Field of the Invention

This invention relates in general to the field of photovoltaics. More particularly the invention relates to a device for conversion between electromagnetic radiation and electric current, said device comprising a first and a second crystalline or poly- crystalline material, and a first and a second electrical collecting member.

Background of the Invention The conventional photovoltaic cell of today is based on a thin- film pn-diode of silicon. This is used for photon harvesting, i.e. for example in solar cells and photon detectors. The pn-diode based photovoltaic cell is a so-called single-junction device. The photon absorption takes place at a single layer, i.e. single pn-junction. A p-n junction is a junction formed by combining p-type and n-type semiconductor materials. This limits the efficiency of the photovoltaic cell, when using it for solar cells, as a result of the broad solar spectrum and the single band gap energy.

Figure 1 illustrates a pn-junction-based photovoltaic cell and it shows the energy behavior and the physical construction of such a nanowire-based photovoltaic cell. The left-hand-side of Fig. 1 shows the operation principle of a photovoltaic cell with a traditional/conventional pn-junction.

Utilizing so-called planar multi-junction or tandem cells may increase the efficiency of solar cells. These cells are composed of several layers of epitaxially deposited films. By combining several planar layers of various alloys of III -V compound semiconductors the absorption spectrum of the device may be broadened. The band-gap energy of each layer can then be tuned to absorb a specific band of the solar spectrum and the total absorption spectrum of the device will, thereby, be much broader than that of a single-junction device.

The ability to optimize the respective band gaps of the various junctions is, though, hampered by the requirement that each layer must be lattice matched or nearly matched to all other layers.

Additional possibilities of tuning the band gap energy, and thereby also tuning the absorption spectrum, can be introduced to the photovoltaic cell by combining the epitaxial films with an increased quantum confinement of the charge carriers. This has been done e.g. in photovoltaic cells based on quantum wires (nanowires) and quantum dots. The work on nanowire-based photovoltaic cells has so far been based on the conventional pn-junction. Recently, this junction has been introduced into nanowires, See e.g. J. Am. Chem. Soc, vol 130, 9224 (2008).

Fabrication techniques based on self-organized growth are well known and well controlled on an academic research level. Several larger companies are currently also performing extensive research on the commercialization of nanowire technologies.

The efficiency of currently available commercial Si solar cells is between 5% (amorphous Si) and 17% (monocrystalline Si). This is far from the theoretical maximum efficiency of 28%, which even that is fairly low. The highest reported efficiency of a solar cell is at the moment just above 40%. Even this value is not very high, although it was achieved using a very complex planar multi-junction device.

New device strategies will be needed in order to increase the light-to-electrical- current conversion efficiency of solar cells. The current high-efficiency technologies are very expensive and, thereby, affordable only in few environments and devices. The price of these devices reflects the difficulties in the fabrication and device optimization. It has proven to be hard to combine materials with optimal band gap energies into the planar multi-junction cells without creating strain-induced dislocations.

The sensitivity is typically an important characteristic of any photon detector. Most photovoltaic cells, based on pn-junctions, are not very sensitive. A relatively large photon current is needed to obtain a good signal-to-noise ratio. Currently studied nanowire-based photovoltaic cells are commonly based on doped nanowires with a built-in pn-junction. However, when doping a nanowire there is a risk that the doping affects the growth and thereby also the quality of the nanowire.

The doping technology of wires is presently under development for III-V materials, e.g. GaAs, with p-type doping being more challenging. M.J. Romero and M.M. Al-Jassim, Journal of Applied Physics 93, 626 (2003) discloses the use of piezoelectricity in solar cells, wherein the solar cells comprises planar layers of strained piezoelectric material inserted in a pn-junction in order to optimize and increase the cell efficiency. However problems of such solar cells are that pn-junctions are needed and that several material interfaces are present in the transport way for the photon and/or electron to pass, which may increase resistance in the cell, and thus lowering the efficiency of the cell. Also, the manufacturing method for such layering of materials is time and fund consuming.

Hence, there is a need for a device suitable for converting electromagnetic radiation into an electric current, which device does not require doping or only a limited amount of doping. Summary of the Invention

Accordingly, the present invention seeks to mitigate, alleviate or eliminate one or more of the above-identified deficiencies and to provide a device for conversion between electromagnetic radiation and electric current, said device comprising: a first crystalline or poly- crystalline material, wherein said first material is a semiconductor with piezoelectric property, said first material having at least two ends; a second crystalline or poly-crystalline material; said first and second materials being connected, such that a strain in the first material is obtained; and said first material being conductively connected to a first and a second collecting/providing means at said at least two ends, respectively, for collection or provision of electric current between said at least two ends.

The device according to some embodiments allows for a large tunability of the band gap energies and the field strength of the active region, in comparison with conventional photovoltaic cells. This enables also tuning of the absorption spectrum.

The device according to some embodiments may be used, by combining different materials in either radial or axial heterostructures, to create photovoltaic cells absorbing several different spectral bands.

The device according to some embodiments does not have to be processed on III-V compound semiconductor wafers. The photovoltaic cells based on piezoelectric nanowires can probably be made even on cheap and low-quality Si wafers. This will considerably reduce the cost of a device, in particular in comparison with planar multi- junction devices of III-V compound semiconductors.

The device according to some embodiments does not require doping in order to create an electric field in the active region. The fabrication of the device is, thereby, easier than that of devices based on nanowires with pn-junctions.

The device according to some embodiments offer a very good photosensitivity since it offers a control of the electron transport at a single-electron level and the light absorption/emission at a single photon level. The strain engineering technique according to some embodiments may be used to create polarization dependent photon detectors.

Brief Description of the Drawings

These and other aspects, features and advantages of which the invention is capable of will be apparent and elucidated from the following description of embodiments of the present invention, reference being made to the accompanying drawings, in which

Figure 1 is an illustration of a traditional photovoltaic cell;

Figure 2a is an illustration of a core-shell geometry; Figure 2b is a schematic illustration of a true core shape;

Figure 2c is another schematic illustration of a true core shape;

Figure 3 a to 3 c illustrates a device according to an embodiment, respectively;

Figure 4 is an illustration showing the energy characteristics of a nanowire 11 according to an embodiment; Figure 5 illustrates very schematically the stresses in the device of Fig. 2a;

Figure 6 illustrates the elastic strain, piezoelectric potential and band edges in a nanowire according to an embodiment of the invention;

Figure 7a illustrates a device comprising three crystalline semiconducting materials according to an embodiment of the invention; Figure 7b illustrates a device comprising more than one shell according to an embodiment of the invention;

Figure 7c illustrates a device comprising an axial heterostructure according to an embodiment of the invention;

Figure 8 illustrates theoretical field strength in various material combinations (zinc blende crystalline materials) in a [111] -oriented nanowire with core-shell according to an embodiment;

Figures 9a to 9e illustrate a manufacturing method according to an embodiment of the present invention, wherein nanowires are grown using catalyst particles;

Figures 10a to 1Of illustrate a manufacturing method according to another embodiment of the present invention, wherein nanowires are grown using a mask;

Figure 11 illustrates a sheet geometry according to one embodiment of the present invention;

Figure 12 depicts the photovoltaic mechanism in strained core-shell nanowires and the schematic band diagram in a strained core-shell nanowire between two metallic contacts, according to one embodiment of the present invention;

Figures 13a to 13f show the elastic strain on a cross section of embodiments of a core-shell nanowires;

Figure 14 shows the transversal component (x/y component) of the piezoelectric field at a cross section of an InAs/InP core-shell nanowire, according to one embodiment of the present invention; Figure 15 shows the piezoelectric field in a QD WZ structure, according to one embodiment of the present invention;

Figure 16a shows schematically a core shell nanowire which is aligned along a [111] direction of the zinc blende crystal direction or along the [0001] direction of a wurtzite crystal, according to one embodiment of the present invention;

Figure 16b shows schematically an axial heterostructure super lattice of InAs and InP segments in a nanowire, according to one embodiment of the present invention;

Figure 16c shows an epitaxially strained InAs quantum dot in an InP nanowire, according to one embodiment of the present invention; Figure 17a shows schematically the symmetry and principal crystal axes of the

[111] oriented zinc blende crystal, according to one embodiment of the present invention;

Figure 17b shows schematically the symmetry and principal crystal axes of the [0001] oriented wurtzite material with respect to the nanowire geometry, according to one embodiment of the present invention;

Figures 18a and 18b show the piezoelectric potential, in terms of isosurfaces for selected values of the potential, in a ZB and a WZ phase core-shell heterostructure, according to one embodiment of the present invention;

Figure 19a and 19b show the piezoelectric potential in an axial superlattice heterostructure, in terms of isosurfaces, according to one embodiment of the present invention; and

Figure 20a and 20b show the piezoelectric potential using isosurfaces in a ZB and WZ QD heterostructure, according to one embodiment of the present invention.

Description of embodiments

Several embodiments of the present invention will be described in more detail below with reference to the accompanying drawings in order for those skilled in the art to be able to carry out the invention. The invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. The embodiments do not limit the invention, but the invention is only limited by the appended patent claims. Furthermore, the terminology used in the detailed description of the particular embodiments illustrated in the accompanying drawings is not intended to be limiting of the invention. The following description focuses on embodiments of the present invention applicable to the field of photo voltaics, and in particular to a photovoltaic cell.

The present embodiments provide a device with a built-in piezoelectric field. The device may be an electronic semiconductor device. The device is used for photon harvesting, whereby a built-in piezoelectric field in the device replaces or assists the electric field of a pn-junction, which may be used in conventional photovoltaic cells. Photon harvesting is meant to be interpreted as the conversion of a photon to an electron-hole pair. The electron-hole pair is then split into an electron and a hole by steering or urging the electrons one way and the holes in another way. The piezoelectric field is induced by epitaxially connecting at least a first material and a second material, said first and second materials being of crystalline or poly-crystalline materials. At least one of the materials is a semiconductor with piezoelectric property. The first and the second material are joined so that a strained material interface is formed and the involved materials are strained. The first and the second material may be of different crystal phases, e.g. zinc blende and wurtzite. The first and the second material may be of the same crystal phase but have different lattice constants.

The electric field of the piezoelectric device in general has a non-zero axial component. This implies that there will be an electric potential difference between two ends of the first and/or the second material, i.e. the crystalline or poly-crystaline, semiconducting material with piezoelectric property. Since the two materials are electrically connected, the same electric potential difference is present in both materials.

In general, the matching of the first and second material may be performed by self-assembly of atoms. In an embodiment, wherein the first material is arranged as a core within a tubular shell of a second material, the core of a first material may for example be grown with a regular and homogeneous crystal structure. When growing the shell of a second material around the core, the shell material may be introduced in the vapor phase.

In metalorganic vapor phase epitaxy the growth material may be supplied by metalorganic (group III material) and hybrid precursors (group V material) in vapor phase. The precursors are compounds that decompose when heated, to elemental group III and V material. For example GaAs can be grown from trimethylgallium, (CH ₃ ) ₃ Ga ₅ and arsine, ASH ₃ . Further examples of precursors are: (CH ₃ ) ₃ A1, (C ₂ Hs) ₃ Al, (CH ₃ ) ₃ Ga, (C ₂ Hs) ₃ Ga, (CHs) ₃ In, (C ₂ Hj) ₃ In, NH ₃ , (CH ₃ ) ₂ N ₂ H ₄ , (C ₄ Hg) ₄ N ₂ H ₄ , (C ₄ Hg) ₄ PH ₃ , PH ₃ , (C ₄ Hg) ₄ AsH ₃ , AsH ₃ , (CH ₃ ) ₃ Sb, (C ₂ Hj) ₃ Sb, (SiMe ₃ ) ₃ Sb, (CH ₃ ) ₂ Cd, (C ₂ Hj) ₂ Cd, (CHs) ₂ Te, (C ₂ Hs) ₂ Te, (C ₃ Hy) ₂ Te, H ₂ Se, (CH ₃ ) ₂ Se, (C ₃ Hy) ₂ Se, (CH ₃ ) ₂ Zn, (C ₂ Hj) ₂ Zn, [t-

Bu ₂ GaSb(SiMe ₃ ) ₂ ] ₂ , [Cl ₂ GaSb(t-Bu) ₂ ] ₃ , [Me ₂ GaSb(t-Bu) ₂ ] ₃ , [Me ₂ InSb(t-Bu) ₂ ] ₃ , Si ₂ H ₆ , (CH ₃ ) ₂ S ₂ , (C ₂ Hs) ₂ S ₂ , H ₂ S, (C ₄ Hg) ₄ SH, (CH ₃ )SeH, (C ₄ Hg) ₄ SeH, MASe, DASe, and (CH ₃ )SeH. Using e.g. (CH ₃ ) ₃ Ga, (CH ₃ ) ₃ In, (CH ₃ ) ₃ A1, AsH ₃ , and PH ₃ any alloy of In, Ga, Al and As, P can be grown. The atoms of the second material in this vapor attach to the surface of the core crystal of the first material. The atoms of the second material then form chemical bonds with the core atoms of the first material. This creates also a material interface (heterojunction). The single layer of shell atoms of the second material can be formed on the core crystal of the first material, even if the natural crystalline structure of the shell material has another lattice constant than the core crystal has.

The growth of strained epitaxial nanowire structures is a stochastic process and proceeds roughly as follows: Once a first monolayer of shell atoms of the second material is formed on the core of the first material, the next atoms attach themselves and bond to the outermost atoms, i.e. the latest deposited shell atoms. A crystal structure of shell atoms is thereby successively created onto the core crystal despite possible differences in terms of lattice constants in the natural crystal structures of the first and second material. In particular, if the lattice constants are not too different, then the joining of the two materials will succeed. Close to the interface between the first and the second material, the separation of neighboring atoms of core or shell material will be different from that of the interatomic separation in a natural bulk crystal. This follows from the matching of these two dissimilar materials. That is the materials are strained. Notably, both the first and the second material is in general strained, i.e. the atoms on both sides of the material interface will "adjust" their inter-atomic spacing in order to allow this joining of two dissimilar crystals.

Mathematically, the strain field, epsilon (ε), is the relative extension or compression. At the interface, the local mismatch, in order to obtain a pair-wise match of core and shell atoms, is given by

[l+ε(r _A )] * c _rA =[l+ε(r _B )] * c _rB

wherein the coordinates r _A and re approach each other from side A (i.e. the first material) and side B (i.e. the second material) of the interface, respectively, and C ₁ - _A and CrB are the tangential components of the lattice constants (or vectors) of the first material (A) and the second material (B), respectively, and E(T _A ) is the strain in the first material (A) at the interface and ε(rβ) is the strain in the second material (B) at the interface. This can be seen as a source term for the elastic equation of equilibrium leading to an overall deformation in the combined structure of the first material and the second material. In this way a strain in the first material can be created.

Other ways of obtaining a strain in the first material is also within the gist of the present invention, such as mechanical bending and thermal expansion. Since the piezoelectric field resulting from the strain engineering is built into the device it is induced in the device. The first and the second material may be configured into a core- shell (according to above and below) or sheet geometry.

According to one embodiment of the present invention an electronic semiconductor device based on nanowires with a built-in piezoelectric field is provided. The nanowires are used for photon harvesting, whereby a built-in piezoelectric field in the nanowires replaces the electric field of a pn-junction, which is used in conventional photovoltaic cells. In this way additional possibilities of tuning the band gap energy, and thereby also tuning the absorption spectrum, can be introduced to the photovoltaic cell. Grown nanowires are typically tens or hundreds of nanometers wide and up to several micrometers long. Nanowires are presently fabricated by liquid phase epitaxy, vapor phase epitaxy, or molecular beam epitaxy. These processing steps have been demonstrated in experimental labs and documented in the literature.

By altering the growth conditions, a structure is obtained with either radial or axial heterostructure and in some cases both. In the radial case, the device consists of a core and one or several shells of different materials with different lattice constants. In the axial heterostructure, varying materials are situated along the axis of the wire. Combining both radial and axial heterostructures is useful for embedding quantum dots in a wire.

A heterojunction may be achieved using nanowires with a core-shell configuration, as is indicated in Fig. 2. A heterojunction is the interface that occurs between two layers or regions of dissimilar crystalline semiconductors. In an embodiment, according to Fig. 2a, there is provided a nanowire 10, intended to be applied in a device for converting electromagnetic radiation into an electric current. The nanowire 10 comprises a first and a second crystalline or poly- crystalline semiconducting material 11, 12. The first material 11 is comprised in a longitudinal core of the nanowire 10, and said second material 12 is comprised in a longitudinal shell of said nanowire 10. The shell then encloses the core. The shell may be looked upon as a tubular element, with a first and second end, and a central lumen. The central lumen is filled with the core, such that the core also extends from the first to the second end. Thus, the nanowire 10 has a core-shell configuration or structure.

The first material 11 is strained within the shell of the second material 12 and has piezoelectric characteristics. The nanowire is manufactured using strain engineering, resulting in a piezoelectric field within the device. The two semiconductor materials 11, 12 may be deposited by utilizing a metalorganic vapor phase epitaxy method, as indicated above.

It is also possible to deposit the two semiconductor materials 11, 12 by using e.g. liquid phase epitaxy, vapor phase epitaxy, or molecular beam epitaxy, which methods are known to the skilled artisan. The epitaxial connection may be complete or partial.

The piezoelectric field may for example be induced by epitaxially connecting at least the first material 11 and the second material 12. The first and the second materials 11, 12 may be of the same crystal phase but have different lattice constants. The first and the second materials 11, 12 may also be of different crystal phases (e.g. zinc blende and wurtzite crystals). The term heterojunction refers to the interface between two materials of different semiconductor crystals. This interface can be a source of mechanical strain if the crystal lattices of these materials have different lattice constants. If the two materials form an ideal interface an epitaxial interface is obtained. In this way a strain in the materials 11 and 12 can be created.

Figure 2a, illustrates a core 11 - shell 12 geometry of a nanowire 10, according to one embodiment. Figures 2b and 2c show schematically a geometry of a core 1 Ia of a nanowire 10, said core 11a being of a zinc blende or wurtzite crystalline material, fabricated by catalyst-assisted growth. The surface geometry of the core 11a of the nanowire 10 can be composed of small micro facets, forming a modulated nanowire surface. This is disclosed in figure 2b. The surface geometry of the core 1 Ia of the nanowire 10 may also be hexagonal. This is disclosed in figure 2c. The surface geometry may in some cases be a mixture of the modulated geometry and the hexagonal one. The sphere 13 at the end represents a catalyst particle. The cores l la and 1 Ib of nanowires 10, according to Figs. 2b and 2c consist only of one material - and the catalyst particle. The nanowire 10 of Fig. 2a may be grown from the core geometry of Fig. 2b or 2c by growing a shell around this. Suitable materials for catalyst may for example be gold and silver. The catalyst may be of a material selected from the group comprising Fe, Co, Ni, Pt, Pd, Cu, Ag, and Au. The exact geometry of a nanowire depends on the growth conditions and the local type of crystal structure in the nanowire.

If the interface between the materials, i.e. a heterojunction, is good enough and if the geometry is chosen correctly, elastically strained materials and a piezoelectric field with a non- vanishing component along the device may be obtained. A heterojunction strains in general both materials on both sides of the interface. This gives rise to strain energy in both materials. During the growth of this kind of interface atoms of the second material are deposited on/around a crystal of the first material. The deposited atoms of the second material tend to minimize their energy by (i) forming a strained lattice on the lattice of atoms of the first material, where each atom on both sides of the interface finds a counterpart on the other side of the interface and forms a bond with that. If the mismatch between the natural atomic lattices of the first and the second material is too large, the atoms of the second material will (ii) form a broken crystal on the crystal of atoms of the first material. The amount of dislocations at the interfaces depends on the differences between the natural crystal of atoms of the first material and that of atoms of the second material. It depends also on the geometry of the interface and on the growth conditions.

It is usually easier to create a dislocation- free material interface with strain, if the interface is curved like it is in core-shell nanowires. The growth of nanowires and nanowire heterostructures enables thereby the combination of lattices and materials, which are less alike than what is possible during the growth of planar structures.

The elastic strain of a piezoelectric material is coupled to an internal electric field, i.e. a piezoelectric field. It follows that for a piezoelectric material we have the direct piezoelectric effect, the converse piezoelectric effect and the interplay between these. If a stress is applied to a piezoelectric material there is developed an electric potential difference whose magnitude is proportional to the applied stress. This is referred to as the direct piezoelectric effect. The converse effect refers to the fact that applying an electric field to a piezoelectric material induces an elastic strain in the material. The situation is in reality, however, an interplay between the direct and the converse effect.

The elasticity of a piezoelectric medium is described by the constitutive equations:

σ _u = c _ljM ^E ε _M - e _k i _j E _k (summing over j , and k) D ₁ = C _1J k S _J k + κ _u ^s E _j , where σ is the stress tensor, c ^E is the elastic stiffness tensor, ε is the strain tensor, e is the piezoelectric tensor, E is the electric field, D is the electric displacement, and κ ^s is the dielectric tensor. If a piezoelectric material is strained, i.e. 8 _M ≠ 0, this will in general induce an electric field Ek ≠ 0. In the present invention this electric field is used for the separation of photon generated electron-hole pairs. In a core-shell geometry of dissimilar crystalline materials there will be an elastic strain due to the lattice mismatch between the crystal of the core and the shell or between different shells. This elastic strain induces also an electric field in the structure as described above.

The orientation and the magnitude of the electric field depends on the magnitude of the lattice mismatch, the elastic stiffness constants CykΛ the piezoelectric constants e _u k, the dielectric constants κ _u ^s and the orientation of the crystal. An increase in the piezoelectric constant will typically increase the piezoelectric field. A decrease in the dielectric constant will typically increase the piezoelectric field.

For example the following cases have been analyzed and are beneficial for photovoltaic cells:

A core-shell wire comprises a core and a shell of a first and a second zinc blende crystalline material, respectively, said first and second material having different lattice constants. The axis of the wire is aligned in parallel with a [111] direction of the zinc blende crystalline materials of the wire. This will induce a strain in the wire, which induces a non-zero component of the piezoelectric field along the axis of the wire.

A core-shell wire comprises wurtzite crystalline materials with different lattice constants. The axis of the wire is aligned in parallel with a [0001] direction (also referred to as the c-axis) of the wurtzite crystalline materials. This will induce a strain in the wire, which induce a non-zero component of the piezoelectric field along the axis of the wire.

A planar structure comprises two thin sheets of zinc blende crystalline materials with different lattice constants. The orientation of the sheets is such that a [H I] crystal direction of the zinc blende crystalline materials lies within the plane of the sheets. This will induce a strain in both sheets, which induces a non-zero component of the piezoelectric field within the plane of the sheets.

A set of nanowires are then connected to a first and a second electrical contact 31, 33, in accordance with Figs. 3a to 3c, at both of its longitudinal ends. Electrical circuits and contacts for nanowires are commonly known and have been demonstrated experimentally for solar cells, sensors, transistors and electronic logic. Figure 3a illustrates the device 10 according to an embodiment. The device comprises a nanowire 10 connected at a first longitudinal end to a first electrical contact 31 and at a second longitudinal end to a substrate 32. The second longitudinal end of the nanowire 10 is connected to a first side of the substrate 32. On a second side of the substrate the second electrical contact 33 is connected. The first and second electrical contact 31, 33 of the device may be connected to a resistance or load 34 by leads 35, 36, respectively. The resistance 34 may for example be an apparatus to be driven by the current obtained from the device, or a battery etc. The first and second contacts 31, 33 - specifically the second contact 33, since it is to be situated in a position, which the photons may have to pass to reach the piezoelectric material - may be manufactured from a material or a combination of materials selected from the group comprising nickel, aluminum, titanium, chromium, SnO _x , Indium tin oxide (ITO), silver, or gold. These materials allows for the manufacturing of transparent contacts. If no necessity of transparency of the contacts 31, 33 exist, the contacts 31, 33 may be manufactured from any conductive material, such as metal or metallic material. It is also possible to arrange non-uniform sheet geometry of contact 33 onto an array of nano wires 10. Such a non-uniform sheet geometry may for example be achieved by arranging strips of contact material onto the ends of the nanowires. Other patterns than strips are also possible, such as nets, within the scope of the present invention. Non-uniform sheet geometry of contact 33 allows for increased permeability of photons.

Figure 3b illustrates a device according to an embodiment, comprising a number of nanowires, wherein each nanowire is connected to a first electrical contact 31 in form of a first electrical point contact 31a. Figure 3c illustrates a device according to an embodiment, comprising a number of nanowires, wherein the first electrical contact

31 is provided in the shape of a second electrical sheet contact 31b, which provides for easier manufacturing of the device.

The electric field of the piezoelectric device in general has a non-zero axial component. This applies in particular for a [111] oriented zinc-blende crystalline core- shell wire and [0001] oriented wurtzite crystalline core-shell wire of dissimilar materials. This implies that there will be an electric potential difference between the two ends of the nanowire 10. The axial component of the piezoelectric field will be approximately constant, far from the ends of the wire, if the wire is long enough, i.e. roughly 4 times longer than its diameter or more. The magnitude of the field will in this case be the same at the whole nanowire 10 cross section, i.e. in the core and in all shells. The piezoelectric field will vary more near the ends of the wire, within a distance of the order of 2*d from the nanowire 10 end, where d is the total diameter of the wire. The electrical contacts may be applied to both the core 11 and shell 12 at the end of the wire. It may in some embodiments be beneficial to contact only one material, such as the core 11 of the first material, at each end of the nanowire 10. Contacting only one material and different materials at the two ends of the nanowire might be beneficial if one aims for a nanowire 10 in which the electron and hole currents are confined to different regions of the nanowire 10. For example the electron and hole currents may be confined to the core 11 and shell or two different shells, respectively. Figure 4 illustrates the energy characteristics of a nanowire 10 according to an embodiment. Figure 4 shows schematically the energy characteristics of strained core- shell nanowires. The conduction and valence band edges are shown to the left as a function of z, i.e. along the axis of a wire. The constant slope in the bands, in the center of the wires, is due to the piezoelectric field, which is induced by the strain. An electron is excited from the valence bands to the conduction band during the absorption of a photon. This generates an electron in the conduction band and a hole in the valence bands. These are then separated by the piezoelectric field, which causes a drift of electrons towards one end and a drift of holes towards another end of the nanowire. This drift corresponds to an electric current that is to be collected at the ends of the nanowires. The geometry of a nanowire array is shown schematically to the right.

Figure 5 shows schematically the stress in a core-shell nanowire of dissimilar crystalline semiconductors. The core will be under compression or tension if the lattice constant of the core is greater or less than that of the shell. This will induce forces in the crystals where the two materials will find a strained state - where both materials are strained - according to the least total strain energy. Figure 6 illustrates the elastic strain, piezoelectric potential and band edges in core-shell nanowires. Figure 6 illustrates the elastic strain (b), the piezoelectric potential (c), and the band edge energies (d) on the axis of a core-shell nanowire according to an embodiment of the present invention (a). This particular embodiment consists of a cylindrical InAs core, into which the charge carriers are confined, and a shell of InP. The InAs core becomes compressed by the InP shell. This strain will, however, be present only inside or in the vicinity of the shell. The strain gives rise to a piezoelectric field within the shell. The electric field along the axis of the nanowires is seen as a gradient in the piezoelectric potential in Figure 6(c). The total potential energy, i.e. the band energies in Fig. 6(d), of the carriers is the combined effect of the bulk band edge energies, the piezoelectric potential, and the strain- induced deformation potentials of the respective band edges. The diameter of the nanowire or whisker is in the order of 1 nm to 10 μm.

In one embodiment the nanowire is manufactured such that the diameter is in the interval of 1 to 700 nm. When the nanowire is manufactured such that the diameter falls within this interval of thinner/smaller diameters a great number of advantages may be obtained. A thinner core-shell nanowire, i.e. having a smaller diameter, has a larger interface curvature, i.e. the curvature radius is smaller. The large curvature allows for a more effective strain relaxation without forming defects. The thinner the wire is the stronger is the polarization dependence of the photon harvesting. This could be used to fabricate photon detectors with an intrinsic polarization preference. The growth of thinner nanowires tends to give rise to nanowires with a wurtzite crystalline structure. These nanowires have a simpler geometry (hexagonal cross section). This is also a way of adjusting the material properties of the wire. Strained core-shell nanowires of a wurtzite lattice usually have a larger piezoelectric field than the corresponding zinc blende structure. Thin wurtzite nanowires seem to have a different polarization dependence of the photon absorption for solar cell applications than zinc blende ones. The combination and/or variation of these two crystal phases in nanowires may allow for tuning and optimization.

In another embodiment the diameter is in the interval of 1 to 10 μm. When the nanowire is manufactured such that the diameter falls within this interval diameter the robustness of the device is increased and the polarization dependence of the photon harvesting decreases. In this respect it is believed that the polarization dependence can be overcome in solar cell applications by regulation of the device design. When the nanowire is manufactured such that the diameter falls within this interval of thick/large diameters some advantages may be obtained. Making electrical contacts may become easier. The dielectric material between the nanowires may in certain instances be omitted.

The polarization dependence of the photon harvesting is a combined result of the confinement of charge carriers along the wire, the band structure of the wurtzite or zinc blende crystal, and the dielectric confinement of light. Confining electrons and holes into a wire, aligned along the z-axis, gives rise to electron and hole ground states of a particular character (symmetry properties). The optical selection rules, between these states, define which kind of light polarization that can be absorbed with these states. In the case of zinc blende materials the optical selection rule is such that photons, polarized along the z-axis, are less likely to be absorbed with electron-hole pairs confined along the z-axis, in comparison to photons polarized perpendicular to the nanowires axis. The symmetry character of the electron and hole states becomes more isotropic when increasing the nanowires diameter. The polarization dependence decreases and vanishes, thereby, when approaching bulk volumes. The device is not limited to the use of nanowires having a very small diameter but may be equally applicable for larger wires. The effect of the device according to some embodiments is scale independent, although the growth or manufacturing of the nanowires might not be. Note that the strain and the piezoelectric fields are indeed scale invariant, but the piezoelectric potential is indeed scale-dependent. The scale- dependence is important for the device characteristics. However, the main idea of the invention is not limited to any particular scale.

It should be appreciated that the piezoelectric core-shell device can be achieved, not only by discontinuous material interfaces between clearly dissimilar materials, but also with a smooth transition between core and shell materials or between two shell materials. This implies that the material composition of the shells and the core do not have to be homogeneous. By smoothly varying the material composition along the growth of the shell one can achieve non-uniform shells and even geometries where the material composition changes smoothly from core material to shell material along the nanowire radius. This might allow growing heterostructures between more dissimilar materials with less strain-induced dislocations. Smooth material interfaces, i.e. smooth transitions between different materials, along the axis of the wire might also be beneficial for the same reason.

In an embodiment, the device comprises two or more shells of different crystalline or poly-crystalline semiconducting materials. Figure 7a illustrates such a device. An additional inner shell 71, in accordance with Fig. 7b, which is a cross section of a device according to one embodiment, may be deposited onto the core 11 by using e.g. liquid phase epitaxy, vapor phase epitaxy, or molecular beam epitaxy, which methods are known to the skilled artisan. Onto the inner shell 71 the shell 12 of the second material is deposited by using e.g. liquid phase epitaxy, vapor phase epitaxy, or molecular beam epitaxy. In this way a radial heterostructure is obtained within the shell

12. The radial heterostructure may combine several semiconducting materials with piezoelectric properties and thereby achieve a multi-color device. Thus, the core 11 may be of a semiconducting material with piezoelectric property different from the semiconducting material with piezoelectric property of the inner shell 71. The core 11 may for example be of a material that absorbs low-energy photons (i.e. photons with a long wave length, whereas the inner shell 71 is of a material absorbing the high-energy (short wave-length) photons. The band gap energy would then be higher in the inner shell 71 than in the core 11. In this way, the confinement of electrons and holes to the core 11 may be enhanced. The absorption properties of this device can be tuned by additional inner shells and the shell 12, modifying the strain and the dielectric properties of this device. The absorption of a photovoltaic cell and photon detector and emission of a light emitting device can be tuned by introducing additional carrier confining layers or segments to the structure. This would in principle lower the quality requirements of the shell 12, as the isolation of the charge carriers, such as the core 11, from the shell 12 would be improved. The additional confinement of the inner shell 71 may also be used to design and optimize the absorption/emission spectrum of the device by modifying the density of electron and hole states in the nanowire. In one embodiment of a photovoltaic cell according to the present invention the band gap energy in the inner shell 71 is high and the band gap energy in the core 11 is intermediate. According to one embodiment the high band gap energy may for example be in the interval of 1.7 to 1.9 eV, such as approximately 1.8 eV, and the intermediate band gap energy may for example be in the interval of 0.9 to 1.1 eV, such as approximately 1.0 eV. These estimations are however only intended to be illustrative for the interpretation, and should not be construed as limiting, since these estimations are quite rough. In another embodiment, according to Fig. 7c, which is a cross sectional view of a device, the core 11 may comprise an axial heterostructure. The axial heterostructure may comprise axial segments 72, 74, and 75, said segment 74 being of a different material than segments 72 and 74. In this arrangement the high-energy photons are absorbed at the upper end of the material, i.e. segment 72, of the core 11, and the low energy photons are absorbed in the intermediate material, i.e. segment 74, and in the lower end material, i.e. segment 75, of the core 11.

The magnitude of the strain and the piezoelectric field is created by the outer shell 12, being of a different material than the core 11. Segments 72, 74, and 75 are created of semiconductors with different band gap energies and piezoelectric properties.

The material of segments 72 and 75 may be the same. This may for example be the case when the axial heterostructure is used to provide a quantum dot. Constructing a quantum dot within the nanowire is beneficial in a light emitting device, in accordance with embodiments below. In this respect, materials 72 and 75 may be materials having higher band gap energies than that of segment 74. This construction may confine the charge carriers to the quantum dot segment 74 and enhance the recombination of electrons and holes therein. In a light emitting diode (LED), constructed with an axial heterostructure, the materials of segments 72, 74, and 75 may be selected such that the band gap energy in segment 72 is high, the band gap energy in segment 74 is intermediate or low, and the band gap energy in segment 75 is high.

In one embodiment the segment 72 is of a material having a band gap energy in the interval of 1.7 to 1.9 eV, the segment 74 is of a material having a band gap energy in the interval of 1.2 to 1.3 eV, and the segment 75 is of a material having a band gap energy in the interval of 0.6 to 0.8 eV.

In a photovoltaic cell, constructed with an axial heterostructure, the materials of segments 72, 74, and 75 may be selected such that the band gap energy in segment 72 is high, the band gap energy in segment 74 is intermediate, and the band gap energy in segment 75 is low.

In Figs. 7b and 7c, a dielectric medium 73 may be deposited around the shell, to form a device body. The dielectric medium may be a medium that does not induce strain in the nanowire. Furthermore, the dielectric medium may be a medium that does not conduct electrical current; not have polarization-dependent properties; is transparent in the frequency domain that the device is targeted for. In this respect amorphous (noncrystalline) dielectrics are good. The dielectric medium may for example be selected from the group comprising SiN:H, and Chemical Vapor Deposited silicon (di)oxide (SiOx), but other materials satisfying the demands above are also suitable without departing from the scope of the invention. SiN:H may for example be deposited by plasma-enhanced chemical vapor deposition (PECVD).

It should be appreciated that the invention is not limited to nanowires only having a core-shell configuration, but various geometries could be used in order to obtain a desired piezoelectric field. For example, in some embodiments one of the crystalline semiconducting materials does not fully enclose the other crystalline semi conductor material. The advantages of not using a fully enclosed shell might include; not limiting to a particular geometry and thereby allowing for a larger tunability in the strain-engineering, and becoming easier to form an electrical contact to the core material. Semiconductor materials are commonly classified according to the periodic table groups from which their constituent atoms come. In this respect, the first material 11 may be selected from the group comprising III -V, H-VI, and IV-IV compound semiconductors, or a suitable ceramic with a prerequisite that it is piezoelectric. Suitable semiconductors with piezoelectric characteristics within these groups are BN, BP, BAs, BSb, AlN, AlP, AlAs, GaN, GaP, GaAs, GaSb, InN, InP, InAs, InSb, AlGaAs, InGaAs, AlGaAsP, GaAsP, ZnO, ZnS, ZnSe, ZnTe, CdO, CdS, CdSe, CdTe, MnTe, SiC, and SiGe. Especially, the semiconducting materials in the group comprising AlN, AlP, AlAs, GaN, GaP, GaAs, GaSb, InN, InP, InAs, InSb, AlGaAs, InGaAs, AlGaAsP, GaAsP, ZnO, ZnS, ZnSe, ZnTe, CdS, CdSe, CdTe, MnTe, and SiC, work well. 5 The second material 12 may also be selected from the group comprising III -V,

H-VI, or IV-IV semi conductors, or a suitable ceramic, as long as the equation [l+ε(r _A )] * * CrB substantially is complied with. The equation [1+E(TA)] * CrA=[l+ε(r _B )] * CrB expresses that the tangential components of the lattice vectors (also the inter-atomic separation) C _ΓA , of the first material, and C _ΓB , of the second material, are 0 strained so that they are of the same length, in accordance with above. However, the second material can be any semiconductor or an insulator with a crystalline structure. The second material may also be a material with piezoelectric property. Such suitable material for the second material can for example be selected from the group comprising Si, Ge, C, SiGe, BN, BP, BAs, BSb, AlN, AlP, AlAs, GaN, GaP, GaAs, GaSb, InN, 5 InP, InAs, InSb, AlGaAs, InGaAs, AlGaAsP, GaAsP, ZnO, ZnS, ZnSe, ZnTe, CdO, CdS, CdSe, CdTe, MnTe, and SiC. Note that Si, Ge and C do not have piezoelectric properties.

If the structure is composed of a core and several shells, then at least one of the involved materials have to be a piezoelectric material. Thus, in one embodiment the O core may lack piezoelectric property, while at least one of the at least one shell has piezoelectric property.

The first and the second materials of a core-shell geometry may be of the same crystalline structure. Typically they are either zinc blende or wurtzite. However, these structures do very often contain regions of different crystals. The properties of the 5 nanowire may be tuned by changing the materials and structure. For example, changing the nanowire length, core-shell diameters, and using several shells changes the structure. The change of structure involves either a change of the geometry, a change of the materials or a change of both of these.

Multi- shell structures offer a larger tunability but the dependence of the O properties on the geometrical parameters may be harder to predict.

One way to modify the properties of the core-shell nanowires is to grow the nanowires along different crystal directions. The [111] direction is the most common for zinc blende growth, whereas the [0001] direction is the most common for wurtzite crystalline growth. The properties, and in particular the magnitude of the piezoelectric 5 field, of a core-shell nanowire depends on the orientation of it with respect to the crystal lattice as a result of the anisotropic stiffness, piezoelectric and dielectric tensors. This applies to both zinc blende and wurtzite nanowires. The main parameters are change of materials, and the radii of the core and the shells, with respect to each other.

In an embodiment the two crystalline semiconducting materials are selected from the group III-V materials, since III -V materials may be suitable for device from an optical perspective. In general the materials should have piezoelectric properties and allow for pseudomorphic matching. The latter could certainly also be obtained for crystal lattices different from zinc blende or wurtzite. III-V materials are promising mainly due to the fairly established processing techniques for these. Most of the III-V materials do also have nice optical properties. A [I I l] orientated core-shell hetero structure of zinc blende crystalline III-V materials is fairly similar to a [0001] orientated core-shell heterostructure of wurtzite crystalline III-V materials. The properties of these two crystals of the same material are pretty alike, although not identical. This is due to the similarities of the crystals at a microscopic scale. The main known differences between structures of these crystals are that the magnitude of the piezoelectric field seems to be larger in a wurtzite structure, and that the absorption of photons is polarized differently. The zinc blende nanowires absorb and emit primarily photons with a polarization along the nanowires. This optical selection rule is weaker for the wurtzite nanowires, which may also absorb and emit primarily photons with a polarization perpendicular to the nanowires

It is, however, noteworthy that the properties of wurtzite crystalline III-V materials are in general poorly understood as these materials normally do not be grown in bulk in the wurtzite phase.

Figure 8 illustrates theoretical field strength in various material combinations (zinc blende crystalline materials) in a nanowire with core-shell configuration. Figure 8 shows how the magnitude of the piezoelectric field depends on the size of the core for a fixed size of the shell. Each line corresponds to one particular combination of a core material and a shell material. The magnitude of the field is plotted as a function of the ratio between the volume of the core and the volume of the shell. Both materials are assumed to be zinc blende crystalline.

The device may be manufactured, in accordance with the embodiment illustrated in Figs. 9a to 9e, by depositing catalyst particles 13 on a first side of a substrate material 32, according to Fig. 9a. The catalyst may be of a material selected from the group comprising Fe, Co, Ni, Pt, Pd, Cu, Ag, and Au. The substrate material 32 may be selected from the group comprising the materials that were suggested for the nanowires themselves, silicon, and sapphire. Other materials might be possible as well, without departing from the gist of the invention. The catalyst particles 13 are dot deposited or lithographically defined on the parts of the substrate 32 where the nanowires are intended to grow. The size of each dot substantially determines the diameter of the core material of the nanowire. Then, catalyst assisted growth of nanowire cores 11 is initiated, in accordance with Fig. 9b, by methods known to the skilled artisan, at the location of the dot depositions. When the core 11 has grown to a suitable length, a shell 12 is grown around the core, in accordance with Fig. 9c, to create a core-shell configuration of a nanowire. Thereafter a dielectric medium 73 is deposited between the nanowires to form a body, in accordance with Fig. 9d.

SiN:H may be deposited by plasma-enhanced chemical vapor deposition (PECVD). Optionally, the catalyst particles are then removed, e.g. by etching. Thereafter, metallic contact sheets 33, 31 are applied on a second side of the substrate, and on top of the nanowires and the dielectric medium (optionally with the catalyst particles on top), in accordance with Fig. 9e. The metallic contact 31 on top of the structure should be substantially transparent for the light to be absorbed by or emitted from the wires. This can be achieved by making the top sheet 33 very thin or by manufacturing it into a pattern (e.g. stripes) that is in contact with the ends of the nanowires but does not cover all of the dielectric 73. According to another embodiment, illustrated in Figs. 10a to 1Of, the device may be manufactured by etching holes 101 in a protecting mask 102 placed on a first side of a substrate material 32, in accordance with Fig. 10a. Thereafter, growth of nanowire cores 11 is initiated in the holes 101, in accordance with Fig. 10b, by methods known to the skilled artisan. Then, the protecting mask 102 is removed, in accordance with Fig. 10c. When the core 11 has grown to a suitable length and the mask 102 has been removed, a shell 12 is grown around the core 11, in accordance with Fig. 1Od, to create a core-shell configuration of a nanowire. Thereafter a dielectric medium 73 is deposited around the shell 12, to form a body, in accordance with Fig. 1Oe. Thereafter, metallic contact sheets 33, 31 are applied on a second side of the substrate, and on top of the nanowires and the dielectric medium, in accordance with Fig. 1Of.

The device according to some embodiments may be used as a device comprising solar cells with a broad absorption spectrum and high efficiency. Thus, in an embodiment, each nanowire of the device is used as a photovoltaic cell, i.e. solar cell. In solar cells, when the sun light hits the surface of the cell, electron-hole pairs are created, whereby an electric field is needed to separate the electron and the hole and create an electrical current through the cell. When a piezoelectric field is created, such as by the embodiments of nanowires, there is no need for pn-junction.

In an embodiment the nanowire is grown with multicolor function for absorption of a wide range of light wavelengths, in accordance with Figs. 7a to 7c above. The multi-color function can be obtained by growing a heterostructure within one or several shells of the nanowire. The interior of the wire will in this case be composed of two or more domains of different semiconductors, each of these being selected for the absorption of a particular section of the spectrum of the solar radiation. These domains can also be quantum dots within the nanowires. Another approach would be to construct a core-shell nanowire with several shells of different materials, each of these being selected for the absorption of a particular spectrum of the solar radiation.

Figures 7b and 7c illustrate these ideas. These structures could allow for a "multi-color" absorption. For example, 11, 71, 72, 74, and 75 may absorb different parts of the solar spectrum if they were of different semiconductors, with different band gap energies. The absorption of light in a semiconductor, with a direct band gap, is most effective for photons having energy slightly above the band gap energy.

The same ideas can also be used for detectors, according to the embodiments described below. Nanowires enable sensitive photon detectors due to the strong confinement of the photon-generated electrical current into very thin nanowires.

The yield of collected and detected photons may be optimized by nanowires designed with separate regions or shells for electron and hole currents, respectively. The different possible band alignments are commonly divided into three classes as type I, II, and III, according to which side of the heterojunction the electrons and holes localize. In

Figure 7a there is used a type II radial interface. This structure allows for separation of electrons and holes, such that the electrons are kept in the core, while the holes are transferred to the shell. In an embodiment wherein the nanowire comprises more than one shell, the electrons and holes may be separated, such that electrons occurs in a core or shell while the holes occur in a shell different from the shell wherein the electrons optionally occur. The benefit of this structure would be: a more effective separation of electrons and holes, leading to a smaller recombination rate. Typically, when a photon is absorbed and an electron-hole pair is created in a photovoltaic cell, there is always a probability of the opposite process - an electron and a hole recombines and a photon is emitted. This process reduces the amount of electrons and holes that can be collected at the contacts.

The device according to some embodiment may be used as a photon detector for a wide variety of spectral bands - ranging from deep infrared electromagnetic radiation to ultraviolet electromagnetic radiation. Ultimately the underlying material should be selected from the point of view of the band gap. A semiconductor can also absorb photons of energies greater than the band gap energy of the semiconductor. A material with a small band gap is thereby suitable for absorption of infra-red light. A material with a larger band gap is on the contrary suitable for absorption of (ultra) violet light.

The exact absorption rate and spectrum are also modified by strain and quantum confinement. The quantum confinement blue-shifts the absorption spectrum. A compressive (tensile) strain typically blue-shifts (red-shifts) the absorption spectrum of a III-V compound semiconductor. The wavelength λ of light is given by

λ = h c / E,

wherein h is Planck's constant (6.626e*10 ^~34 kgm/s), c is the speed of light (299792 km/s), and E is the photon energy. For example:

InSb has a band energy of 0.17 eV, integrating this material in the core-shell nanowire would allow absorption of infra-red light with wavelengths slightly below 7300 run.

InAs has a band energy of 0.36 eV, integrating this material in the core-shell nanowire would allow absorption of infra-red light with wavelengths slightly below

3400 nm.

GaAs has a band energy of 1.43 eV, integrating this material in the core-shell nanowire would allow absorption of near- infra-red light with wavelengths slightly below 867 nm. GaN has a band energy of 3.4 eV, integrating this material in the core-shell nanowire would allow absorption of near-ultraviolet light with wavelengths slightly below 365 nm.

According to one embodiment of the present invention, the device(s) described above may be used for quantum cascade lasers (QCL). Within a bulk semiconductor crystal, electrons may occupy states in one of two continuous energy bands - the valence band, which is heavily populated with low energy electrons and the conduction band, which is sparsely populated with high energy electrons. The two energy bands are separated by an energy band gap in which there are no permitted states available for electrons to occupy. In quantum cascade structures, electrons undergo intersubband transitions and photons are emitted. The electrons tunnel to the next period of the structure and the process repeats. In semiconductor laser diodes, electrons and holes are annihilated after recombining across the band gap and can play no further part in photon generation. However in a QCL, once an electron has undergone an intersubband transition and emitted a photon in one period of the superlattice, it can tunnel into the next period of the structure where another photon can be emitted. This process of a single electron causing the emission of multiple photons as it traverses through the QCL structure gives rise to the name cascade and makes a quantum efficiency of greater than unity possible which leads to higher output powers than semiconductor laser diodes. According to the present invention quantum cascade lasers in piezoelectric core-shell nano wires may be created. The core of the nanowires is then composed of an axial superlattice. The benefit in comparison with traditional QCLs would then be that there would be no need of large sheets of material. The geometry would instead have a fairly small lateral extension in form of the diameters of the nanowires in the array of nanowires. This heterostructure superlattice would then be covered by a shell, inducing a strain and an axial piezoelectric field. The lasing mechanism of the nanowires-based QCL would be exactly that of conventional QCLs, with the difference that the piezoelectric field would drive the electrons from one segment to another. The benefits of this would be that one could fabricate longer superlattices, composed of more "quantum wells". This would increase the number of created photons per one electron. Applications and use of the above-described embodiments according to the invention are various and include exemplary fields, such as photovoltaics.

According to the present invention diodes in piezoelectric core-shell nanowires may be created. This is then a piezoelectric diode, in contrast to diode with a pn- junction. The mechanism of creating current from photons, by creating electrons and holes from said photons, is then reversed. This means that the core-shell nanowires practically can be used as light emitting diodes (LEDs) without the need of dopants.

Even though the embodiments of the present invention eliminates the need of dopants, it is still within the scope of the present invention to include dopants, such as pn-junctions in the devices, such as the nanowires, and specifically the core of the nanowires, without departing from the gist of the invention. Thus, in still another embodiment a p-n junction is integrated in the core of the nano wires according to the embodiments above. The right-hand-side of Fig. 1 shows how this pn-junction could be integrated in cores of nanowires in order to fabricate core-shell nanowire-based photovoltaic cells. Its operation principle is based on the conventional charge-separation mechanism of a pn-junction. Around said cores shells are then arranged, in accordance with the embodiments above, to create a piezoelectric field in between the ends of the core.

According to one embodiment the piezoelectric device, the first and the second material may be configured into a sheet geometry, in accordance with figure 11. The device consists of a first and a second sheet 111, 112. The first and the second sheet 111, 112 are parallel and epitaxially connected. The first sheet 111 may be a rigid substrate onto which the other sheet is grown using e.g. vapor phase epitaxy, liquid phase epitaxy or molecular beam epitaxy. The orientation of the first and the second zinc blende crystalline materials 111, 112 is such that the interface between the two sheets is perpendicular to the [-1 0 1] crystal direction. In this respect the first sheet 111 may be called a [-1 0 1] oriented substrate. If the two sheets 111, 112 are of crystalline materials with different lattice constants both materials will in generally be strained. If one of these sheets is a substrate, i.e. this sheet is much thicker than the other one, this sheet will be less strained. There will be a strain-induced piezoelectric field within the strained material when this material is piezoelectric. Thus, in accordance with figure 11, the second sheet 112 may have piezoelectric properties, such that a piezoelectric field will be directed within the second sheet 112 along the [010] crystal direction. Contacts 113 may then be arranged at ends of the strained piezoelectric sheet 112. The electrical contacts could be manufactured into two parallel stripes, both of them being aligned perpendicularly to the [010] crystal direction. The same materials could be used for the first and second materials of the sheet geometry as was suggested for the first and second materials of the core-shell nanowire. The same materials could be used for the electrical contacts as was suggested for the electrical contacts of the core-shell nanowire. Figure 12 depicts the photovoltaic mechanism in strained core-shell nanowires and the schematic band diagram in a strained core-shell nanowire 1202 between two metallic contacts 1201, 1203. The piezoelectric field 1204 induces a linear slope in the conduction 1208 and valence band edges 1209, in the center of the nanowire. The bands are, however, curved by charge-exchange effects near the metals. The absorption of photons 1207 excites electrons 1205 from the valence bands 1209 of the nanowire into the conduction band 1208, giving rise to holes 1206 in the valence bands. These electrons 1205 and holes 1206 drift in opposite directions because of the piezoelectric field 1204 and the slope of the conduction 1208 and valence bands 1209.

Figures 13a to 13f show the elastic strain on a cross section (far from the ends) of embodiments of core-shell nanowires. Figure 13(a) shows the exx component of the strain in a zinc blende crystalline nanowire. Figure 13(b) shows the exx component of the strain in a wurtzite crystalline nanowire. Figure 13(c) shows the ezz component of the strain in a zinc blende crystalline nanowire. Figure 13(d) shows the ezz component of the strain in a wurtzite crystalline nanowire. Figure 13(e) shows the exy component of the strain in a zinc blende crystalline nanowire. Figure 13(f) shows the exy component of the strain in a wurtzite crystalline nanowire.

Figure 14 shows the transversal component (x/y component) of the piezoelectric field at a cross section of an InAs/InP core-shell nanowire. The nanowire consists of an InAs core and an InP shell, both materials being in the zinc blende crystal phase and the axis of the wire (equals to the Z axis) being aligned with the [111] crystal direction. The magnitude of the arrow is proportional to the magnitude of the piezoelectric field. The maximum absolute value of the piezoelectric field is in Figure 13 approximately (Ex2/Ey2) 1/2 ~ 13 mV/nm. The longitudinal component Ez of the piezoelectric field is constant at the whole cross section of the nanowire. Figure 15 shows the piezoelectric field in a QD WZ structure with the isopotential lines included.

Figure 16a shows schematically a core shell nanowire which is aligned along a [111] direction of the zinc blende crystal direction or along the [0001] direction of a wurtzite crystal. The [0001] direction of wurtzite is commonly also referred to as the c axis of the crystal.

Figure 16b shows schematically an axial heterostructure super lattice of InAs and InP segments in a nanowire. This structure is composed of epitaxially strained segments of piezoelectric semiconductors. It is possible to obtain an average piezoelectric field along the nanowire axis by a proper combination of materials and by careful choice of segment thicknesses hi and h2. The magnitude and direction of the axial piezoelectric field depends on the crystal structure (e.g. zinc blende and wurtzite), on the combined materials, and on the geometry (the relations between the nanowire radius and the thicknesses hi and h2).

Figure 16c shows an epitaxially strained InAs quantum dot in an InP nanowire. The size of the quantum dot (hi and rl) and the radius of the nanowire (r2) can be tuned during the nanowire growth. The magnitude of the elastic strain and piezoelectric field will in this case depend (and can be tuned by varying) hi, rl and r2.

Figure 17a shows schematically the symmetry and principal crystal axes of the [111] oriented zinc blende crystal and 17b of the [0001] oriented wurtzite material with respect to the nanowire geometry. The crystal symmetries are depicted in terms of a rotated cube and a hexagon.

Figures 18a and 18b show the piezoelectric potential, in terms of isosurfaces for selected values of the potential, in a ZB and a WZ phase core-shell heterostructure. In figs. 18a and 18b rl=4nm and r2=9nm are used as an example. The axial component Ez of the piezoelectric field is constant everywhere within the radial heterostructures and far from the ends of the nanowire. The WZ heterostructure does not develop any x or y components of the piezoelectric field, i.e. Ex = Ey = 0. The corresponding isosurfaces of the potential are therefore flat. The ZB heterostructure does, however, induce a nonvanishing electric field within the xy plane as well and the isosurfaces of the potential are correspondingly nonplanar for the ZB heterostructure. These isosurfaces are saddle surfaces with three maxima and three minima. The saddle surfaces of the potential in a ZB core-shell nanowire is expected to induce a channeling of electron and hole currents into different areas of the nanowire. This means that the electrons and holes, which are drifting along the axis of the nanowire, are also separated in the cross sectional xy plane by the piezoelectric field. This is expected to reduce the electron-hole recombination in a favorable way.

Figure 19a and 19b show the piezoelectric potential in an axial superlattice heterostructure, in terms of isosurfaces. In figs. 19a and 19b, r2=9nm, hl=8nn, h2=8nm are used as an example. The magnitude of the piezoelectric field i may be smaller in an axial superlattice heterostructure, in a nanowire, than in a quantum well superlattice.

This is a result of the strain relaxation by a bulging of the nanowire exterior. In an ideal quantum well superlattice (for which r2 approaches infinity) there is no bulging of the geometry and the piezoelectric field becomes constant within each material.

Figure 20a and 20b show the piezoelectric potential using isosurfaces in a ZB and WZ QD heterostructure. In figs. 20a and 20b, r2=4nm, r2=9nm, and hl=8nn are used as an example. The potential is remarkably different in these two structures. There are five local maxima and five local minima in the potential of the ZB heterostructure. Three equal maxima and three equal minima are located at the circumference of the shell. A fourth local maximum (and minimum) of the potential is found within the InAs core. The fifth maximum (and minimum) is located on the axis of the nanowire, in the shell, above (below) the InAs core. This potential shows a three-fold rotational symmetry about the z-axis. The absolute value of the potential |V(r)| is inversion symmetric but the potential V(r) itself is not. The piezoelectric potential of the WZ heterostructure has only one maximum and one minimum. These are located on the axis of the nanowire, close to the InAs/InP interface. This potential is cylindrical symmetric. The potential is inversion symmetric except for the sign, as in the case of the ZB structure. We note also that the magnitude of the piezoelectric potential is much larger in the WZ heterostructure than in the ZB heterostructure. The piezoelectric field is E ~6 mV/nm, in the center of the ZB InAs QD. In the WZ QD we find E ~ 14 mV/nm. The large piezoelectric field within the QD will affect the operation of this structure in electronic applications.

The direction and the magnitude of the piezoelectric field are the same in the core and the shell in an infinitely long strained core-shell nanowire. This follows from the governing equations of electrostatic fields, which require that the tangential components of an electric field are constant at any material interface. It follows that the electric field is constant (i.e. Ez has the same value and sign in the core and the shell) also in a finite core-shell nanowire, far from the ends of the nanowire. The direction of the piezoelectric field in a strained core-shell nanowire depends on the magnitude of the strain, the magnitude and sign of the piezoelectric constants and the dielectric constants of the core and shell materials.

The total axial piezoelectric field will be smaller if the piezoelectric constants of the core and shell are of the same sign, since the fields are directed in opposite directions because the core and the shell have opposite strains. The total piezoelectric field will in this case be a weighted average of the fields induced by the strained core and by the shell. The magnitude of the piezoelectric field will be larger if the core and shell materials have piezoelectric constants of opposite signs. The two materials will in this case induce a piezoelectric charge displacement in the same direction along the nanowire axis. This enhances the total piezoelectric field.

The lattice constant of a typical semiconductor crystal may be in the range 0.54 - 0.66 nm. It is more difficult to form an epitaxial heterostructure if the relative difference (al-a2)/al between the lattice constants of the materials of a heterostructure is too large. The maximum thickness of a material that can be epitaxially grown onto another one depends on the the relative difference (al-a2)/al between the lattice constants of the materials. The growth of core-shell nano wires is however more flexible because of the curved geometry and the elastic flexibility of very thin nanowire. It is possible to grow epitaxial core-shell heterostructures as long as the relative difference (al-a2)/al between the lattice constants of the materials is of the order of a few percent, for example below 10%, such as about 3%. On the other hand, if the relative difference is small, the piezoelectric field will be low. Thus, the relative difference should be above a predetermined low limit, such as above 0.5% for example above 1%, such as about 3%.

If the connection between the materials is epitaxial, the strain in the material will be strong and more predictable. Thus, epitaxial connection or growth of the crystals is desired. However, the connection between the crystals may not be epitaxial over the entire material interface of a heterojunction. If a large portion of the heterojunction is epitaxial, an efficient strain may be ontained, such as above 50%, for example above 80%, such as above 90%.

Although the present invention has been described above with reference to specific embodiments, it is not intended to be limited to the specific form set forth herein. It should be appreciated that the above-mentioned embodiments, even if described individually, could be combined within the scope of the invention. Rather, the invention is limited only by the accompanying claims and other embodiments than the specific above are equally possible within the scope of these appended claims.

In the claims, the term "comprises/comprising" does not exclude the presence of other elements or steps. Furthermore, although individually listed, a plurality of means, elements or method steps may be implemented by e.g. a single unit or processor. Additionally, although individual features may be included in different claims, these may possibly advantageously be combined, and the inclusion in different claims does not imply that a combination of features is not feasible and/or advantageous. In addition, singular references do not exclude a plurality. The terms "a", "an", "first", "second" etc do not preclude a plurality. Reference signs in the claims are provided merely as a clarifying example and shall not be construed as limiting the scope of the claims in any way.