes before these two elements recombine. There are mainly two types of recombination. The first is due to a physical process of spontaneous emission (radiative recombination), the second mechanism is that resulting from the recombination by traps which can be impurities and defects present in the material. These impurities introduce energy levels in the forbidden band and thus create centers of recombination.

 When the electron-hole pairs dissociate, the free charges must diffuse to the electrodes. The mobility of charge carriers defines the ability of these electrons or holes to move within a material. The collection device requires a large electric field to separate the electron-hole pairs. It can be created by juxtaposing the semiconductor in intimate contact with a metal (Schottky barrier) or with another semiconductor. In the latter case, the semiconductors may be of the same material but doped differently (homojunction) or different materials (heterojunction).

Several types of components can be made from this principle, such as photodiodes, phototransistors or photocells.

An example of a component widely used in the solar panel industry is the silicon PIN photodiode, consisting of a P-doped zone, a zone undoped so-called intrinsic and an N-doped zone. This component operates with a bias voltage.

 A limitation of these components is the limited absorption of silicon, which does not absorb photons beyond a wavelength of 0.8 μιτι. To improve the conversion efficiency, direct-gap semiconductor-based components, such as gallium arsenide (GaAs), are developed. But the main disadvantage of these components is their high cost.

 Another example of a component is the solar cell, which does not require a bias voltage. A commonly used photocell consists of a PN junction in monocrystalline silicon or PIN in amorphous silicon. This component does not require a bias voltage.

The main disadvantages are the inhomogeneous absorption of monocrystalline silicon and the short diffusion length of the carriers in the amorphous silicon, thus limiting its conversion efficiency.

More recently, a new technological brick has emerged with one-dimensional filamentary structures, which take the form of nanotubes or nanowires (nW).

Carbon nanotubes (CNT) are macromolecular systems that have unique physical properties: according to their chirality (or else their helicity) they can be metallic or semiconductors (called in this case s-CNT). Semiconductor CNTs are direct band and intrinsically photoconductive.

Their small diameters and long lengths make them practically ideal one-dimensional (1 D) systems. They are very much studied at present, in particular for the realization of field effect transistors, because they appear as one of the most promising ways to continue the miniaturization of MOS components beyond the next decade. After the first demonstration of the possibility of using carbon nanotubes as field-effect transistors (CNT-FETs) in 1998, inverter-type demonstrators, NOR gate, SRAM memory or even more complex circuits as well as very high frequency transistors have already emerged in several research centers. These demonstrations made it possible to highlight the potential of these nanomaterials for electronic or optoelectronic applications because of their excellent electrical transport and absorption properties. In the field of optoelectronics, IBM researchers have shown that it is possible to excite optical transitions of CNTs, simultaneously injecting electrons and holes in the channel of a FET transistor. Due to the small gap of single-wall CNTs, the radiation is emitted in the near infrared; moreover, it is polarized parallel to the main axis of the CNT. IBM researchers have also shown that it is possible to create a photocurrent in the FET when it is illuminated by infrared light, even if the wavelength in the infrared is greater than the size of the device. Potential barriers at the contact level have a great influence on the injection and extraction of carriers within the CNTs. Among the means implemented for the control of the potential barriers, there is for example the realization of multi-grids addressable independently.

 The semiconductor nanowires (denominated in this case s-NW) are for example based on silicon or germanium with a stress dimensions below 100 nm. Remember that Si or Ge are still today the materials of choice for photovoltaic applications, the best solar cell yields being obtained with these materials in their monocrystalline form. In the form of nanowires, the constrained axial dimension can induce a confinement of the volume of the carriers while the unstressed longitudinal dimension can be exploited to control the transport of carriers or photons. These nano-objects are therefore very promising materials for the manufacture of optoelectronic components. In addition, this type of crystalline material can be synthesized on a wide variety of substrates (for example glass, metal strips, etc.) thus significantly reducing production costs.

These devices, however, have several disadvantages, mainly related to the lack of control (direction, positioning, density) of the synthesis of CNT. Thus, the nanotubes cross the electrodes in random directions. Due to the dependence of the angle with which the tubes cross the contacts, the length of the channel of the nanotubes is different. These variations in the length of the channel and the density of nanotubes in the channel lead to wide variations in the electrical properties of the devices, which is clearly an obstacle to the reproducible fabrication of devices.

An object of the invention is to provide a novel device based on semiconductor filaments making it possible to produce a rectifier-type electronic device or an optoelectronic device capable of generating a photocurrent under illumination, as an alternative to semiconductor-based components. conventional conductors, and not having the aforementioned drawbacks.

DESCRIPTION OF THE INVENTION

To achieve this goal, the invention proposes an electronic device comprising:

 a first electrode comprising a first part comprising a metallic material and a second porous part comprising a network of substantially aligned pores, said first electrode being deposited on a substrate made of insulating material,

a second electrode comprising a metallic material,

 a plurality of semiconductor filaments, a filament comprising a root and an end, the root being disposed in a pore and in electrical contact with said first part, the end of a filament being made in electrical contact with at least one nanoparticle comprising a metallic material, the nanoparticle being mounted in electrical contact with the second electrode,

 the electrical contact between the end of the filament and the nanoparticle being of the rectifier type so that there exists an energy barrier for the passage of carriers between the filament and the at least one particle,

the filament being without direct electrical contact with the second electrode.

The device according to the invention uses dense networks of semiconductor filaments. A high density of nanotubes increases the density of the current able to be transported in the device. The filament network constitutes the conduction channel of the device. According to one variant, the semiconductor filament is a carbon nanotube (s-CNT).

 According to another variant, the semiconductor filament is a semiconductor nanowire (s-nW), for example silicon or germanium.

 In the device according to the invention, the metal nanoparticle is located at the end of the semiconductor filament, between the filament and the second metal electrode, and is in electrical contact with the filament on one side and with the second electrode of the other side. The nature of the metal of the nanoparticle and the semiconductor filament is such that there is an energy barrier, called a Schottky barrier, at the filament / metal junction of the nanoparticle. This contact between the semiconductor, in this case the filament, and the metal, in this case the nanoparticle, is commonly referred to as a rectifying contact.

The existence of this potential barrier is an essential and original feature of the filament-based device, making it suitable for making a rectifying contact in electronics and generating a photocurrent under illumination for an optoelectronic application.

 In electronics, an example of use of the device according to the invention is to provide a rectifier in the field of microwave waves, which requires the transport of a large power.

 S-CNT and s-NW are intrinsically photoconductive. This property is used to make opto-electronic components.

In optoelectronics, a light flux illuminating the photoconductive filament creates in it electron pairs / holes that can migrate over large distances without recombination, up to the level of the Schottky potential (contact) potential barrier for which the photogenerated free carriers by Absorption of light in the filaments can be separated by the internal field created at the filament / electrode junction. This effect can generate an electromotive force and a photocurrent under illumination (photocell operation) and has the effect of reducing the height of the energy barrier, thus allowing optical modulation of the conductivity of the device (photodetector or photoc switch / modulator operation). It is thus possible to produce, from the device according to the invention, various types of optoelectronic components, such as photodiode, photocell, photocoupler ("photoswitch" in English), photodetector (fast and broadband) ...

The pore geometry of the filaments makes it possible to orient the filaments preferentially in one direction, which leads to a good organization of the filaments in the channel of the devices. In addition, the use of filaments as semiconductor material and photoconductor involves several advantages.

For s-CNTs, graphene is a very "black" material, that is to say that it has a very broad absorption spectrum of the light spectrum, much wider than that of silicon. The absorption spectrum of the carbon nanotubes comprises the entire visible spectrum and the infrared, typically from 200 nm to 10 μιτι-15 μιτι. A first advantage consists in an increased sensitivity of the optoelectronic device according to the invention.

In addition, the device is of great adaptability when used as a switch controlled by a light source (photoc switch) modulable in intensity, such as a laser or a light emitting diode. Because of the broad absorption spectrum, there is an important choice, in terms of wavelength, of sources of illumination.

Most commercial photovoltaic cells use silicon. However, this material has a relatively low absorption coefficient, which requires the use of high thicknesses, typically hundreds of microns to obtain a 90% absorption of light. These high thicknesses go against an efficient collection of generated charges. With s-NW, it becomes possible to exploit a significant optical absorption length in the unstressed longitudinal direction while the presence of a constrained axial dimension (<100 nm) induces containment effects to achieve very high carrier mobilities. Thus, s-CNT or s-NW have very high carrier mobilities, of the order of a thousand to a few thousand cm ² / Vs (typically 100 times more than in silicon).

 In a photoc switch application this high mobility allows a very high switching frequency.

Another advantage is the generation of a large amount of photoinduced current, since the filament growth processes provide a large number of filaments in a small volume. In addition, because the potential barrier is high, carriers only pass through tunneling. So the probability of crossing the barrier is very low, which induces a very weak current of darkness.

In addition, the device according to the invention is compatible with inexpensive insulating substrates such as glass.

Advantageously, the first part of the first electrode comprises aluminum and the second part of the first electrode comprises porous alumina. According to one variant, the pores are oriented substantially parallel to the plane of the substrate and the second electrode is also deposited on the substrate.

According to another variant, the pores are oriented substantially vertically to the plane of the substrate and the second electrode comprises a metal film carried on the filaments.

Advantageously, the metal of a metal nanoparticle has an output work greater than 5 eV.

Advantageously, the metal of a metal nanoparticle comprises nickel.

Advantageously, the device according to the invention further comprises a plurality of metal particles fixed on the wall and on the end (17). filaments, so that at least one of the metal particles attached to the end makes the electrical contact of the rectifier type.

It is also proposed, according to another aspect of the invention, a Schottky type diode comprising a device as described above and means adapted to apply a potential difference between the first part of the first electrode and the second electrode.

It is also proposed, according to another aspect of the invention, an optoelectronic device as described above in which the filaments are arranged to be able to receive light radiation.

It is also proposed, according to another aspect of the invention, an optoelectronic assembly comprising at least one photodiode and / or a photocell and / or a photodetector and / or a photoc switch incorporating at least one optoelectronic device as described above.

It is also proposed, according to another aspect of the invention, a particle detector comprising a scintillator and an optoelectronic device as described above.

It is also proposed according to another aspect of the invention a method of manufacturing an electronic device comprising:

 the production of a first electrode deposited on a substrate of insulating material, comprising a first part comprising a metallic material and a second porous part comprising a network of substantially aligned pores,

 the production of a second electrode comprising a metallic material,

the production of a plurality of semiconductor filaments comprising a step of depositing catalytic elements inside the pores, the elements comprising at least one nanoparticle comprising a metallic material, then a phase of vapor phase growth of the filaments from the catalytic elements, a filament comprising a root disposed in a pore and in electrical contact with the first part, and an end in electrical contact with the nanoparticle, the growth being carried out up to that the nanoparticle is in electrical contact with the second electrode, avoiding any direct electrical contact between the filament and the second electrode, the contact between the end of the filament and the nanoparticle being of the rectifier type so that there is a barrier of energy for the passage of carriers between the filament and the nanoparticle.

An advantage of the method according to the invention is that it solves the problems concerning the organization, positioning and handling of nanowires or carbon nanotubes in conjunction with their controlled growth.

According to one embodiment, the growth of the filaments takes place in a confined environment, in nanopores of porous alumina membranes obtained by anodic oxidation of the aluminum films according to a known method.

Thus, the process is carried out from an aluminum layer in which the second porous part is produced by an anodic oxidation process of aluminum.

According to a method according to the state of the art, the resumption of contacts is carried out by burying the ends of the filaments in a metal.

On the other hand, the process according to the invention continues to grow until it comes into contact with the second electrode, avoiding any direct electrical contact between the filament 15, including its end 17, and the second electrode. The process according to the invention achieves the growth of very dense organized networks, with a filament density greater than 10 ⁹ / cm ² .

According to one variant, the second electrode is produced before the growth step and on the same substrate, the second electrode having a substantially vertical face facing a substantially vertical face of the first electrode comprising the pores, and the production of The pores of the first electrode are made so as to obtain pores oriented substantially parallel to the plane of the substrate, the growth taking place so that filaments extend substantially parallel to the substrate over a sufficient length to reach the face. look. According to another variant, the pores of the first electrode are produced so as to obtain pores oriented substantially perpendicular to the plane of the substrate, and the growth is carried out so that the filaments rise vertically, and the production of the second electrode comprises the transfer of a metal film on the nanoparticles. Advantageously, the metal film comprises a thin layer of graphene.

 Advantageously, the process as described above also comprises a step of decorating the filaments carried out after the growth step, in which metal particles are fixed on defects and at the ends of said filaments, so that at least one of the particles attached to the end makes the electrical contact rectifier type.

Other features, objects and advantages of the present invention will appear on reading the detailed description which follows and with reference to the appended drawings, given by way of non-limiting example, and in which:

 FIG. 1 represents an exemplary device according to a first variant of the invention,

 FIG. 2 represents an exemplary device according to a second variant of the invention,

FIG. 3 illustrates the energy band diagram before and after contacting the filament with the electrode on the one hand and the metal particle on the other hand;

 FIG. 4 represents a variant of the device according to the invention comprising metal particles fixed on the defects and at the ends of the filaments,

 FIG. 5 represents an optoelectronic device according to the invention for which the filaments are arranged to receive a light radiation,

FIG. 6 illustrates an example of an electrical characteristic of the electronic device according to the invention, FIG. 7 illustrates examples of characteristics of the optoelectronic device according to a first variant of the invention, with and without illumination by a luminous radiation,

 FIG. 8 illustrates an example of variation of the current flowing in a device according to the invention as a function of the illumination received and as a function of the applied voltage,

 FIG. 9 illustrates an example of variation of the responsivity of a device according to the invention as a function of the power of the illumination laser.

FIG. 10 illustrates the standardized responsivity of a device according to the invention as a function of the frequency of an illumination laser diode.

 FIG 1 1 represents the steps of a first variant of a method of manufacturing an electronic device

 FIG. 12 represents the steps of a second alternative method of manufacturing an electronic device

FIG. 13 represents an optional step of the method of manufacturing an electronic device

DETAILED DESCRIPTION OF THE INVENTION FIG. 1 schematically represents a device according to a first variant of the invention. A first metal electrode 9 and a second metal electrode 13 are deposited on an insulating substrate 14. Typically the electrodes have a thickness of a few tens to a few hundred nm and are spaced about 10Ομιτι to 1 mm.

This insulating substrate may comprise glass, quartz, sapphire or another insulating substrate depending on the intended application, or a layer of SiO ₂ or another insulator deposited on any substrate, such as silicon or a metal.

 The first electrode 9 comprises two parts. A first portion 10 comprising a metallic material and a second porous portion 11 comprising a network of pores, these pores being substantially aligned parallel to the plane of the substrate 14.

Preferably, the first part comprises aluminum and the second part corresponds to a part of the electrode comprising aluminum which has been converted to porous alumina by anodic oxidation according to a known method.

 Inside some pores, a filament 15 is born. Its root 16, disposed inside the pore, is in electrical contact with the first metal part 10 of the first electrode 9.

 Preferably, for the first electrode, a metal is chosen whose energy diagram is matched to that of the semiconductor filament 15, so that the circulation of the carriers is easy. Typically the output work of the metal is very close to that of the filament. The output of a metal is defined as the minimum energy that must be supplied to an electron to pull it out of the metal. For a semiconductor, the output job is defined as the difference between the vacuum level and the Fermi level in the semiconductor.

 A semiconductor carbon nanotube typically has an OCNT output work of about 4.5 eV, typically between 4.5 eV and 4.8 eV.

A Si or Ge nanowire typically has an output work Φ _3-Ν τ typically between 4.3 eV and 4.7 eV.

 Aluminum, for example, with an output work of between 4.06 eV and 4.26 eV, is a suitable metal.

A nanotube typically has a diameter of a few nm to a few tens of nm.

 A nanowire typically has a diameter of one to a few tens of nm.

The device according to the invention comprises a plurality of filaments, preferentially organized according to a dense network.

The filament 15, which takes root at 16, comprises an end 17, and extends over the entire distance between the pore outlet and the second metal electrode 13. The growth process is such that the obtained filament is semiconductor.

 In the device according to the invention, there is no direct contact between the filament 15, including its end 17 and the second electrode 13.

The passage of electrical charges between the filament and the second electrode takes place via the metal nanoparticle 18. In fact, in the device according to the invention, the end 17 of the filament 15 is mounted in electrical contact with at least one metal nanoparticle 18, which itself is mounted in electrical contact with the electrode 13.

In general, the term "end of the filament" is understood to mean the last few tens of nm of this filament.

Typically, the metal nanoparticle has a size of between a few nm and a few tens of nm, and its diameter must be close to the diameter of the filament.

 The metal of the nanoparticle is chosen so that the electrical contact between the end 17 of the filament 15 and the nanoparticle 18 is a semiconductor / metal contact of the rectifier type, that is to say that there is a barrier of energy for the passage of carriers between the filament 15 and the nanoparticle 18. A potential barrier of this type is commonly called Schottky barrier. According to a first variant described in FIG. 1, the pores 12 are oriented substantially parallel to the insulating substrate 14 and the second electrode 13 is also deposited on the substrate 14.

 Preferably, the material of the second electrode 13 is identical to that of the first part of the first electrode 10, in order to simplify the method of producing the device, the two electrodes being able to be deposited simultaneously. Thus the contact with the second electrode is facilitated. The lateral geometry of the pore network also makes it possible to control the density of filaments by controlling the characteristics of the porous membrane.

According to a second variant described in FIG. 2, the pores 12 are oriented perpendicularly to the substrate 14. The second electrode 13 comprises a metallic film 20 transferred onto the filaments and in electrical contact with the metal particles 18.

FIG. 3 depicts the front energy band diagram (FIG. 3a) and after (FIG. 3b) the contacting of the filament 15 with the electrode 9 at the level of the root 16 and secondly with the metal particle 18 at its end 17. In FIG. 3a, the line 30 corresponds to the vacuum level, the line 31 corresponds to the maximum energy that an equilibrium electron can have in the metal of the first electrode 9, the line 33 corresponds to the maximum energy that can have an equilibrium electron in the metal of the metal nanoparticle 18, the line 32 corresponds to the Fermi level of the filament 15. Φ1 represents the work of the metal of the electrode 9, Φ ₀ ΝΤ the work of the output of the filament 15 and Φρ the output work of the metal of the nanoparticle 18. FIG. 3b describes the thermodynamic equilibrium energy band diagram when there is an electrical contact between, on the one hand, the filament 15 and the electrode 9 at the root 16 and secondly the filament 15 and the metal nanoparticle 18 at its end 17. The level of Fermi 34 is constant throughout the structure. The curvature of the band diagram is represented by the contacts on the electrode 9 / filament 15 / nanoparticle 18 assembly. X. _C NT represents the electronic affinity of the semiconductor filament 15. The electronic affinity of a semiconductor is defines as the energy to bring to a free carrier (bottom of the conduction band) to tear it to the semiconductor.

Between the filament 15 and the nanoparticle 18, there is the characteristic diagram of a semiconductor junction / Schottky type metal. Vs, referred to as the surface potential, is the barrier that a carrier must cross to pass from the semiconductor filament to the metal of the nanoparticle 18. Ob is the barrier that the carrier must pass to pass from the metal of the nanoparticle 18 to the semiconductor of the nanoparticle 18. filament 15. Ob is commonly referred to as Schottky barrier.

 Preferably, the output work of the filament is at least 0.2 eV less than the work output of the metal of the nanoparticle. An example of a possible structure is:

Electrode 9 in aluminum (Φ1 = 4.06-4.26 eV) / electronic affinity of a carbon nanotube of about 4.5 eV / nickel metal nanoparticle 18 (Φρ = 5.04-5.35 eV). Preferably, the work output of the metal of the nanoparticle 18 is greater than 5 eV, for example between 5 eV and 6 eV, which is the case for example of metals such as nickel, cobalt, palladium and platinum . Indeed, this value makes it possible to obtain a difference of at least 0.2 eV with the Fermi level of the filament.

Advantageously, the metal of the nanoparticle 18 comprises nickel.

Nickel particles are used as a catalytic element in a known method of filament growth, vapor phase growth. The nickel particle is located at the end of the filament at the end of growth. The manufacturing process of the device is thus simplified in this case.

 According to one variant, other metallic particles may be obtained in the device according to the invention, by a "decoration" process carried out subsequent to the growth of the filaments described below.

 The metal particles are, at the end of this process, fixed on the defects of the filament, these defects being distributed over the entire length of the filament. Some of them will attach to the end of the filament, because of the existence of bonds hanging there and thus become the particles 48 realizing the Schotky barrier as shown in FIG.

 Thus the metal particles 48 attached to the end of the filament perform the electrical rectifier contact in place of the metal particle 18 from the catalyst.

 The "decoration" metal is integrated into the device after having achieved the growth of the filaments. The growth phase and the metallization phase are decoupled. It is thus possible to choose a metal from a wider range than when it is imposed by the nature of the particle which served as a catalytic element.

 One advantage is the possibility of varying the work output of the metal particles. For example for rectifier or photoswitch operation, the leakage current (or darkness) is lowered by increasing the height of the Schottky barrier.

In another example, for photopile operation, increasing the value of the Schottky barrier increases the charge separation efficiency, which increases the efficiency of the photocell. Conventionally, a Schottky type barrier makes it possible, when a potential difference is applied, to separate the electron / hole pairs, the carrier density increases close to the barrier and the tunneling probability of tunneling increases. We then obtain a Schottky diode. The device according to the invention therefore behaves as a Schottky diode when a potential difference is applied between the first part of the first electrode 10 and the second electrode 13. When filaments 15 are arranged so as to be able to receive a light radiation 50 As illustrated in FIG. 5a for the first horizontal variant and FIG. 5b for the second vertical variant, electron / hole pairs are generated by photon absorption, due to the photoconductivity of the filaments, s-CNT or s-nW. The existence of a rectifying contact at one end of the filament allows the device according to the invention to generate a photocurrent under illumination. Thus a device according to the invention is optoelectronic when the filaments are arranged to be able to receive light radiation.

For the vertical variant described in FIG. 5b, the metal film 20 is then chosen optically transparent, such as for example a graphene thin film. Thus, the surface capable of receiving illumination is maximized, which allows a large production of energy.

When the filament 15 joins the metal of the first electrode 9 and a junction with the metal of the nanoparticle 18, the filament band diagram curves. The output work of the metal of the first electrode Φ1 is placed close to the conduction band of the filament, and the output work of the metal of the nanoparticle Φ _Ρ is placed near the valence band of the filament.

When a forward voltage is applied between the two electrodes, the Schottky barrier decreases as the voltage increases, and the current flows freely.

When a reverse voltage ("reverse" in English) is applied between the two electrodes, the Schottky barrier increases when the voltage increases. The device is in the off state and the current does not flow. FIG. 6 illustrates an example of an electrical characteristic, corresponding to the intensity flowing in the device as a function of the voltage applied between the two electrodes, called characteristic IV, for an electronic device according to the first variant of the invention with CNT.

 The device was made according to a method described below. In this example the first part 10 of the electrode 9 and the electrode 13 are made of aluminum, with a thickness of 200 nm. The second part of the electrode 1 1 porous alumina, and the nickel metal nanoparticle 18.

The nanotubes have a diameter of about ten nm, the distance separating the two electrodes is about 100 μιτι.

The first part of the electrode 10 is connected to ground and a potential V is applied to the second electrode 13.

The characteristic IV of FIG. 6, illustrated by the curve 51, corresponds to the current state when a negative V potential is applied to the metal of the nanoparticle via the second electrode 13. A Schottky diode type operation is therefore obtained for the device according to the invention, to which an electric voltage is applied between the first portion 10 of the electrode 9 and the electrode 13.

Operation of the same type can of course be observed for any metal of the metal nanoparticle capable of making a rectifier type contact between the semiconductor filament 15 and the nanoparticle 18.

FIGS. 7a to 7d illustrate examples of characteristics IV of the optoelectronic device according to the first variant of the invention with s-CNTs as described above, a curve corresponding to the characteristic of the device without illumination and a curve with the characteristic of the same. device with illumination by light radiation. In these examples, the device and more particularly the s-CNT are illuminated by a laser diode emitting in the infrared light around the wavelength λ equal to 850 nm.

FIG. 7a corresponds to the characteristic obtained with the metal particle at the nickel end, resulting from the growth process starting from a catalytic nickel element. In FIG. 7a the curve 70 corresponds to the characteristic IV without illumination and the curve 71 corresponds to the characteristic IV under illumination. For a negative voltage, the curves 70 and 71 are identical, no photonic response is observed. On the other hand, for a positive voltage V, a current flows in the device under illumination. This current corresponds to a photogenerated current by illumination. Thus, the device according to the invention is capable of generating a photocurrent under illumination. Figs. 7b to 7d illustrate three further examples of device characteristic IV of s-CNT. In these examples, metal particles 18 forming the rectifying contact were positioned at the end of the nanotube by means of a decoration process described later, subsequent to the growth of the nanotubes.

In Figure 7b, the metal of particle 18 is platinum, working output 5.12 eV-4.93 eV. Curve 74 clearly shows the generation of a photocurrent, as for the case of nickel on the curve of FIG. 7a. FIGS. 7c and 7d show the characteristics l-V respectively corresponding to a particle 18 made of iron (7c), working output 4.67 eV-4.81 eV, and tin (7d), working output 4.42 eV. In the latter two cases, the output work is very close to that of the carbon nanotube. There is no potential barrier, so no carrier separation, and therefore no rectifier operation. There is no generation of photocurrent, the curves without and with illumination are confused.

 Thus, the device according to the invention to which a voltage is applied between its electrodes shows a diode-type electronic operation without illumination and an optoelectronic operation of photodiode type under illumination.

More generally, the device according to the invention has a photocurrent response under illumination, and thus constitutes an elementary brick adapted to be integrated in different types of optoelectronic components such as photodiode, photocell, photodetector or photocututor to realize optoelectronic assemblies. According to one example, the device according to the invention is in blocked mode without illumination and becomes on when illuminated.

 In another example, a photocell incorporating a device according to the invention is connected so as to charge a capacitance for producing a photodetector.

 According to another example, a device according to the invention is combined with a source of intensity-adjustable illumination to produce a photoc switch.

 According to another example, a device according to the invention is combined with a scintillator to produce a particle detector. Due to the speed of operation of the device, the photons are detected with greater precision.

FIG. 8 describes an example of variation 81 of the current I flowing in a device according to the invention as a function of the illumination received, as a function of the applied voltage V. When no voltage is applied (V = 0), the current is non-zero: a photocurrent is generated by the device. The device according to the invention is capable of generating a current when it is illuminated, it operates in this case in a photocell.

FIG. 9 describes an example of a variation of the responsivity ("responsivity") of a device according to the invention, expressed in Ampere per Watt, as a function of the power of the illumination laser. The responsivity is equal to the variation of the response of the device in relation to the variation of the power of the laser.

 The applied voltage is in this example 4V, but the device can of course operate with a zero applied voltage.

The points marked with a cross correspond to an illumination laser emitting in the blue at 458 nm, the points marked with circles correspond to an illumination laser emitting in the green at 514 nm, the points marked with triangles correspond to a laser emitting in the red at 633 nm.

The responsivity of the device decreases when the power of the laser increases. Responsibilities up to 10 ⁴ A / W are obtained with a device according to the invention. The device according to the invention has a wide sensitivity throughout the visible spectrum. Figure 10 shows the normalized responsivity versus frequency of an illuminating laser diode emitting around 850 nm. It is found that the responsivity decreases relatively little when the frequency of the laser increases. For a photoc switch application, the device according to the invention is therefore able to operate at a high frequency. In this example the frequency of the laser pulses is limited to 100 KHz, but operation at higher frequencies is possible.

Another aspect of the invention is a method of manufacturing an electronic or optoelectronic device comprising nanostructures in filaments.

 The method of manufacturing an electronic device according to the invention is represented in FIGS. 11 and 12 and comprises:

 the production of a first electrode 9 deposited on a substrate made of insulating material 14, comprising a first part 10 comprising a metallic material and a second porous part 11 comprising a network of substantially aligned pores 12,

 the production of a second electrode 13 comprising a metallic material,

the production of a plurality of semiconductor filaments comprising a step of depositing catalytic elements inside the pores, the elements comprising at least one nanoparticle 18 comprising a metallic material, then a phase of vapor phase growth filaments from the catalytic elements, a filament 15 comprising a root 16 disposed in a pore 12 and in electrical contact with the first part 10, and an end 17 in electrical contact with the nanoparticle 18, the growth being carried out until that the nanoparticle 18 is in electrical contact with the second electrode 13, avoiding any direct electrical contact between the filament 15 and the second electrode 13, the contact between the end 17 of the filament 15 and the nanoparticle 18 being of type rectifier so that there is an energy barrier for the passage of carriers between the filament 15 and the nanoparticle 18.

Advantageously, the first electrode 9 is made from an aluminum layer in which the second porous part 11 is produced by an anodic oxidation process of aluminum.

 According to one variant, the semiconductor filament is a carbon nanotube (s-CNT).

 According to another variant, the semiconductor filament is a semiconductor nanowire (s-nW), for example silicon or germanium.

Advantageously, the dense network of semiconductor filaments is obtained by "chemical vapor deposition" or CVD type growth in nanoporous matrices. The CVD growth of nanotubes or semiconductor nanowires is known and controlled, and is based on the use of catalysts (for example Fe, Co, Ni for CNT or Au, Cu, Ni, Al for s-NW) in the form of of nanoparticles.

 Thus the method according to the invention is based on a controlled technology for producing a dense network of filaments.

In the case of growth that starts in nanopores, the confined nature of the pores induces that at the end of growth the nanoparticles of the catalyst are necessarily found at the end of the CNT / s-NW.

The end of the growth step according to the invention is carried out by avoiding any direct electrical contact between the filament 15, including its end 17 and the second electrode 13.

 Preferably, an encapsulation layer is deposited on the second electrode in order to avoid obtaining filaments some of which is in direct contact with the upper face of the electrode.

The time of the growth step according to the invention is optimized to prevent the filaments bending in contact with the second electrode and thus a portion of the filament is in direct contact with the second electrode 13. An example of a first variant of the process for manufacturing an electronic or optoelectronic device based on filamentous nanostructures according to the invention using two aluminum-based electrodes is described below and illustrated by FIGS. 1 1 f, which show the main stages. This variant is called horizontal variant.

 According to a first step illustrated in FIG. 11a, an aluminum layer 101 is deposited on an insulating substrate 14. The thickness is from a few tens of nanometers to a few micrometers depending on the desired application. Indeed, the thickness in conjunction with the width will mainly define the number of nanofilaments and therefore the value of the electric current that can be transported by the device.

As illustrated in FIG. 11b, a lithography of the aluminum layer is carried out so as to define a first electrode 9 and a second electrode 13.

According to the first horizontal variant, the two electrodes preferably each have a vertical face facing each other. The distance between the two electrodes is for example between 1 μιτι and 5 μιτι.

 According to this example, the two aluminum electrodes are advantageously made simultaneously in the same step.

 As illustrated in FIG. 11c, an encapsulation layer 102 is deposited,

103 over respectively the first and the second electrode 9, 13.

The encapsulation layer comprises for example AI ₂ 0 ₃ .

 As illustrated in FIG. 11, the first electrode 9 is made by transforming a portion 11 of the first electrode 9 into a pore network.

12 substantially aligned, the other portion 10 of the electrode being retained.

Advantageously, a known anodic oxidation process is used (FR

2888041 - US2009-0035908).

 According to the first horizontal variant, the pores 12 are made to be oriented substantially parallel to the plane of the substrate 14.

As illustrated in FIG. 1, catalytic elements are deposited inside said pores by an electrochemical process. In a pore, these catalytic elements are composed of at least one nanoparticle 18 comprising a metallic material such as nickel or cobalt. As illustrated in FIG. 1, filament growth is achieved by a chemical vapor deposition (CVD) technique from the catalytic elements.

The filaments comprise a root 16 in a pore 12 in electrical contact with the first portion 10 of the first electrode 9 and an end 17 in electrical contact with a nanoparticle 18, which is present at the end of the filament during the growth phase, because of the confined nature of the latter, and which corresponds to the remaining part of the catalytic element having initiated the growth.

Due to the pore orientation, the growth is generally parallel to the plane of the substrate.

 The growth of the filaments takes place over a sufficient length until the nanoparticle 18 is in electrical contact with the second electrode 13.

According to the first horizontal variant of the method, the electrical contact is preferably effected with the vertical face of the second electrode 13 facing the first electrode 9. The encapsulation layer 103 makes it possible to prevent the contact between a filament 15 and the second electrode 13 is performed without passing through the nanoparticle 18, which would disrupt the operation of the device by increasing its leakage current.

 According to the state of the art, the contact is effected by evaporation of a metal layer which "buries" the ends of the CNTs or nanowires, thus making a direct contact between the terminal part or the end of the filament and the second electrode.

The materials respectively of the filament 15 and the nanoparticle 18 are chosen so that the contact between the end 17 of the filament 15 and the said nanoparticle 18 is of the rectifier type, so that there is an energy barrier for the passage of carriers. between the filament 15 and the particle 18. Thus, an asymmetrical junction is formed between the two electrodes.

The horizontal variant of the process is easy to integrate according to current microelectronic technologies which are all planar type. Advantageously, for the production of optoelectronic components, the intrinsically photoconductive semiconductor filaments are arranged so as to be able to receive light radiation. An example of a second variant of the process for manufacturing an electronic or optoelectronic device based on filament nanostructures according to the invention using an aluminum-based electrode is described below and illustrated by FIGS. 12a to 12e. which show the main stages. This variant is called vertical variant.

According to a first step illustrated in FIG. 11a, an aluminum layer 1 10 is deposited on an insulating substrate 14. Typically, the layer has a thickness of between a few hundred nm and 2 μιη.

As illustrated in FIG. 12b, the first electrode 9 is made by transforming a part 11 of the first electrode 9 into a network of substantially aligned pores 12, the other part of the electrode being retained. Advantageously, a known anodic oxidation process is used (FR 2888041 - US2009-0035908).

 According to the second vertical variant, the pores 12 are made to be oriented substantially perpendicular to the plane of the substrate 14.

As illustrated in FIG. 12c, catalytic elements are deposited inside said pores by an electrochemical process. In a pore, these catalytic elements are composed of at least one nanoparticle 18 comprising a metallic material such as nickel or cobalt.

As illustrated in FIG. 12d, filament growth is achieved by a chemical vapor deposition (CVD) technique from the catalytic elements. The filaments comprise a root 16 in a pore 12 in electrical contact with the first portion 10 of the first electrode 9 and an end 17 in electrical contact with a nanoparticle 18, which is present at the end of the filament during the growth phase and which corresponds to the remaining part of the catalytic element having initiated the growth.

Due to the orientation of the pores, the growth of the nanotubes is generally perpendicular to the plane of the substrate: the filaments stand perpendicular to the substrate. The growth time is optimized to avoid a loss of directivity of the growth of the filaments once out of the pores. Advantageously, the length of the emerging portion of the filaments at the pore outlet is between a few hundred nm and 2 μιη.

The step of producing the second electrode is illustrated in FIG. 12e and comprises the transfer of a metal film 20 onto the nanoparticles 18 in contact with the ends of the filaments. The carry is such that it ensures electrical contact between the nanoparticle 18 and the metal film 20, avoiding any direct electrical contact between the filament 15 and the metal film 20.

 For example, the metal film is transferred by being disposed on two insulating supports 120, 121 located on either side of the filaments.

The materials respectively of the filament 15 and the nanoparticle 18 are chosen so that the contact between the end 17 of the filament 15 and the said nanoparticle 18 is of the rectifier type, so that there is an energy barrier for the passage of carriers. between the filament 15 and the particle 18. Thus, an asymmetrical junction is formed between the two electrodes.

Advantageously, for producing optoelectronic components, the metal film is transparent, so that the semiconductor and photoconductive filaments can receive light radiation. Advantageously, the metal film comprises at least one thin layer of graphene. An advantage of the vertical process is a large dimension of the active surface, suitable for photovoltaic applications.

Conventionally, metallization of a small portion of the metal film is then carried out in order to be able to take contacts, so as to be able to apply an electrical voltage across the electrodes, as shown in FIG. 13.

The innovative structure of the electrodes and filaments, combined with the control of the properties and density of the filaments as well as the control of the rectifying contact between the filament and the nanoparticle thanks to the process according to the invention makes it possible, in a reproducible manner, to produce electronic and optoelectronic components as represented in FIGS. 1 and 2.

According to one embodiment, the filaments are semiconducting carbon nanotubes and the metal nanoparticles are nickel or cobalt.

 According to another embodiment, the filaments are silicon or germanium nanowires, the metal nanoparticles are gold, nickel or copper.

According to an optional variant of the method according to the invention, a filament decoration step is carried out after the growth step. This step consists in fixing metal particles on defects and at the ends 17 of the filaments 15. The metal particles 48 fixed at the ends are then in electrical contact with the end 17 of the filament on the one hand and with the second electrode 1 3 d on the other hand, in place of the metal particle 18 from the catalyst.

Preferably the decoration step is performed by electro-filing.

An advantage of the decorating step is the establishment of an electrical charge passage between filament ends and the second electrode via the particles 48, for ends that were not initially in contact with the electrode via the particle 18, for example because of a filament too short.

 In addition, the decorative metal can be chosen from a wider variety of metals than those capable of carrying out a catalyst function. Thus, it is possible to vary the value of the work of output of the metal particles making the contact rectifier.

 In addition, the "decoration" metal being integrated in the device after making the growth of the filaments, the growth phase and the metallization phase are decoupled.

Claims

1. Electronic device comprising:  a first electrode (9) comprising a first part (10) comprising a metallic material and a second porous part (1 1) comprising a substantially aligned network of pores (12), said first electrode (9) being deposited on a substrate in insulating material (14),  a second electrode (13) comprising a metallic material,  a plurality of semiconductor filaments (15), a filament (15) comprising a root (16) and an end (17),  said root (16) being disposed in a pore (1 2) and in electrical contact with said first portion (10), ^* Said end (17) of a filament (15) being mounted in electrical contact with at least one nanoparticle (18,48) comprising a metallic material,  said nanoparticle (18,48) being mounted in electrical contact with said second electrode (13),  the electrical contact between said end (17) of the filament (15) and said nanoparticle (18, 48) being of the rectifier type so that there exists an energy barrier for the passage of carriers between said filament (15) and said at least one particle (18),  said filament (15) being without direct electrical contact with said second electrode (13). The device of claim 1 wherein said first portion of said first electrode comprises aluminum and said second portion of said first electrode comprises porous alumina. 3. Device according to one of the preceding claims wherein said pores are oriented substantially parallel to the plane of the substrate and said second electrode (13) is also deposited on said substrate (14). 4. Device according to one of claims 1 or 2 wherein said pores are oriented substantially vertically to the plane of the substrate and said second electrode (13) comprises a metal film (20) carried on said filaments (15). 5. Device according to one of the preceding claims wherein the metal of said at least one metal nanoparticle (18) has an output work greater than 5 eV. 6. Device according to one of the preceding claims wherein the metal of said at least one metal nanoparticle (18) comprises nickel. 7. Device according to one of the preceding claims further comprising a plurality of metal particles fixed on the wall and on said end (17) of said filaments (15), so that at least one of said particles (48) attached to said end (17) performs said rectifier type electrical contact. 8. Diode Schottky type comprising a device according to one of the preceding claims and means adapted to apply a potential difference between said first portion of said first electrode and said second electrode. 9. Optoelectronic device according to one of the preceding claims wherein said filaments are arranged to be able to receive light radiation. Optoelectronic assembly comprising at least one photodiode and / or a photocell and / or a photodetector and / or a photoc switch incorporating at least one optoelectronic device according to claim 9. 1 1. Particle detector comprising a scintillator and an optoelectronic device according to claim 9. 12. A method of manufacturing an electronic device comprising: the production of a first electrode (9) deposited on a substrate of insulating material (14), comprising a first part (10) comprising a metal material and a second porous portion (1 1) comprising a substantially aligned network of pores (12),  the production of a second electrode (13) comprising a metallic material, the production of a plurality of semiconductor filaments (15) comprising: ^* A step of depositing catalytic elements within said pores, said elements comprising at least one nanoparticle (18) comprising a metallic material,  then a step of vapor phase growth of said filaments from said catalytic elements, a filament (15) comprising a root (16) disposed in a pore (12) and in electrical contact with said first portion (10), and an end (17) in electrical contact with said nanoparticle (18),  the growth being carried out until said nanoparticle (18) is in electrical contact with said second electrode (13), avoiding any direct electrical contact between said filament (15) and said second electrode (13), ^* The contact between said tip (17) of filament (15) and said nanoparticle (18) rectifier type being so that there is an energy barrier for carriers passage between said filament (15) and said at least one nanoparticle (18). 13. The method of claim 12 wherein said first electrode (9) is made from an aluminum layer wherein said second porous portion (1 1) is made by an anodic oxidation process of aluminum. 14. The method of claim 12 or 13 wherein said second electrode (13) is made prior to said growth step and on the same substrate (14), said second electrode (13) having a substantially vertical face facing a substantially vertical face of said first electrode comprising said pores (12), and wherein the realization of said pores of said first electrode is performed so as to obtain pores oriented substantially parallel to the plane of said substrate (14), said growth being effected such that filaments extend substantially parallel to said substrate (14) over a sufficient length to reach said facing face. 15. The method of claim 12 or 13 wherein the production of said pores of said first electrode is performed so as to obtain pores oriented substantially perpendicular to the plane of said substrate, said growth being effected so that said filaments stand vertically, and wherein the realization of said second electrode (13) comprises the transfer of a metal film (20) on said nanoparticles (18). The method of claim 15 wherein the metal film comprises a graphene thin film. 17. Method according to one of claims 12 to 16 further comprising a step of decorating said filaments produced after the growth step, wherein metal particles are fixed on defects and at the ends (17) of said filaments (15). ), so that at least one of said particles (48) attached to said end (17) makes said rectifier type electrical contact.