A semiconductor structure

Abstract

A semiconductor structure (10) comprising a substrate (12) and an epitaxial metal layer (14) There is a first active layer (16) comprising a quantum confinement structure arranged to emit at a first wavelength (λ1) and a second active layer (20) comprising a quantum confinement structure arranged to emit at a second wavelength (λ2). The semiconductor is suitable to form a multicoloured µLED having emission at two or more wavelengths from the same epitaxial stack.

Description

A semiconductor structureA semiconductor structure, particularly but not exclusively a light-emitting diode (LED) with multicolour capability.It is known to produce an LED from a semiconductor structure to emit light at a specific wavelength. Typically green and blue emission is achieved in a structure formed in gallium nitride (GaN) material system, that is the layers comprise GaN and alloys with GaN. Typically red emission is achieved in a gallium arsenide (GaAs) based structure where the active (emission) layer comprises an alloy of GaAs, the substrate is GaAs and there are intermediate layers of GaAs or alloys thereof. To form a multicoloured LED it is necessary to use individual dies cut from single colour wafers and then to place them in a close-packed array on a backplane.One problem with this method is that as demand for higher display resolution shrinks the required pixel size it becomes more difficult to accurately pick and place the dies. So called microLEDs (pLEDs) have a die width of less than 1pm, compared to standard LEDs with a die width of 0.08-20mm, meaning that arrays may have many thousands to tens of thousands of dies in the space of a single LED die. When picking and placing so many of such dies the probability of misalignment or damage increases. Thus a different method to formulate an array of pLEDs is needed.The present invention provides a semiconductor structure comprising, sequentially: a substrate; an epitaxial metal layer; a first active layer comprising a quantum confinement structure arranged to emit at a first wavelength; and a second active layer comprising a quantum confinement structure arranged to emit at a second wavelength. Advantageously the semiconductor structure of the present invention provides two emission wavelengths from a single structure meaning that a multicoloured pLED, for example, can be formed more easily by growth in a single sequence in an epitaxial reactor.The epitaxial metal layer may be arranged to function as a mirror and / or an electrical contact. Advantageously the epitaxial metal layer can be grown as part of the semiconductor structure. Advantageously metal can be reflective to light at visible wavelengths so that some of the light emitted by the first and / or second active layers can be redirected to emit through the top of the semiconductor structure and is not lost to absorption by or emission through the substrate.The first active layer may comprise (lnySci.y)xGai.xN quantum dots, where 0<x,y<l. Advantageously inclusion of scandium in the composition reduces the required amount of indium which in turn reduces the likelihood of indium segregation in the layer. Advantageously quantum dots confine the charge into a one-dimensional feature which increases the probability of charge recombination and thus emission intensity. Furthermore quantum dots are not as sensitive to growth temperature as quantum wells. Thus forming the first active layer of quantum dots may allow more successful formation of the second active layer, whether it is formed of quantum dots or quantum wells.The first wavelength may be in the range 620nm to 740nm. Thus the first wavelength is red in the visible spectrum. The first wavelength may be in the range 645nm to 700nm. Thus the first wavelength is a pure red. The first wavelength may be approximately 630nm. Advantageously the first active layer emits red light from a GaN-based material which is compatible with other emission from GaN-based materials.The second active layer may comprise (lnySci.y)xGai.xN quantum dots, where 0<x,y<l. Advantageously inclusion of scandium in the composition reduces the required amount of indium which in turn reduces the likelihood of indium segregation in the layer. Advantageously quantum dots confine the charge into a one-dimensional feature which increases the probability of chargerecombination and thus emission intensity. Furthermore quantum dots are not as sensitive to growth temperature as quantum wells. Thus forming the second active layer of quantum dots may reduce the likelihood of damage to the first active layer, whether it is formed of quantum dots or quantum wells.The second wavelength may be in the range 500nm to 570nm. Thus the second wavelength is green in the visible spectrum. The second wavelength may be in the range 525nm to 550nm. Thus the first wavelength is a pure green. The second wavelength may be approximately 530nm. Advantageously the second active layer emits green light from a GaN-based material which is compatible with the red emission from the first active layer.The semiconductor structure may further comprise a third active layer comprising a quantum confinement structure arranged to emit at a third wavelength. Advantageously the third wavelength is different to the first and second wavelengths meaning the semiconductor structure provides emission at three different wavelengths, for example red, green and blue.The third active layer may comprise (Alylni.y)xGai.xN quantum dots, where 0<x,y<l.. Advantageously quantum dots confine the charge into a one-dimensional feature which increases the probability of charge recombination and thus emission intensity. Furthermore quantum dots are not as sensitive to growth temperature as quantum wells. Thus forming the third active layer of quantum dots may reduce the likelihood of damage to the first active layer or second active layer, whether they are formed of quantum dots or quantum wells.The third wavelength may be in the range 450nm to 505nm. Thus the second wavelength is blue in the visible spectrum. The third wavelength may be in the range 450nm to 470nm. Thus the first wavelength is a pure blue. The third wavelength may be approximately 460nm. Advantageously the third active layer emits blue light from a GaN-based material which is compatible with the green emission from the second active layer and the red emission from the first active layer.The semiconductor structure may further comprise a fourth active layer comprising quantum dots or quantum wells, a quantum confinement structure, arranged to emit at a fourth wavelength. Advantageously the fourth wavelength is shorter than the other wavelengths meaning that the fourth active layer optically pumps (that is, emits photons which excite electrons in) one or more of the first, second and third active layers.The fourth active layer may comprise AlxGai.xN quantum dots, where 0<x<l. Advantageously quantum dots confine the charge into a one-dimensional feature which increases the probability of charge recombination and thus emission intensity. Furthermore quantum dots are not as sensitive to growth temperature as quantum wells. Thus forming the fourth active layer of quantum dots may reduce the likelihood of damage to the first active layer, second active layer or third active layer, whether they are formed of quantum dots or quantum wells.The fourth wavelength may be less than 450nm. Thus the fourth wavelength may be ultraviolet in the electromagnetic spectrum. The fourth wavelength may be in the range 450nm to 200nm. Thus the fourth wavelength may be UVA, UVB or UVC with no visible light. The fourth wavelength may be approximately 279nm. Advantageously the fourth active layer emits short wavelength light from a GaN-based material which is compatible with the blue emission from the third active layer, the green emission from the second active layer and the red emission from the first active layer.The fourth active layer may be provided on or over the third active layer. Advantageously the fourth active layer may emit visible or ultraviolet light. Advantageously the fourth active layer mayoptically pump one or more of the other active layers. Alternatively the fourth active layer may be provided between the substrate and the first active layer. Advantageously it can optically pump the first active layer with minimal absorption or scattering losses in this arrangement.The semiconductor structure may further comprise a first contact layer on the first active layer. Advantageously the first contact layer can be used with the epitaxial metal layer to control the first active layer.The semiconductor structure may further comprise a second contact layer on the second active layer. Advantageously the second contact layer can be used with the first contact layer to control the second active layer. When both the first and second contact layers are provided the first and second active layers can be controlled independently.The first contact layer may comprise a highly doped material. The second contact layer may comprise a highly doped material. Advantageously the contacts have lower resistivity when more highly doped.The first contact layer may comprise a p++-n junction. The second contact layer may comprise a p++-n junction. Advantageously the p++-n junction controls charge flow to be through the desired active layer only.The semiconductor structure may further comprise a first reflective layer between the epitaxial metal layer and the first active layer. Advantageously the first reflective layer may be configured to reflect light at or around the wavelength emitted by the first active layer so that it is redirected into the preferred emission direction. Advantageously the first reflective layer acts to increase the emission of light at the first wavelength.The semiconductor structure may further comprise a second reflective layer between the first active layer and the second active layer. Advantageously the second reflective layer is configured to reflect light at or around the wavelength emitted by the second active layer so that it is redirected into the preferred emission direction. Advantageously the second reflective layer acts to increase the emission of light at the second wavelength.The semiconductor structure may further comprise a third reflective layer between the second active layer and the third active layer. Advantageously the third reflective layer is configured to reflect light at or around the wavelength emitted by the third active layer so that it is redirected into the preferred emission direction. Advantageously the third reflective layer acts to increase the emission of light at the third wavelength.The semiconductor structure may further comprise a fourth reflective layer adjacent to the fourth active layer. Advantageously the fourth reflective layer is configured to reflect light at or around the wavelength emitted by the fourth active layer so that it is redirected into the preferred emission direction. That direction may be towards other active layers in the semiconductor structure in order to optically pump one or more of those layers. Advantageously the fourth reflective layer acts to increase the emission of light at the fourth wavelength.The semiconductor structure may further comprise a third contact layer on the third active layer. Advantageously the third contact layer can be used with the second contact layer to control the second third active layer. When the first, second and third contact layers are provided the first, second and third active layers can be controlled independently.The semiconductor structure may further comprise a lower contact layer between the epitaxial metal layer and the first active layer. Advantageously the lower contact layer can be used with the first contact layer to control the first active layer. In this arrangement the epitaxial metal layer can be used as a mirror or reflector to reflect light at or around the first wavelength into the preferred emission direction.The semiconductor structure may further comprise an upper contact. The upper contact may be on or over the second active layer. Alternatively the upper contact may be on or over the third active layer. Alternatively the upper contact may be on or over the fourth active layer. Advantageously the upper contact may be used with the epitaxial metal layer, lower contact layer, first contact layer, second contact layer and / or third contact layer to control one or more active layers in the semiconductor structure. When all four active layers are provided the upper contact may be used with the third contact layer to control the fourth active layer or may be used with the second contact layer to control the third active layer, whichever is at the top of the semiconductor structure.The upper contact may comprise indium oxide (InjOa). Advantageously this material is transparent to light at visible wavelengths. Thus the light emitted in the active layers can be emitted through the upper contact without significant absorption, reflection or scattering.The first contact layer may comprise a first contact sublayer, an insulation sublayer and a second contact sublayer. The second contact layer may comprise a first contact sublayer, an insulation sublayer and a second contact sublayer. The third contact layer may comprise a first contact sublayer, an insulation sublayer and a second contact sublayer.The first contact sublayer and the second contact sublayer may have opposite conductivity type. Thus the first contact sublayer may be p-type and the second contact sublayer may be n-type or vice versa.The insulation sublayer may comprise a lattice matched, wide bandgap material. The insulation sublayer may comprise GaN alloyed with one or more of aluminium (Al), indium (In), scandium (Sc) and boron (B). The insulation sublayer may comprise wurtzite aluminium scandium nitride (AIScN). Conveniently wide bandgap materials introduce a lower density of defects into the structure because they are crystallographically compatible with the active layer. Additionally they can be grown in the same epitaxial reactor as the active layer. The insulation sublayer may comprise a dielectric material. The dielectric material may be any one of an oxide, glass, a crystalline rare earth oxide, and scandium oxide (Sc2O3). Advantageously oxide materials have superior insulating properties.The semiconductor structure may further comprise an electrode in or on the epitaxial metal layer. Alternatively the electrode may be in or on the lower contact layer. The semiconductor structure may further comprise an electrode in or on any or each of the first contact layer, second contact layer and third contact layer. The semiconductor structure may further comprise an electrode in or on the upper contact layer. Advantageously each electrode can be connected to an electrical supply.The present invention also provides a multicoloured LED (light emitting diode) comprising a semiconductor structure as described in any of the previous paragraphs. Advantageously each active layer can be configured to emit a different colour of visible light. Advantageously the semiconductor structure can therefore provide a red-green-blue (RGB) LED without the need to pick and place individual coloured pixels from separate colour structures.The present invention further provides a method of forming a semiconductor structure comprising steps to: grow a metal layer on a substrate; grow a first active layer on the metal layer, the first activelayer comprising a quantum confinement structure, quantum dots or quantum wells, arranged to emit at a first wavelength; and provide a second active layer on the first active layer, the second active layer comprising a quantum confinement structure, quantum dots or quantum wells, arranged to emit at a second wavelength.Advantageously the method enables production of a multicoloured light emitting structure, such as an LED, without the need to pick and place from separately grown structures. Advantageously this enables small pixel sizes and therefore makes this method suitable for producing microLED (pLED) structures. Advantageously the semiconductor structure may be epitaxially grown in a single growth process. Advantageously epitaxial growth results in very high quality material meaning that very small pixel sizes can be achieved with minimal defects.The method may further comprise a step to provide a third active layer on the second active layer, the third active layer comprising a quantum confinement structure, quantum dots or quantum wells, arranged to emit at a third wavelength. Advantageously the third active layer may be configured to emit at a different wavelength to the first active layer and the second active layer. The third wavelength may be shorter than the first or second wavelengths. Advantageously the method therefore produces tricolour emission, for example RGB emission which enables the full range of visible light generation.The method may further comprise a step to provide a fourth active layer, the fourth active layer comprising a quantum confinement structure, quantum dots or quantum wells, arranged to emit at a fourth wavelength. Advantageously the fourth active layer may be configured to emit at a different wavelength to the first active layer, the second active layer and the third active layer. The fourth wavelength may be shorter than the first, second and third wavelengths. Advantageously the method therefore produces tricolour emission, for example RGB emission which enables the full range of visible light generation.The fourth active layer may be grown before the first active layer. Advantageously this puts the shortest wavelength below longer wavelength active layers so that the shorter wavelength optically pumps the longer wavelengths. Optical pumping may be increased by growing a fourth reflective layer below the fourth active layer, that is on the side that is distal from the or each other active layer.Alternatively the fourth active layer may be grown after the first active layer, after the second active layer, or after the optional third active layer. Advantageously the fourth active layer may emit at a short wavelength which optically pumps one or more of the other active layers. Optical pumping may be increased by growing a fourth reflective layer on the side that is distal from the or each active layer to be pumped. Where the fourth active layer is grown at the top of the semiconductor structure the fourth reflective layer may therefore be grown on or over the fourth active layer.The method may further comprise a step to grow a first contact layer on the first active layer. The method may further comprise a step to grow a second contact layer on the second active layer. The method may further comprise a step to grow a third contact layer on the third active layer. The method may further comprise a step to grow an upper contact layer on the semiconductor structure. The contacts may be used in pairs to individually control (illuminate) the active layers.Each step to grow a layer may comprise epitaxial growth. Advantageously epitaxial growth results in high quality layers with minimal defects, a smooth surface, and beneficial properties. Each epitaxial growth may comprise growth in a molecular beam epitaxy (MBE) reactor or in a metal-organic chemical vapour deposition (MOCVD) reactor. Advantageously the whole method can be performed in a single reactor, which minimises the opportunities to introduce contaminants. Alternatively oneor more layer may be epitaxially grown in an MBE reactor and one or more layer may be grown in an MOCVD reactor. Advantageously this hybrid method means layers in which the properties are critical may be grown in the MBE reactor, in order to obtain the highest quality, whereas other layers can be grown in the MOCVD reactor which results in quicker growth.The present invention will be more fully described by way of example with reference to the accompanying drawings, in which:Figure 1 to Figure 10 each illustrate a semiconductor structure according to the present invention.Epitaxy or epitaxial means crystalline growth of material, usually via high temperature deposition. Epitaxy can be effected in a molecular beam epitaxy (MBE) tool in which layers are grown on a heated substrate in an ultra-high vacuum environment. Elemental sources are heated in a furnace and directed towards the substrate without carrier gases. The elemental constituents react at the substrate surface to create a deposited layer. Each layer is allowed to reach its lowest energy state before the next layer is grown so that bonds are formed between the layers. Epitaxy can also be performed in a metal-organic vapour phase epitaxy (MOVPE) tool, also known as a metal-organic chemical vapour deposition (MOCVD) tool. Compound metal-organic and hydride sources are flowed over a heated surface using a carrier gas, typically hydrogen. Epitaxial deposition occurs at much higher pressure than in an MBE tool. The compound constituents are cracked in the gas phase and then reacted at the surface to grow layers of desired composition.Deposition means the depositing of a layer on another layer or substrate. It encompasses epitaxy, chemical vapour deposition (CVD), powder bed deposition and other known techniques to deposit material in a layer.A compound material comprising one or more materials from group III of the periodic table with one or more materials from group V is known as a lll-V material. The compounds have a 1:1 combination of group III and group V regardless of the number of elements from each group. Subscripts in chemical symbols of compounds refer to the proportion of that element within that group. Thus Alo.zsGaAs means the group III part comprises 25% Al, and thus 75% Ga, whilst the group V part comprises 100% As.Crystalline means a material or layer with a single crystal orientation. In epitaxial growth or deposition subsequent layers with the same or similar lattice constant follow the registry of the previous crystalline layer and therefore grow with the same crystal orientation. In-plane is used herein to mean parallel to the surface of the substrate; out-of-plane is used to mean perpendicular to the surface of the substrate.Throughout this disclosure, as will be understood by the skilled reader, crystal orientation <100> means the face of a cubic crystal structure and encompasses

[0100] ,

[0010] and

[0001] orientations using the Miller indices. Similarly <0001> encompasses

[0001] and [000-1] except if the material polarity is critical. Integer multiples of any one or more of the indices are equivalent to the unitary version of the index. For example, (222) is equivalent to, the same as, (111).Substrate means a planar wafer on which subsequent layers may be deposited or grown. A substrate may be formed of a single element or a compound material, and may be doped or undoped. For example, common substrates include silicon (Si), gallium arsenide (GaAs), silicon germanium (SiGe), silicon germanium tin (SiGeSn), indium phosphide (InP), and gallium antimonide (GaSb).A substrate may be on-axis, that is where the growth surface aligns with a crystal plane. For example it has <100> crystal orientation. References herein to a substrate in a given orientation alsoencompass a substrate which is miscut by up to 20° towards another crystallographic direction, for example a (100) substrate miscut towards the (111) plane.Vertical or out of plane means in the growth direction; lateral or in-plane means parallel to the substrate surface and perpendicular to the growth direction.Doping means that a layer or material contains a small impurity concentration of another element (dopant) which donates (donor) or extracts (acceptor) charge carriers from the parent material and therefore alters the conductivity. Charge carriers may be electrons or holes. A doped material with extra electrons is called n-type whilst a doped material with extra holes (fewer electrons) is called p- type.Lattice matched means that two crystalline layers have the same, or similar, lattice spacing and so the second layer will tend to grow isomorphically on the first layer. Lattice constant is the unstrained lattice spacing of the crystalline unit cell. Lattice coincident means that a crystalline layer has a lattice constant which is, or is close to, an integer multiple of the previous layer so that the atoms can be in registry with the previous layer. Lattice mismatch is where the lattice constants of two adjacent layers are neither lattice matched nor lattice coincident. Such mismatch introduces elastic strain into the structure, particularly the second layer, as the second layer adopts the in-plane lattice spacing of the first layer. The strain is compressive where the second layer has a larger lattice constant and tensile where the second layer has a smaller lattice constant.Where the strain is too great the structure relaxes to minimise energy through defect generation, typically dislocations, known as slip, or additional interstitial bonds, each of which allows the layer to revert towards its lattice constant. The strain may be too great due to a large lattice mismatch or due to an accumulation of small mismatches over many layers. A relaxed layer is known as metamorphic, incoherent, incommensurate or relaxed, which terms are also commonly interchangeable.A pseudomorphic system is one in which a single-crystal thin layer overlies a single-crystal substrate and where the layer and substrate have similar crystal structures and nearly identical lattice constants. In a pseudomorphic structure the in-plane lattice spacing of the thin layer adopts the in-plane lattice constant of the substrate and is therefore elastically strained, either compressively where the layer has a larger lattice spacing than the substrate or tensilely where the layer has a smaller lattice spacing than the substrate. A pseudomorphic structure is not constrained in the out-of-plane direction and so the lattice spacing of the thin layer in this direction may change to accommodate the strain generated by the mismatch between lattice spacing. The thin layer may alternatively be described as "coherent", "commensurate", "strained" or "unrelaxed", which terms are often used interchangeably. In a pseudomorphic structure all the layers adopt the lattice spacing of the substrate in their respective in-plane lattice spacing.A layer may be monolithic, that is comprising bulk material throughout. Alternatively it may be porous for some or all of its thickness. A porous layer includes air or vacuum pores, with the porosity defined as the proportion of the area which is occupied by the pores rather than the bulk material. The porosity can vary through the thickness of the layer. For example, the layer may be porous in one or more sublayer. The layer may include an upper portion which is porous with a lower portion that is non-porous. Alternatively the layer may include one or more discrete, non-continuous portions (domains) that are porous with the remainder being non-porous (with bulk material properties). The portions may be non-continuous within the plane of a sublayer and / or through the thickness of the layer (horizontally and / or vertically in the sense of the growth direction). The portions may be distributed in a regular array or irregular pattern across the layer, and / or through it. The porosity maybe constant or variable within the porous regions. Where the porosity is variable it may be linearly varied through the thickness, or may be varied according to a different function such as quadratic, logarithmic or a step function.A porous layer means that pores have been formed through bulk material so that voids are intentionally introduced. Porosity is expressed in percentages which refers to the volume of bulk material which has been removed so 25% porosity means that the 25% of the equivalent volume of bulk material is voided.A fully depleted porous layer means a layer in which there are no charge carriers.A crystalline bixbyite oxide layer may be a rare earth oxide layer. The rare earth elements are scandium (Sc), yttrium (Y) and all of the lanthanoid series which is lanthanum (La), cerium (Ce), praseodymium (Pr), neodymium (Nd), promethium (Pm), samarium (Sm), europium (Eu), gadolinium (Gd), terbium (Tb), dysprosium (Dy), holmium (Ho), erbium (Er), thulium (Tm), ytterbium (Yb) and lutetium (Lu). The bixbyite oxides are bixbyite in crystal structure. Other bixbyite oxides include indium oxide (ln2O3), vanadium oxide (V2O5), iron oxide (Fe2O3), manganese oxide (Mn2O3) and ternary compounds of a rare earth, a metal and oxygen (RE-M-O).Where a device is described it should be understood that it will typically be formed on a circular substrate wafer of 4" (100mm), 6" (150mm), 8" (200mm), 12" (300mm) or greater diameter. After growth, deposition, bonding and other fabrication steps the devices are separated by dicing the wafer and layers into devices (chips) of appropriate dimensions. Typically tens, hundreds or thousands of devices are cut from a single wafer.The invention will now be described more particularly with reference to Figure 1 which shows a semiconductor structure 10. The semiconductor structure 10 may be fabricated into a light-emitting diode (LED), for example a microLED (pLED) meaning that each pixel is of the order of 1pm or less in diameter. There is a substrate 12. The substrate 12 may be silicon, for example Si (100) which is convenient for integration with driving electronics. Alternatively the substrate 12 may be gallium nitride (GaN) meaning subsequent layers are close to lattice matched and the semiconductor structure 10 exhibits minimal stress. Alternatively the substrate 12 may be sapphire (which is transparent to visible light), Si (111) or another material.Grown on the substrate 12, for example epitaxially grown, is a metal layer 14. The metal layer 14 may be configured as a contact, for example a low resistivity contact. The metal layer 14 may also be configured as a reflector, and may be a highly efficient reflector. In particular it may be configured to reflect light which is emitted from one or more subsequent layers of the semiconductor structure 10. For example, it may be configured to reflect light with a wavelength that is close to, preferably identical to, the wavelength of light emitted from a subsequent layer.Grown on the metal layer 14, for example epitaxially grown, is a first active layer 16. The first active layer 16 comprises quantum dots which are arranged to emit at a first wavelength X3. The first wavelength X3is preferably in the range 620nm to 740nm which is red light in the visible spectrum. For example the first wavelength Xi may be between 645nm and 700nm so that it emits a pure red. The first active layer 16 may comprise a matrix material in which quantum dots are suspended or positioned, for example GaN. The first active layer 16 may comprise an array of quantum dots which are closely packed. A quantum dot confines charges into a one-dimensional feature which increases the probability of charge recombination and therefore increases the emission power. Typically a layer of quantum dots resembles an array of hemispherical protuberances on the previous layer; they may be tessellated and / or distorted so that there are no gaps between the quantum dots.The first active layer 16 may comprise indium gallium nitride, InGaN. The composition may be lnxGai.xN, where x=0.4 such that it emits at the desired first wavelength Xi, approximately 630nm depending on the size of the quantum dots. In some embodiments the first wavelength Xi is the longest wavelength of those emitted by layers in the semiconductor structure 10.The first active layer 16 may alternatively comprise a quaternary compound of indium scandium gallium nitride, (lnySci.y)xGai.xN, where x and y are each non-zero (i.e. 0<x,y<l). The quaternary compound may have wurtzite structure. For example a quantum dot composition having x=0.7 and y=0.7, thus (Ino.ySco.aJo.zGao.sN, will emit red light at approximately 630nm. In the more general case the quaternary compound can be expressed as (lnySci.y)xGai.xN where 0<x,y<l which encompasses the previously described lnxGai.xN when y=l. By adding Sc the concentration of In can be reduced which is beneficial, particularly for red light emission, since it reduces the required In concentration therefore reducing the probability of In segregation.Grown or provided on, for example epitaxially grown on or bonded to, the first active layer 16 is a second active layer 20. The second active layer 20 comprises quantum dots which are arranged to emit at a second wavelength X2. The second wavelength X2is preferably in the range 500nm to 570nm which is green light in the visible spectrum. For example the second wavelength X2may be between 525nm and 550nm so that it emits a pure green. The second active layer 20 may comprise a matrix material in which quantum dots are suspended or positioned, for example GaN. Alternatively the second active layer 20 may comprise an array of closely packed quantum dots.The second active layer 20 may comprise InGaN. The composition may be lnxGai.xN, where x=0.3 such that it emits at the desired second wavelength X2, approximately 530nm depending on the size of the quantum dots. Preferably the second wavelength X2is shorter than the first wavelength Xi. Light emitted from the first active layer 16 is able to travel through the optional first contact layer 18 and through the second active layer 20 to be emitted through an upper surface of the semiconductor structure 10. Thus the second active layer 20 is transparent to the first wavelength X2.The second active layer 20 may alternatively comprise a quaternary compound of indium scandium gallium nitride, (lnySci.y)xGai.xN, where x and y are each non-zero (i.e. 0<x,y<l). The quaternary compound may have wurtzite structure. For example a quantum dot composition having x=0.5 and y=0.7, thus (Ino.ySco.sJo.sGao.sN, will emit green light at approximately 530nm. In the more general case the quaternary compound can be expressed as (lnySci.y)xGai.xN where 0<x,y,<l which encompasses the previously described lnxGai.xN when y=l. By adding Sc the concentration of In can be reduced which is beneficial, particularly for red light emission but also for green light emission, since it reduces the required In concentration therefore reducing the probability of In separation.Advantageously, the semiconductor structure 10 is able to emit light at two different wavelengths with appropriate control and activation of the first and second active layers 16, 20. However, it can be grown in a single sequence in an epitaxial reactor without the requirement to remove the semiconductor structure 10 and perform additional steps in a different reactor or tool. In the alternative the second active layer 20 may be bonded to the first active layer 16 having been grown (or bonded) in a separate process.Optionally there may be a third active layer 24 grown or provided on, for example epitaxially grown on or bonded to, the previous layers. The third active layer 24 comprises quantum dots which are arranged to emit at a third wavelength X3. The third wavelength X3 is preferably in the range 450nm to 505nm which is blue light in the visible spectrum. For example the third wavelength X3may be between 450nm and 470nm so that it emits a pure blue. The third active layer 24 may comprise amatrix material in which quantum dots are suspended or positioned, for example GaN. Alternatively the third active layer 24 may comprise an array of closely packed quantum dots.The third active layer 24 may comprise InGaN. The composition may be lnxGai.xN, where x=0.2 such that it emits at the desired third wavelength X3, approximately 460nm depending on the size of the quantum dots. Preferably the third wavelength X3 is shorter than the first wavelength Xi or the second wavelength X2. Light emitted from the first active layer 16 is able to travel through the subsequent layers, second active layer 20, and third active layer 24 to be emitted through an upper surface of the semiconductor structure 10. Similarly light emitted from the second active layer 20 is able to travel through the third active layer 24 to be emitted through an upper surface of the semiconductor structure 10. Thus the third active layer 24 is transparent to the first wavelength Xi and second wavelength X2.The third active layer 24 may alternatively comprise GaN or aluminium gallium nitride, AlxGai.xN, quantum dots. For the latter x may equal 0.1. Advantageously GaN quantum dots in an AIGaN matrix have a lower growth temperature than AIGaN dots in an AIN matrix, which may decrease the probability of In segregation in InGaN layers below the third active layer 24. Advantageously the growth temperatures of the GaN dots and AIGaN matrix are similar whereas InN (or high-ln content InGaN) and AIN have incompatible growth temperatures and therefore an increased probability of In segregation. Alternatively the third active layer 24 may comprise a quaternary compound of aluminium indium gallium nitride, (Alylni.y)xGai.xN. where 0<x,y<l. The quaternary compound may have wurtzite structure. In the case that y=l this is the tertiary compound AlxGai.xN as mentioned above. By using a quaternary compound the proportions of Al and In can be optimised for bandgap, therefore emission wavelength, and lattice mismatch.Therefore the semiconductor structure 10 may have three active layers which each emit light at a different wavelength. However, it can be grown in a single sequence in an epitaxial reactor without the requirement to remove the semiconductor structure 10 and perform additional steps in a different reactor or tool. In the alternative the second active layer 20 and subsequent layers may be bonded to the first contact layer 18 having been grown (or bonded) in a separate process.Optionally there may be a fourth active layer 28 grown or provided on, for example epitaxially grown on or bonded to, the previous layers. The fourth active layer 28 comprises quantum dots which are arranged to emit at a fourth wavelength X4. The fourth wavelength X4is preferably less than 450nm, for example in the range 200nm to 450nm which includes ultraviolet light in the electromagnetic spectrum. For example the fourth wavelength X4may be between 280nm and 400nm. Ultraviolet light can be used to optically pump longer wavelength layers. That is, photons with short wavelength excite electrons in layers with a narrower bandgap (longer wavelength). The fourth active layer 28 may comprise a matrix material in which quantum dots are suspended or positioned, for example (AI)GaN. Alternatively the fourth active layer 28 may comprise an array of closely packed quantum dots.The fourth active layer 28 may comprise AIGaN. The composition may be AlxGai.xN, where x=0.4 such that it emits at the desired fourth wavelength X4, 279nm. Preferably the fourth wavelength X4is shorter than the first wavelength Xi, the second wavelength X2or the third wavelength X3. Light emitted from the first active layer 16 is able to travel through the subsequent layers including the second active layer 20, third active layer 24 and fourth active layer 28 to be emitted through an upper surface of the semiconductor structure 10. Similarly light emitted from the second active layer 20 is able to travel through the third active layer 24 and fourth active layer 28 to be emitted through an upper surface of the semiconductor structure 10. Also light emitted from the third active layer 24 isable to travel through the fourth active layer 28. Thus the fourth active layer 28 is transparent to the first wavelength Xi, second wavelength X2 and third wavelength X3.The semiconductor structure 10 therefore has up to four active layers which each emit light at a different wavelength. However, it can be grown in a single sequence in an epitaxial reactor without the requirement to remove the semiconductor structure 10 and perform additional steps in a different reactor or tool. In the alternative the second active layer 20 and subsequent layers may be bonded to the first contact layer 18 having been grown (or bonded) in a separate process.Each of the first, second, third and fourth active layers 16, 20, 24, 28 may comprise a plurality of periods with alternate sublayers comprising quantum dots, for example comprised of InGaN or AIGaN, and GaN barrier layers. By providing multiple layers of quantum dots the intensity of light emitted is higher than for a single layer.The fourth active layer 28 may be provided, not to emit light out of the semiconductor structure 10, but to optically excite one or more of the other active layers 16, 20, 24. This is because short wavelengths, such as ultraviolet or blue, will optically pump longer wavelengths, such as green or red. Thus the fourth active layer 28 may be arranged to emit in the ultraviolet wavelength range in order to optically excite the blue emitting third active layer 24. The ultraviolet emission may also or alternatively optically excite the green emitting second active layer 20 and / or the red emitting first active layer 16. The shorter the wavelength, the higher the photon energy. Therefore electrons can be excited from the conduction band into the valence band of an active region by photons emitted by a different action region. Emission of light from the fourth active layer 28 is multidirectional and therefore can simultaneously be emitted from the semiconductor structure 10 as useful blue light and be an excitation source for longer wavelengths in the first, second and / or third active layers 16, 20, 24. The optical excitation of the longer wavelength active layers may be in addition to electrical excitation. Advantageously more emission is caused by greater excitation as more electron-hole pairs are recombined to emit photons. Alternatively the optical excitation may be the sole source of excitation meaning that only the shortest, or multiple shorter, wavelength active layers need be electrically excited to excite all active layers. Advantageously fewer contacts are needed in the semiconductor structure 10.The third active layer 24 may be provided exclusively or partially to optically excite one or more of the first and second active layers 16, 20. Emission of light from the third active layer 24 is multidirectional and therefore can simultaneously be emitted from the semiconductor structure 10 as useful blue light and be an excitation source for longer wavelengths in the first and / or second active layers 16, 20. The optical excitation of the longer wavelength active layers may be in addition to electrical excitation. Advantageously more emission is caused by greater excitation as more electronhole pairs are recombined to emit photons. Alternatively the optical excitation may be the sole source of excitation meaning that only the shortest, or multiple shorter, wavelength active layers need be electrically excited to excite all active layers. Advantageously fewer contacts are needed in the semiconductor structure 10.The second active layer 20 may additionally optically excite the first active layer 16. Emission of light from the second active layer 20 is multidirectional and therefore can simultaneously be emitted from the semiconductor structure 10 as useful green light and be an excitation source for longer wavelengths in the first active layer 16. The optical excitation of the longer wavelength first active layer 16 may be in addition to electrical excitation. Advantageously more emission is caused by greater excitation as more electron-hole pairs are recombined to emit photons. Alternatively the optical excitation may be the sole source of excitation meaning that only the shortest, or multipleshorter, wavelength active layers need be electrically excited to excite all active layers. Advantageously fewer contacts are needed in the semiconductor structure 10.The first active layer 16 may thus be optically excited (pumped) by each of the second, third and fourth active layers 20, 24, 28 meaning that it may get maximum increase in excitation and therefore maximise electron-hole recombination resulting in maximum light emission at red wavelengths. Advantageously the additional pumping of the first active layer 16 may counteract the weaker emission of red light inherent from GaN-based materials. Similarly the green second active layer 20 may be optically excited by the third and fourth active layers 24, 28 meaning its emission has a double boost. Green emission may be weaker than blue emission from GaN-based materials and therefore the additional optical excitation may help to balance the emission intensity between the colours.The semiconductor structure 10 may include an upper contact 30 at the top of the structure, that is grown or bonded to the uppermost layer. The upper contact 30 is electrically conductive and may be optically transparent. For example, it may comprise indium oxide InjOa. The upper contact layer 30 together with the metal layer 14 controls the first active layer 16, second active layer 20, optional third active layer 24 and optional fourth active layer 28 by enabling a current to be applied across the layers. Thus the emitted light is a combination of the light emitted from each of the active layers. If the first, second and third wavelengths Xi, X2, X3 are tuned to emit red, green and blue light respectively then the resultant light emitted from the semiconductor structure 10 can be tuned to emit across the visible colour spectrum, including white light.The upper contact 30 may be non-transparent to short wavelength (ultraviolet) light. In this case it will reflect light emitted from the optional fourth active layer 28 to redirect it from emission through the top of the semiconductor structure 10 to pass into the semiconductor structure 10 where it optically excites one or more of the other active layers 16, 20, 24.Figure 2 is similar to Figure 1 except that it additionally shows more optional layers. Grown on the first active layer 16, for example epitaxially grown, is an optional first contact layer 18. The first contact layer 18 together with the metal layer 14 controls the first active layer 16 by enabling a current to be applied across the first active layer 16. The first contact layer 18 may comprise a highly doped nitride. The first contact layer 18 may be highly p-doped, p++, or may be highly n-doped, n++. The first contact layer 18, where present, additionally acts as a contact for the second active layer 20.Grown or provided on, for example epitaxially grown on or bonded to, the second active layer 20 is an optional second contact layer 22. The second contact layer 22 in combination with the optional first contact layer 18 enables control of the second active layer 20 by applying a current across it. The second contact layer 22 may comprise a highly doped nitride. The second contact layer 22 may be highly p-doped, p++, or may be highly n-doped, n++.Grown or provided on, for example epitaxially grown on or bonded to, the third active layer 24 is an optional third contact layer 26. The third contact layer 26 in combination with the second contact layer 22 enables control of the third active layer 24 by applying a current across it. The third contact layer 26 may comprise a highly doped nitride. The third contact layer 26 may be highly p-doped, p++, or may be highly n-doped, n++.The third contact layer 26 may be further configured to reflect some or all of the light emitted from the third active layer 24 towards the second active layer 20 and the first active layer 16. By reflecting the light the third active layer 24 can be used to optically pump the first active layer 16 and / or the second active layer 20. The second contact layer 22 may be configured to permit blue light from the third active layer 24 to pass therethrough. Similarly the first contact layer 18 may be configured topermit blue light from the third active layer 24 and / or green light from the second active layer 20 to pass therethrough to optically pump the first active layer 16.In the absence of the first and second contact layers 18, 22 the third contact layer 26 may alternatively act with the epitaxial metal layer 14 to inject carriers to each of the active layers 16, 20, 24. Furthermore, the third contact layer 26 in combination with the upper contact layer 30 enables control of the fourth active layer 28 by applying a current across it.An optional lower contact layer 32 may be grown on the metal layer 14 so that it is below the first active layer 16. The lower contact layer 32 together with the first contact layer 18 controls the first active layer 16 by enabling a current to be applied across the first active layer 16. The lower contact layer 32 may comprise a highly doped nitride. The lower contact layer 32 may be highly p-doped, p++, or may be highly n-doped, n++. Lower contact layer 32 may also function as a buffer layer for growth of subsequent layers. The metal layer 14 is a low resistivity contact and may provide more uniform current spreading.Advantageously, each active layer can be independently controlled by providing a contact layer above and below it so that current can be applied to one active layer but not to the other active layers, or a different level of current can be applied simultaneously to different ones of the active layers.Figure 3 shows the semiconductor structure 10 of Figure 2 with electrodes coupled with the contact layers. A lower electrode 46 may be provided that is coupled to the epitaxial metal layer 14 or the lower contact layer 32. A first electrode 48 is coupled to the optional first contact layer 18. The lower and first electrodes 46, 18 together apply a voltage across the first active layer 16 in order to inject carriers. Alternatively the lower electrode 46 may be formed or grown as part of the epitaxial metal layer 14 or the lower contact layer 32, and / or the first electrode 34 may be formed or grown as part of the first contact layer 18.A second electrode 50 is coupled to the optional second contact layer 22. The second electrode 50 and first electrode 48 together apply a voltage across the second active layer 20 in order to inject carriers. Alternatively the second electrode 50 is formed or grown as part of the second contact layer 22.A third electrode 52 is coupled to the optional third contact layer 26. It acts with the second electrode 50 to apply a voltage across the third active layer 24 to inject carriers. Alternatively the third electrode 52 may be formed or grown as part of the third contact layer 26.An upper electrode 54 is coupled to the upper contact layer 30. It acts with the third electrode 52 to apply a voltage across the fourth active layer 28 to inject carriers, or with the second electrode 50 to apply a voltage across the third active layer 24 when no fourth active layer 28 is present, or with the first electrode 48 to apply a voltage across the second active layer 20 when no third active layer 24 is present. Alternatively the upper electrode 54 may be formed or grown as part of the upper contact layer 30.Alternatively the upper electrode 54, coupled to the upper contact layer 30, and the lower electrode 46, coupled to the epitaxial metal layer 14 or lower contact layer 32, act to apply a voltage across all the active layers 16, 20, 24, 28 to inject carriers.Figure 4 is a similar semiconductor structure 10 as illustrated in Figure 3. Thus the substrate 12, epitaxial metal layer 14, first active layer 16, second active layer 20, optional third active layer 24, optional fourth active layer 28 and upper contact layer 30 are the same as described with respect to Figure 3. However, the optional first and / or second and / or third contact layers 18, 22, 26 are different.The first contact layer 18 may comprise a first sublayer 18a and a second sublayer 18b. Between the first and second sublayers 18a, 18b is a first insulation sublayer 34. The first sublayer 18a and second sublayer 18b may have the same composition as each other. The first insulation sublayer 34 may comprise a lattice matched, wide bandgap material. It acts to electrically insulate the first sublayer 18a from the second sublayer 18b such that the first sublayer 18a acts as a contact for the first active layer 16 with the epitaxial metal layer 14 whereas the second sublayer 18b acts as a contact for the second active layer 20. The first insulation sublayer 34 may be optically transparent. It should be optically transparent to red light emitted from the first active layer 16. It may also be optically transparent to one or more of the shorter wavelengths emitted by the second, third and fourth active layers 20, 24, 28 so that these layers may optically pump the first active layer 16.The first sublayer 18a may be the opposite conduction type to the second sublayer 18b. Thus the first sublayer 18a may be p-type whilst the second sublayer 18b is n-type or vice versa. This enables the layers to be used to control the active layers therebetween, for example a p-type first sublayer 18a in combination with an n-type lower contact 32 can control the first active layer 16.The first insulation sublayer 34 may be undoped GaN alloyed with one or more of Al, In, Sc and B. For example, the first insulation sublayer 34 may comprise AlxGai.xN where x is greater than 50%, for example 60%. Having a large Al content ensures a wide bandgap insulator material. Alternatively the first insulation sublayer 34 may be AlxSci.xN in wurtzite crystal orientation. For example, x may be 0.82 such that the first insulation sublayer 28 is closely lattice matched to the adjacent first and second contact sublayers 18a, 18b, the structure is wurtzite and the bandgap is greater than, for example, 4.5eV. Such a bandgap is large enough to prevent charge carriers moving across the layers in normal operation. Alternatively the first insulation sublayer 34 may be a dielectric material such as an oxide, glass, or a crystalline rare earth oxide such as scandium oxide (Sc2O3). The first insulation sublayer 34 is transparent to visible light, particularly to light at the wavelengths emitted by the first active layer 16.Similarly the second contact layer 22 may comprise a first sublayer 22a and a second sublayer 22b. Between the first and second sublayers 22a, 22b is a second insulation sublayer 36. The first sublayer 22a and second sublayer 22b may have the same composition as each other. The second insulation sublayer 36 may comprise a lattice matched, wide bandgap material. It acts to electrically insulate the first sublayer 22a from the second sublayer 22b such that the first sublayer 22a acts as a contact for the second active layer 20 with the second sublayer 18b or first contact layer 18 whereas the second sublayer 22b acts as a contact for the third active layer 24.The first sublayer 22a may be the opposite conduction type to the second sublayer 22b. Thus the first sublayer 22a may be p-type whilst the second sublayer 22b is n-type or vice versa. This may help to simplify the electric circuit to control the semiconductor structure 10.The second insulation sublayer 36 may be undoped GaN alloyed with one or more of Al, In, Sc and B. For example, the second insulation sublayer 36 may comprise AlxGai.xN where x is greater than 50%, for example 60%, so that it is a wide bandgap insulator material. Alternatively the second insulation sublayer 36 may be AlxSci.xN in wurtzite crystal orientation. For example, x may be 0.82 such that the second insulation sublayer 36 is closely lattice matched to the adjacent first and second contact sublayers 22a, 22b and the bandgap is greater than 4.5eV. Such a bandgap is large enough to prevent charge carriers moving across the layers in normal operation. Alternatively the second insulation sublayer 36 may be a dielectric material such as an oxide, glass, or a crystalline rare earth oxide such as scandium oxide (Sc2O3). The second insulation sublayer 36 is transparent to visible light,particularly to light at the wavelengths emitted by the first active layer 16 and the second active layer 20.The second insulation sublayer 36 may be optically transparent. It should be optically transparent to green light emitted from the second active layer 20 and red light emitted from the first active layer 16. It may also be optically transparent to one or more of the shorter wavelengths emitted by the third and fourth active layers 24, 28 so that these layers may optically pump the first active layer 16 and / or the second active layer 20.Similarly the third contact layer 26 may comprise a first sublayer 26a and a second sublayer 26b. Between the first and second sublayers 26a, 26b is a third insulation sublayer 38. The first sublayer 26a and second sublayer 26b may have the same composition as each other. The third insulation sublayer 38 may comprise a lattice matched, wide bandgap material. It acts to electrically insulate the first sublayer 26a from the second sublayer 26b such that the first sublayer 26a acts as a contact for the third active layer 24 with the second sublayer 22b whereas the second sublayer 26b acts as a contact for the fourth active layer 28.The first sublayer 26a may be the opposite conduction type to the second sublayer 26b. Thus the first sublayer 26a may be p-type whilst the second sublayer 26b is n-type or vice versa. This may help to simplify the electric circuit to control the semiconductor structure 10.The third insulation sublayer 38 may be GaN alloyed with one or more of Al, In, Sc and B. For example, the third insulation sublayer 38 may comprise AlxGai.xN where x is greater than 50%, for example 60%, so that it is a wide bandgap insulator material. Alternatively the third insulation sublayer 38 may be AlxSci.xN in wurtzite crystal orientation. For example, x may be 0.82 such that the third insulation sublayer 38 is closely lattice matched to the adjacent first and second contact sublayers 26a, 26b and the bandgap is greater than 4.5eV. Such a bandgap is large enough to prevent charge carriers moving across the layers in normal operation. Alternatively the third insulation sublayer 38 may be a dielectric material such as an oxide, glass, or a crystalline rare earth oxide such as scandium oxide (SC2O3). The third insulation sublayer 38 is transparent to visible light, particularly to light at the wavelengths emitted by the first active layer 16, the second active layer 20 and the third active layer 24.The third insulation sublayer 38 may be optically transparent. It should be optically transparent to blue light emitted from the third active layer 24, green light emitted from the second active layer 20 and red light emitted from the first active layer 16. It may also be optically transparent to the shorter wavelength emitted by the fourth active layer 28 so that this layer may optically pump the first active layer 16 and / or the second active layer 20 and / or the third active layer 24.Each contact sublayer has its own contact electrode. Thus the first sublayer 18a of the first contact is coupled to electrode 47 and works in combination with the lower contact 32 and its electrode 46 to control the first active layer 16. The second sublayer 18b of the first contact is coupled to electrode 48 and works in combination with the first sublayer 22a of the second contact that is coupled to electrode 49 to control the second active layer 20. The second sublayer 22b of the second contact is coupled to electrode 50 and works in combination with the first sublayer 26a of the third contact that is coupled to electrode 51 to control the optional third active layer 24. The second sublayer 26b of the third contact is coupled to electrode 52 and works in combination with the upper contact 30 and its electrode 54 to control the optional fourth active layer 28.In the alternative the first contact layer 18 may comprise two sublayers forming a p++-n tunnel junction. Additionally or alternatively the second contact layer 22 may comprise two sublayersforming a p++-n tunnel junction. Additionally or alternatively the third contact layer 26 may comprise two sublayers forming a p++-n tunnel junction. A tunnel junction may allow a portion of the current flowing into the n-type sublayer to flow as a current of opposite sign into the adjacent p-type sublayer. Thus a single electrode may supply current to two adjacent active layers. Alternatively a tunnel junction may allow the current from an active region to flow into the next active region as a current of opposite sign, thus avoiding the need for additional electrodes between the active regions. Advantageously, in either case the number of layers / electrodes is reduced.Figure 5 is a similar semiconductor structure 10 to that in Figure 1 with additional layers. Firstly, on, for example grown on, the metal layer 14 is a first reflective layer 39 that reflects wavelengths close to the first wavelength Xi emitted by the first active layer 16. This means that any light emitted by the first active layer 16 towards the substrate 12 is reflected by the first reflective layer 39 and thus redirected in the desired emission direction, for example through the optional transparent upper contact layer 30 at the top of the semiconductor structure 10 as shown. The first reflective layer 39 may be tuned to reflect light emitted by the first active layer 16 meaning it has a wavelength close to, preferably identical to, the first wavelength Xi. The metal layer 14 may be configured as an additional reflector as well as its primary purpose as an electrical contact in this arrangement.The first reflective layer 39 may comprise multiple periods of two sublayers having different refractive indices. The compositions of the sublayers are chosen to reflect light of a similar wavelength to the first active layer 16. For example, the first reflective layer 39 may comprise a distributed Bragg reflector (DBR) comprising sublayers of AI(Ga)N and GaN. For example, a first sublayer of 67.8nm of GaN and a second sublayer of 78.7nm of AIN repeated 10 times will give a theoretical reflectivity of 86% at a typical red wavelength (630nm). Repeating the sublayers 20 times will increase reflectivity to 99%. Alternatively a ternary composition can be used for one of the sublayers which will decrease the difference in refractive indices and therefore increase the required number of periods to achieve a given level of reflectivity (and thus complexity and growth time), but reduces lattice mismatch and / or strain effects.Advantageously reflecting the red light emitted from the first active layer 16 increases the amount of usefully emitted light for a small increase in number of layers and thickness of the semiconductor structure 10. Alternatively the metal layer 14 can be configured to be a reflective layer with a broad spectrum of wavelengths so that it reflects light emitted from two or more of the active layers 16, 20, 24, 28.When the metal layer 14 is arranged to be a first reflective layer it may be necessary to provide a separate lower contact layer 32 rather than to use the periodically varying metal layer 14 as the contact.Grown on the first active layer 16, for example epitaxially grown, or bonded thereto is a second reflective layer 40. The second reflective layer 40 comprises multiple periods of two sublayers having different refractive indices. The compositions of the sublayers are chosen to reflect light of a similar wavelength to that generated by the second active layer 20. Thus the first reflective layer 40 acts to reflect emission close to the second wavelength X2. In particular it reflects light which is emitted towards the substrate 12, thus preventing propagation in this direction, and redirects it in the opposite, preferred, emission direction for example through the optional transparent upper contact layer 30. The second reflective layer 40 may have two or more periods of sublayers forming a superlattice structure. The second reflective layer 40 may be tuned to reflect light emitted by the second active layer 20 meaning it has a wavelength close to, preferably identical to, the wavelength of the second active layer 20. For example, the second reflective layer 40 may comprise a distributedBragg reflector (DBR) comprising sublayers of AI(Ga)N and GaN. For example, a first sublayer of 54.8nm of GaN and a second sublayer of 64.4nm of AIN repeated 10 times will give a reflectivity of 89% at a typical green wavelength (520nm). Repeating the sublayers 20 times will increase reflectivity to 99%. Alternatively a ternary composition can be used for one of the sublayers which will decrease the difference in refractive indices and therefore increase the required number of periods to achieve a given level of reflectivity (and thus complexity and growth time), but reduces lattice mismatch and / or strain effects.Advantageously using the second reflective layer 40 results in more of the emitted light being directed in the desired emission direction so that there is greater intensity of light of the second wavelength X2. Thus the intensity of light emitted at the second wavelength X2from the second active layer 20 can match the intensity of the light emitted at the first wavelength Xi from the first active layer 16. A further advantage is that this is achieved with both active layers 16, 20 formed in the same material family, i.e. GaN-based materials. Formerly it was necessary to produce green emission from a GaN-based material and red from GaAs-based material with the consequent challenge of integrating both colours in one device.The second reflective layer 40 must be reasonably transparent to emission at the first wavelength Xi from the first active layer 16, although some light from the first active layer 16 will be absorbed or reflected.Grown on the second active layer 20, for example epitaxially grown, or bonded thereto is a third reflective layer 42. The third reflective layer 42 comprises multiple periods of two sublayers having different refractive indices. The compositions of the sublayers are chosen to reflect light at a different wavelength to the second reflective layer 40, at a similar wavelength to that generated by the third active layer 24. Thus the third reflective layer 42 acts to reflect emission at the third wavelength X3. In particular it reflects light which is emitted towards the substrate 12, thus preventing propagation in this direction, and redirects it in the opposite, preferred, emission direction for example through the optional transparent upper contact layer 30. The third reflective layer 42 may have two or more periods of sublayers forming a superlattice structure.Advantageously using the third reflective layer 42 results in more of the emitted light being directed in the desired emission direction so that there is greater intensity of light of the third wavelength X3. Thus the intensity of light emitted at the third wavelength X3 from the third active layer 24 can match the intensity of the light emitted at the first wavelength Xi from the first active layer 16 and / or the light emitted at the second wavelength X2from the second active layer 20. A further advantage is that this is achieved with all the active layers 16, 20, 24 formed in the same material family, i.e. GaN-based materials. Formerly it was necessary to produce green and blue emission from a GaN-based material and red from GaAs-based material with the consequent challenge of integrating all three colours in one device.The third reflective layer 42 may be configured to reflect light at both the third and fourth wavelengths X3, X4, emitted from the third and fourth active layers 24, 28. The third reflective layer 42 must be transparent to emission at the first wavelength Xi from the first active layer 16 and at the second wavelength X2from the second active layer 20. The third reflective layer 42 may be tuned to reflect light emitted by the third active layer 24 meaning it has a wavelength close to, preferably identical to, the wavelength of the third active layer 24. For example, the third reflective layer 42 may comprise a distributed Bragg reflector (DBR) comprising sublayers of AI(Ga)N and GaN. For example, a first sublayer of 47.4nm of GaN and a second sublayer of 56.6nm of AIN repeated 10 times will give a reflectivity of 93% at a typical blue wavelength (460nm). Repeating the sublayers 20 times willincrease reflectivity to over 99%. Alternatively a ternary composition can be used for one of the sublayers which will decrease the difference in refractive indices and therefore increase the required number of periods to achieve a given level of reflectivity (and thus complexity and growth time), but reduces lattice mismatch and / or strain effects.Grown on the fourth active layer 28, for example epitaxially grown, or bonded thereto is a fourth reflective layer 44. The fourth reflective layer 44 comprises multiple periods of two sublayers having different refractive indices. The compositions of the sublayers are chosen to reflect light at a different wavelength to the second reflective layer 40 and third reflective layer 42, at a similar wavelength to that generated by the fourth active layer 28. Thus the fourth reflective layer 44 acts to reflect emission at the fourth wavelength X4. In particular it reflects light which is emitted towards the optional transparent upper contact layer 30, thus preventing propagation in this direction, and redirects it in the opposite, preferred, emission direction for example towards the first, second and third active layers 16, 20, 24. Thus the fourth reflective layer 44 reflects short wavelength light, for example at ultraviolet (UV) frequencies to optically pump one or more of the other active layers 16, 20, 24. The fourth reflective layer 44 may have two or more periods of sublayers forming a superlattice structure (DBR).The fourth reflective layer 44 must be transparent to emission at the first wavelength Xi from the first active layer 16, at the second wavelength X2from the second active layer 20, and at the third wavelength X3 from the third active layer 24. The fourth reflective layer 44 may be tuned to reflect light emitted by the fourth active layer 28 meaning it has a wavelength close to, preferably identical to, the wavelength of the fourth active layer 28. For example, the fourth reflective layer 44 may comprise a distributed Bragg reflector (DBR) comprising sublayers of AI(Ga)N and GaN. For example, a first sublayer of 40nm of GaN and a second sublayer of 50nm of AIN repeated 10 times will give a reflectivity of 94% at a typical UV wavelength (390nm). Repeating the sublayers 20 times will increase reflectivity to 99%. Alternatively a ternary composition can be used for one of the sublayers which will decrease the difference in refractive indices and therefore increase the required number of periods to achieve a given level of reflectivity (and thus complexity and growth time), but reduces lattice mismatch and / or strain effects.Advantageously using the fourth reflective layer 44 results in more of the emitted light being directed in the desired emission direction so that there is greater intensity of light of the fourth wavelength X4. A further advantage is that this is achieved with all the active layers 16, 20, 24, 28 formed in the same material family, i.e. GaN-based materials. Formerly it was necessary to produce green and blue emission from a GaN-based material and red from GaAs-based material with the consequent challenge of integrating all three colours in one device.Each reflective layer 14, 40, 42, 44 will have reflectivity fringes on either side of its design wavelength. These fringes will interact with the peak (design) wavelength of the other reflective layers 14, 40, 42, 44 and behave as noise. Therefore the reflective layers 14, 40, 42, 44 require careful design to account for this effect.Thus each reflective layer 14, 40, 42, 44 reflects light from the active layer adjacent to it and thereby increases the intensity of the light emitted from the semiconductor structure 10, for example through the optional transparent upper contact layer 30. Semiconductor structures 10 may have one or more of the reflective layers. The configuration of the or each reflective layer may be designed so that the relative intensity of visible light from each active layer 16, 20, 24 is approximately equal.One or more of the reflective layers 14, 40, 42, 44 may be combined with the contact layers 32, 18, 22, 30 shown in and discussed in relation to Figure 2, Figure 3 and Figure 4. In this case each reflective layer 14, 40, 42, 44 is grown on or bonded onto its respective contact layer 32, 18, 22 (below upper contact layer 30) so that the reflective layer 14, 40, 42, 44 is closer to the respective active layer 16, 20, 24, 28. This means that the light emitted towards the substrate 12 is redirected efficiently even if the contact layer and / or insulation sublayer 34, 36, 38 partially absorbs the relevant wavelength.Figure 6 illustrates an alternative semiconductor structure 10 which is similar to that shown in Figure 1. Grown on or bonded to the substrate 12 is an epitaxial metal layer 14 which acts as a contact and / or a reflective layer. It may also function as a current spreading layer. There is a first active layer 16 which is configured to emit light at the first wavelength, Xi. For example, the first wavelength Xi may be red visible light in the range 620nm to 740nm. For example the first wavelength Xi may be between 645nm and 700nm so that it emits a pure red. There is a second active layer 20 grown on or over the first active layer 16 which is configured to emit light at the second wavelength X2. For example, the second wavelength Xjmay be green visible light in the range 500nm to 570nm which is green light in the visible spectrum. For example the second wavelength X2may be between 525nm and 550nm so that it emits a pure green.Optionally there is a third active layer 24 grown on or over the second active layer 20. The third active layer 24 is configured to emit light at the third wavelength X3. For example, the third wavelength X3may be blue visible light in the range 450nm to 505nm which is blue light in the visible spectrum. For example the third wavelength X3 may be between 450nm and 470nm so that it emits a pure blue. Grown on or over the third active layer 24, if present, or over the second active layer 20 if not is an optional transparent upper contact 30.Figure 6 differs from Figure 1 in that the optional fourth active layer 28 is positioned below the first active layer 16 rather than at the top of the semiconductor structure 10. The fourth active layer 28 is configured to emit light at the fourth wavelength X4. For example, the fourth wavelength X4may be less than 450nm, for example 200nm to 450nm. Since ultraviolet light has a shorter wavelength than visible light, it optically pumps the longer wavelength active layers 16, 20, 24. By positioning the fourth active layer 28 below the first active layer 16 the majority of the optical pumping is experienced by the first active layer 16. Some of the (UV) light will pass through the first active layer 16 to impact on, and optically pump, the second active layer 20 and some may also pass through this layer to optically pump the optional third active layer 24. Advantageously the optical pumping of the first active layer 16 by the fourth active layer 28 below it may increase the intensity to match that of the other colours. This may compensate when the light generated by the first active layer 16 is of a lower intensity than that generated and emitted by the second and third active layers 20, 24.Figure 7 has the same semiconductor structure 10 as shown in Figure 6 with additional layers. In particular the semiconductor structure 10 includes the first reflective layer 39, second reflective layer 40 and third reflective layer 42 below the first, second and third active layers 16, 20, 24 respectively as described with respect to Figure 5. Each of the reflective layers is configured to reflect the wavelength emitted by the active layer above it in order to redirect it to be emitted through the transparent upper contact 30. Furthermore, each reflective layer is substantially transparent to the wavelength or wavelengths emitted by the one or more active layers below it in the semiconductor structure 10.Below the fourth active layer 28, grown on or over the epitaxial metal layer 14, is a fourth reflective layer 44. The fourth reflective layer 44 is as described with respect to Figure 5. The fourth reflective layer 44 reflects the light emitted at the fourth wavelength X4so that more of it passes up through thefirst active layer 16, and potentially beyond into the second active layer 20 and even into the third active layer 24. The (UV) light therefore optically pumps at least the first active layer 16 and therefore increases the intensity of light generated in and emitted from that layer. By positioning the fourth reflective layer 44 below the fourth active layer 28 the maximum amount of light emitted at the fourth wavelength X4is directed through the other active layers to optically pump them.Figure 8 shows a semiconductor structure 10 having two repetitions (periods) of the first, second, third and fourth active layers 16, 20, 24, 28. More than two repetitions (periods) are also possible. Providing more than one of each active layer may increase the overall intensity of light emitted at that wavelength. Only one, only two, only three or all four of the active layers (where used) may be repeated. Different ones of the active layers may be repeated a different number of times in a given semiconductor structure 10. There may be one or more additional periods of first, second, third and fourth active layers 16, 20, 24, 28 in sequence, for example there may be between 3 and 5 periods. By providing more than one set (or period) of the four active layers 16, 20, 24, 28 the wider bandgap layers optically pump the narrower bandgap layers above them. Alternatively the fourth active layer 28 may be below the first active layer 16 in each repetition of the layers.The reflective layers 14, 40, 42, 44 may be provided once below the lowest of the respective active layers 16, 20, 24, 28 in a structure. Alternatively there may be a reflective layer provided adjacent to (beneath) each active layer.There may be a contact layer 32, 18, 22, 26 for each active layer 16, 20, 24, 28 in each period. Preferably each pair of contact layers 18, 22 that controls the first active layers 16 are coupled to the same electrodes 46, 48 so that all the first active layers 16 are activated together. Similarly for the second, third and fourth active layers 20, 24, 28. Alternatively each contact layer may have its own electrode and the voltage be delivered across the active layers in groups.Returning to Figure 3, the first contact layer 18 may comprise two sublayers forming a p++-n tunnel junction. Additionally or alternatively the second contact layer 22 may comprise two sublayers forming a p++-n tunnel junction. Additionally or alternatively the third contact layer 26 may comprise two sublayers forming a p++-n tunnel junction. Such a tunnel junction may act as a barrier to charge carriers. Each tunnel junction reverses the conductivity that is uppermost in the semiconductor structure 10 which may obviate the need for a separate insulating sublayer 34, 36, 38 since voltage will be constrained to flow only across each active layer and not through the contact layer into the next active layer.The semiconductor structure 10 shown in any of the earlier figures can be processed into a three colour pLED by etching three adjacent mesas, as shown in Figure 9 (based on the semiconductor structure 10 illustrated in Figure 1). Atop each mesa a filter 46 can be grown, bonded, deposited or otherwise provided. Thus a red filter 46a can be grown on or bonded onto the first mesa, a green filter 46b can be grown on or bonded onto the second mesa, and a blue filter 46c can be grown on or bonded onto the third mesa. Instead of the single upper contact 54 there can be a separate upper contact 54a, 54b, 54c for each mesa so that each mesa is independently addressable, as shown in Figure 10.Each filter 46 may be a distributed Bragg reflector (DBR) which reflects the non-desirable frequencies back into the semiconductor structure 10 and allows through the desirable frequency (colour). Alternatively it may be an absorber which absorbs light at other frequencies than the desirable one. The filters 46a, 46b, 46c may be grown on the preceding layer, for example epitaxially grown. Alternatively they may be deposited on the preceding layer, for example spin-coated. Such atechnique may be beneficial where the material used in one or more of the filters 46a, 46b, 46c would require a high growth temperature which would damage the semiconductor structure 10 beneath it (i.e. by exceeding the thermal budget of the semiconductor structure 10).The semiconductor structure 10 may be bonded to a second substrate at its upper surface, for example on the surface of the upper contact 30 which is distal from the active layers 16, 20, 24, 28. The growth substrate 10 can then be removed. Beneficially this can enable a semiconductor structure 10 to be grown in a sequence that is beneficial for crystal growth (for example, p above n, or longer wavelengths above shorter wavelengths) and to be assembled in an orientation that is more beneficial for a circuit (for example p down, or longer wavelengths below shorter wavelengths) by bonding it onto a new substrate. The substrate may be transparent to enable emission therethrough.One or more of the contacts is a p-contact and another one or more is an n-contact; an n-contact and a p-contact are required to excite carriers and cause emission from an active layer 16, 20, 24, 28. A p-contact may comprise two or more sublayers. For example a transparent indium-tin-oxide (ITO) sublayer and a Ni-based bonding pad. The p-contact may be transparent to light emitted from the active layers 16, 20, 24, 28 or may be reflective so that it assists in redirecting light from one or more of the active layers towards the desired emission direction.The contacts 14, 32, 18, 22, 26, 30 may have any conventional form. For example they may include metal bumps or be filled trenches. Each contact may be the same as or different to the other contacts.Alternatively the semiconductor structure 10 may include polarisation doping to generate holes in the appropriate contact layers. The dopants may require activation which may also drive out hydrogen that is present due to the growth methods.The semiconductor structure 10 may include additional layers and / or features. For example, it may include one or more device or circuit element such as transistors, diodes, resistors, capacitors, inductors, or electrostatic discharge protection components.Although the active layers have been described with the longest wavelength emission (red) nearest to the substrate 12 and the emission wavelengths decreasing at each subsequent layer other orders are also feasible. For example the first active layer may emit at the shortest wavelength (ultraviolent or blue, for example) and subsequent layers emit at progressively longer wavelengths. Advantageously the shorter wavelengths will optically pump the longer wavelength active layers as the light passes through them prior to emission through the optional transparent optical contact layer 30. Therefore including reflective layers as described in relation to Figure 5 or a broad frequency reflector adjacent to the substrate 12 is beneficial to maximise the light emission from such a semiconductor structure 10. Alternatively the active layers may emit at wavelengths which are not ordered according to position in the semiconductor structure 10.One advantage of the semiconductor structure 10 described herein is that a small pixel size is achievable, because it is epitaxially grown (or a mixture of grown and bonded) with two or more colour emission. Thus pixels are smaller than is possible with die from single colour wafers which are placed together on a backplane. Furthermore the reliability is greater because all pixels are grown in a single semiconductor structure 10 and thus there are bigger pieces to transfer.

Claims

Claims1. A semiconductor structure (10) comprising, sequentially:• a substrate (12);• an epitaxial metal layer (14);• a first active layer (16) comprising a quantum confinement structure arranged to emit at a first wavelength (Xi); and• a second active layer (20) comprising a quantum confinement structure arranged to emit at a second wavelength (Xj);• wherein the first active layer (16) comprises (lnySci.y)xGai.xN, where 0<x,y<l.

2. A semiconductor structure (10) as claimed in claim 1 wherein the epitaxial metal layer (14) is arranged to function as a mirror or an electrical contact.

3. A semiconductor structure (10) as claimed in any of claims 1 to 2 wherein the first active layer (16) comprises quantum dots.

4. A semiconductor structure (10) as claimed in claim 1 or claim 2 wherein the first wavelength (Xi) is in the range 620nm to 740nm.

5. A semiconductor structure (10) as claimed in any preceding claim wherein the second active layer (20) comprises (lnySci.y)xGai.xN , where 0<x,y<l.

6. A semiconductor structure (10) as claimed in any preceding claim wherein the second active layer (20) comprises quantum dots.

7. A semiconductor structure (10) as claimed in claim 5 wherein the second wavelength (Xj) is in the range 500nm to 570nm.

8. A semiconductor structure (10) as claimed in any preceding claim further comprising a third active layer (24) comprising a quantum confinement structure arranged to emit at a third wavelength (X3).

9. A semiconductor structure (10) as claimed in any preceding claim wherein the third active layer (24) comprises (Alylni.y)xGai.xN, where 0<x,y<l.

10. A semiconductor structure (10) as claimed in claim 8 or claim 9 wherein the third active layer (24) comprises quantum dots.

11. A semiconductor structure (10) as claimed in claim 9 wherein the third wavelength (X3) is in the range 450nm to 505nm.

12. A semiconductor structure (10) as claimed in any preceding claim further comprising a fourth active layer (28) comprising a quantum confinement structure arranged to emit at a fourth wavelength (X4).

13. A semiconductor structure (10) as claimed in claim 12 wherein the fourth active layer (28) comprises AlxGai.xN, where 0<x<l.

14. A semiconductor structure (10) as claimed in claim 12 or claim 13 wherein the fourth active layer (28) comprises quantum dots.

15. A semiconductor structure (10) as claimed in claim 12 wherein the fourth wavelength (X4) is in the range 250nm to 450nm.

16. A semiconductor structure (10) as claimed in any preceding claim further comprising a first contact layer (18) on the first active layer (16).

17. A semiconductor structure (10) as claimed in any preceding claim further comprising a second contact layer (22) on the second active layer (20).

18. A semiconductor structure (10) as claimed in claim 16 or claim 17 wherein the first contact layer (18) and / or second contact layer (22) comprises a highly doped nitride material.

19. A semiconductor structure (10) as claimed in any of claims 16 to 18 wherein the first contact layer (18) and / or second contact layer (22) comprises a p++-n junction.

20. A semiconductor structure (10) as claimed in any preceding claim further comprising a first reflective layer (39) between the epitaxial metal layer (14) and the first active layer (16).

21. A semiconductor structure (10) as claimed in any preceding claim further comprising a second reflective layer (40) between the first active layer (16) and the second active layer (20).

22. A semiconductor structure (10) as claimed in any of claims 9 to 21 further comprising a third contact layer (26) on the third active layer (24).

23. A semiconductor structure (10) as claimed in claim 22 wherein the third contact layer (26) comprises a highly doped nitride material.

24. A semiconductor structure (10) as claimed in claim 22 or claim 23 wherein the third contact layer (26) comprises a p++-n junction.

25. A semiconductor structure (10) as claimed in any of claims 9 to 22 further comprising a third reflective layer (42) between the second active layer (20) and the third active layer (24).

26. A semiconductor structure (10) as claimed in any of claims 12 to 25 further comprising a fourth reflective layer (44) adjacent to the fourth active layer (28).

27. A semiconductor structure (10) as claimed in any preceding claim further comprising a lower contact layer (32) between the epitaxial metal layer (14) and the first active layer (16).

28. A semiconductor structure (10) as claimed in any preceding claim further comprising an upper contact layer (30).

29. A semiconductor structure (10) as claimed in claim 29 wherein the upper contact layer (30) comprises InjOa.

30. A semiconductor structure (10) as claimed in any preceding claim wherein the first contact layer (18) and / or second contact layer (22) and / or third contact layer (26) comprises a first contact sublayer (18a, 22a, 26a), an insulation sublayer (34, 36, 38) and a second contact sublayer (18b, 22b, 26b).

31. A semiconductor structure (10) as claimed in claim 30 wherein the first contact sublayer (18a, 22a, 26a) and the second contact sublayer (18b, 22b, 26b) have opposite conductivity type.

32. A semiconductor structure (10) as claimed in claim 30 or claim 31 wherein the insulation sublayer (34, 36, 38) comprises a lattice matched, wide bandgap material.

33. A semiconductor structure (10) as claimed in claim 32 wherein the insulation sublayer (34, 36, 38) comprises GaN alloyed with one or more of Al, In, Sc and B.

34. A semiconductor structure (10) as claimed in claim 32 wherein the insulation sublayer (34, 36, 38) comprises wurtzite AIScN.

35. A semiconductor structure (10) as claimed in claim 30 or claim 31 wherein the insulation sublayer (34, 36, 38) comprises a dielectric material.

36. A semiconductor structure (10) as claimed in claim 35 wherein the insulation sublayer (34, 36, 38) comprises any one of an oxide, glass, a crystalline rare earth oxide, SC2O3.

37. A semiconductor structure (10) as claimed in any preceding claim further comprising an electrode (46) in or on the epitaxial metal layer (14) or the lower contact layer (32).

38. A semiconductor structure (10) as claimed in any preceding claim further comprising an electrode (48, 50, 52) in or on each of the first contact layer (18), second contact layer (22) and third contact layer (26).

39. A semiconductor structure (10) as claimed in any preceding claim further comprising an electrode (54) in or on the upper contact layer (30).

40. A multicoloured LED comprising a semiconductor structure (10) as claimed in any preceding claim.

41. A method of forming a semiconductor structure (10) comprising steps to:• grow a metal layer (14) on a substrate (12);• grow a first active layer (16) on the metal layer (14), the first active layer (16) comprising a quantum confinement structure arranged to emit at a first wavelength (Xi); and• provide a second active layer (20) on the first active layer (16), the second active layer (20) comprising a quantum confinement structure arranged to emit at a second wavelength (X?);• wherein the first active layer (16) comprises (lnySci.y)xGai.xN, where 0<x,y<l.

42. A method of forming a semiconductor structure (10) as claimed in claim 41 further comprising a step to provide a third active layer (24) on the second active layer (20), the third active layer (24) comprising a quantum confinement structure arranged to emit at a third wavelength (X3).

43. A method of forming a semiconductor structure (10) as claimed in claim 41 or claim 42 further comprising a step to provide a fourth active layer (28), the fourth active layer (28) comprising a quantum confinement structure arranged to emit at a fourth wavelength (X4).

44. A method of forming a semiconductor structure (10) as claimed in claim 43 wherein the fourth active layer (28) is grown before the first active layer (16).

45. A method offorming a semiconductor structure (10) as claimed in any of claims 41 to 44 further comprising a step to grow a first contact layer (18) on the first active layer (16).

46. A method of forming a semiconductor structure (10) as claimed in any of claims 41 to 45 further comprising a step to grow a second contact layer (22) on the second active layer (20).

47. A method of forming a semiconductor structure (10) as claimed in claim 42 further comprising a step to grow a third contact layer (26) on the third active layer (24).

48. A method of forming a semiconductor structure (10) as claimed in any of claims 41 to 47 further comprising a step to grow an upper contact layer (30) on the semiconductor structure (10).

49. A method of forming a semiconductor structure (10) as claimed in any of claims 41 to 48 wherein each step to grow a layer comprises epitaxial growth.

50. A method of forming a semiconductor structure (10) as claimed in claim 49 wherein each epitaxial growth comprises growth in a molecular beam epitaxy reactor or a metal-organic chemical vapour deposition reactor.

51. A method of forming a semiconductor structure (10) as claimed in claim 49 wherein one or more layer is epitaxially grown in an MBE reactor and one or more layer is epitaxially grown in an MOCVD reactor.

Images