Technical field

The invention relates generally to solar panels, configured to convert incident electromagnetic energy into electrical energy. In particular, the invention relates to improving existing solar panels so as to increase efficiency at low cost.

Background

Different technologies for converting solar radiation energy into other forms of useful energy have been suggested throughout the years. While various solutions for converting solar energy into thermal energy have been developed, the most challenging objective has been to convert radiation energy into electrical energy. In such a scenario, a solar panel generally refers to a photovoltaic module, including a set of photovoltaic (PV) cells, or solar cells, that generally are electrically connected.

The most prevalent material for solar panels is silicon (Si), and a typical Si PV cell is composed of a thin wafer consisting of an ultra-thin layer of phosphorus-doped (n-type) silicon on top of a thicker layer of boron-doped (p-type) silicon. An electrical field is created near the top surface of the cell where these two materials are in contact, called the p-n junction. When sunlight strikes the surface of a PV cell, this electrical field provides momentum and direction to light-stimulated charged carriers, i.e. electrons or holes, resulting in a flow of current when the solar cell is connected to an electrical load. In a single-junction PV cell, only photons whose energy is equal to or greater than the band gap of the cell material can free an electron for an electric circuit. In other words, the photovoltaic response of single-junction cells is limited to the portion of the sun's spectrum whose energy is above the band gap of the absorbing material, and lower-energy photons are not used. Furthermore, excessive energy above the band gap will be lost as heat.

Solar panels with several p-n junctions of different band gap are known. Such multi junction cells have primarily been developed based on thin film technology. As an example, such a cell may comprise multiple thin films, each essentially a solar cell grown on top of each other by metalorganic vapor phase epitaxy. A triple-junction cell, for example, may consist of the semiconductors: GaAs, Ge, and GalnP. Each layer thus has a different band gap, which allows it to absorb electromagnetic radiation over a different portion of the spectrum. Another solution is suggested in US8664513, in which solar modules including spectral concentrators are described. A solar module includes an active layer including a set of photovoltaic cells, and a spectral concentrator optically coupled to the active layer and including a luminescent material that exhibits photoluminescence in response to incident solar radiation with a peak emission wavelength in the near infrared range.

In spite of extensive research in the area, solar panel technology still faces the challenge of improving efficiency in terms of energy conversion, and the balance of energy gained compared to cost of development and installation. An aspect of this problem is the generation of heat in solar panels, which both means that a part of the incident radiation energy is not successfully converted into electrical energy, and which furthermore, might be detrimental to the function and lifetime of the solar panel.

Summary of the Invention

An embodiment of the invention provides apparatus for converting incident light to electrical energy, said apparatus comprising: a first surface configured to receive incident light, a photo voltaic layer comprising; a plurality of photovoltaic components, the photovoltaic components being interspaced with a bulk, and wherein the photovoltaic components are configured to convert light received from the bulk to electrical energy; and a reflective layer configured to reflect light propagating in the bulk.

Another embodiment of the invention provides apparatus for converting incident light to electrical energy, said apparatus comprising: a first surface configured to receive incident light, a photo voltaic layer comprising; a plurality of photovoltaic components 90, photovoltaic components being interspaced with a bulk, and wherein photovoltaic components are configured to convert light received from the bulk to electrical energy; a reflective layer configured to reflect light propagating in the bulk; a first photo luminescent layer configured to emit light at a first wavelength upon absorption of incident light and arranged between first surface and photo voltaic layer; a spectrally selective mirror configured to reflect light of a first wavelength and arranged between first surface and first photo luminescent layer.

Brief description of the drawings

Figure la shows a top view of a solar panel array according to the state of the art.

Figure lb shows a side view of a solar panel array according to the state of the art.

Figure 2 shows a side view of an embodiment described below. Figure 3 shows a side view of a photovoltaic component comprising anodes.

Figure 4 shows a side view of an embodiment described below with interconnects.

Figures 5a and 5b show a top view of an embodiment described below with interconnects. Figure 6 shows a top view of an embodiment similar to above with another arrangement.

Detailed description of the embodiments

Specific embodiments of the invention will now be described with reference to the accompanying drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. The terminology used in the detailed description of the embodiments illustrated in the accompanying drawings is not intended to be limiting of the invention. In the drawings, like numbers refer to like elements.

Fig. 1A illustrates a top planar view of a state of the art solar panel 1, comprising a plurality of solar cells 2. The solar cells 2 may be of different type, but the most common type on the market is based on a single junction silicon cell 2, having a band gap which corresponds to a certain detection wavelength XC. Each cell 2 is typically made from a Si wafer, though other types of material are also used in the art, such as GaAs. Each cell is provided with an electrode structure. On the back side of the cell 2, the electrode may have any shape, and may comprise a conductive coating (not shown) covering part of or the entire back side. On the front side of the cell 2, a connector grid is normally applied, as illustrated in the drawing of Fig. 1A. Alternatively, a transparent top electrode structure may be used, such as ITO. Further high efficiency alternatives include interdigitated back contacted cells (IBC).

In the present disclosure, the term "detection" refers to conversion of the energy of light into electricity by the photovoltaic effect. A photon that is detected is absorbed by a photovoltaic cell and converted into electrical energy and thermal energy.

In the present disclosure, the term "detection wavelength" is used to describe a wavelength within a limited wavelength range close to and below the band gap wavelength of the PV cell. A photon having the detection wavelength should have a wavelength close enough to the band gap wavelength in order to be absorbed without producing a significant amount of excessive heat due to electron thermalization, but not so close that the absorption probability is too low. If the absorption probability is too low, then the risk of the photon not being absorbed by the PV cell is too high. The term "detection wavelength" is used to simplify the explanation of the design of the converter layer. Fig. IB shows the solar panel 1 of Fig. 1A from the side, though not to scale of any realistic embodiment. This drawing shows how adjacent cells 2 may be serially connected by means of connectors 3.

A known problem related to standard solar panels is that light of shorter wavelengths than the detection wavelength XC are not efficiently converted into electrical energy. The excessive energy of an incident photon absorbed in the cell 2, exceeding the band gap, will typically be lost as heat. Not only does this result in energy loss due to lowered efficiency, but the effect of the heating may also damage the solar cells 2.

Fig. 2 illustrates an embodiment configured to alleviate this problem. In this embodiment, a solar panel 2 comprises a first surface 11 configured to receive the incident light,

a photo voltaic layer 104 comprising a plurality of photovoltaic components 90 interspaced with a bulk 6. The photovoltaic components 90 are configured to convert light received from bulk 6 to electrical energy. The solar panel 2 also has a reflective layer 123 configured to reflect light propagating in bulk 6. As will be explained, a technical effect of the solar panel 2 configuration is that it will lead to improved energy conversion efficiency. Furthermore, this benefit is obtained at a low manufacturing cost relative to existing solar panel manufacturing techniques.

The above embodiment provides several significant advantages over the PV cell configurations of the prior art. Firstly, as the PV cell is effectively made up of separate photovoltaic components 90 interspaced with bulk 6, the usable surface area of the photovoltaic components 90 is potentially 4x that of a plane of silicon where the photovoltaic components 90 are e.g. cube shaped (i.e. lx for each exposed side of the cube) or 5x that of a plane of silicon (where the top surface is also used). This effect is harnessed in the present invention. The embodiment is configured to ensure that light is coupled into the side surfaces of the PV cell as much as possible.

Secondly, a cost saving is introduced. The dominating cost in standard silicon solar panels is the cost of the silicon cell. In the present embodiment, the amount of silicon needed is significantly reduced by concentrating the light onto smaller pieces of silicon.

In the preferred embodiment, light does not enter the photovoltaic components 90 through the top surface facing the top surface 11 of the solar panel, as is normally the case, but rather the light is guided to enter the PV cell through the side faces from bulk 6. This means that it is typically desirable to have a PV cell that is thicker than is typically used for standard solar panels. In one embodiment, photovoltaic components 90 are 0.1-5 mm thick but preferably 0.2-1 mm thick with one preferred embodiment having 0.4-0.6 mm thickness. The bottom face is typically used for electrically and thermally connecting the PV cell.

One advantage of light entering through the side faces of the cell is that the top and bottom surface of the wafer can be completely covered by the anode and cathode. Then, when the wafer is singulated, the side faces are opened up to receiving light.

As the photovoltaic components 90 are wider than the thickness of a typical photovoltaic wafer, the probability of absorption is increased for photons entering through the side. This reduces the probability of long wavelength photons passing through the cell without being converted.

Whilst the surface of a normal PV cell is treated to minimize silicon defects, the photovoltaic components 90 of the present embodiment need to have all surfaces treated to remove silicon defects.

In an embodiment shown in figure 3, photovoltaic components 90 are single junction solar cells, having a band gap corresponding to a detection wavelength XC. In this embodiment, each photovoltaic component 90 comprises a first electrode 7, a second electrode 8, a semiconductor layer 21 arranged between the first electrode 7 and second electrode 8.

In an alternative embodiment, photovoltaic components 90 are multi junction solar cells, having different detection wavelengths with a band gap XCn for each of the junctions. In this embodiment, each photovoltaic component 90 comprises a first electrode 7, a second electrode 8, a semiconductor layer 21 arranged between the first electrode 7 and second electrode 8.

The bulk 6 material is an optically transmissive material that resides between the photovoltaic components 90. Preferably, a bulk material that is both flexible and able to withstand high temperatures is used. The ability to withstand high temperatures is advantageous where temperatures close the solar cell are high due to the warming of the cell due to inefficiency in the conversion process. In an embodiment, the bulk material comprises Polytetrafluoroethylene (PTFE) based materials. PTFE has a low refractive index, advantageously minimizing the reflective losses of the solar panel 2 when used by itself or in combination with an anti-reflective layer. PTFE also has the ability to withstand high temperatures (as high as 200°C) and has a low surface energy, making it difficult for dirt and dust to get stuck to the surface, and allowing easy cleaning. It is also difficult to scratch compared to other polymers and has a long lifetime (25+ years) in high temperature, high humidity environments.

In order to prevent light from leaving a photo voltaic layer 104 through its back surface, reflective layer 123 is used to direct the light to continue propagating within photo voltaic layer 104.

In a preferred embodiment, reflective layer 123 is made of titanium dioxide based paint.

In a preferred embodiment of the invention, the solar panel 2 also includes first photo luminescent layer 101. The first photo luminescent layer 101 is configured to emit light at a first wavelength upon absorption of incident light and arranged between first surface 11 and photo voltaic layer 104. The first photo luminescent layer 101 is preferably configured to emit photo luminescent light, or in other words, down-convert light incident upon it into light, of one or more wavelengths XPL, adapted for absorption by solar cells for conversion into electrical energy. The luminescent layer down shifts the majority of the incoming photons to photons that are close to the band gap of silicon thereby lowering the amount of energy that is converted to heat instead of electricity.

In one embodiment, the first photo luminescent layer 101 is configured to operate together with the photovoltaic components 90, wherein the photovoltaic components 90 are single junction solar cells having a band gap corresponding to a detection wavelength XC. In such an embodiment, the first photo luminescent layer 101 is preferably configured to emit light with a single peak of emission, i.e. light of one wavelength XPL< XC, i.e. of corresponding or larger energy than the band gap of that single junction. In a variant of this embodiment, the first photo luminescent layer 101 is configured to operate together with photovoltaic components 90, wherein photovoltaic components 90 are multi junction solar cells. In such an embodiment, the first photo luminescent layer 101 is preferably configured to emit light at different wavelengths, each with a peak of emission XPLn corresponding to a band gap XCn of the junctions of the solar cells.

In an embodiment where the first photo luminescent layer 101 is configured to operate together with single junction photovoltaic components 90 having a band gap of 1.1 eV and a maximum detection wavelength XC of llOOnm, the photo luminescent wavelength of light emitted from first photo luminescent layer 101 is configured to be within the range of 800- 1100 nm. A photo luminescent wavelength within 200nm of the detection wavelength advantageously minimizes the amount of energy lost as waste heat. Most preferably, said photo luminescent wavelength has an emission peak of 950 +/- 50 nm for a high efficiency cell, or a peak of 850 +/- 50 nm for a low efficiency cell.

In one embodiment shown in figure 2a, the solar panel 2 may include a second photo luminescent layer 111. In such an embodiment shown in figure 2B, layer 111 is provided along in between photovoltaic components 90 adjacent the reflective layer 123. In one embodiment (not shown), layer 111 may be a continuous layer along the reflective layer 123. The second photo luminescent layer 111 may be provided without first photo luminescent layer 101. Alternatively, second photo luminescent layer 111 is provided in combination with first photo luminescent layer 101. The second photo luminescent layer 111 is configured to emit light at the first wavelength upon absorption of incident light and arranged between photo voltaic layer 104 and reflective layer 123.

In one embodiment, the second photo luminescent layer 111 is configured to emit light at a first wavelength upon absorption of incident light and is arranged between photo voltaic layer 104 and the reflective layer 123.

Photo luminescent light may be emitted from photo luminescent layer 101 at different angles with respect to the incident light. The photo luminescent light will be spread out more or less isotropically. The photo luminescent light may subsequently be reflected or scattered in the photo voltaic layer 104, such that it is directed back towards the front surface 11. In one embodiment, a spectrally selective mirror 103 is arranged between the photo luminescent layer 101 and the front surface 11, configured to reflect light of the photo luminescent wavelength XPL. This way, converted light emitted from the photo luminescent layer 101 is trapped in the photo voltaic layer 104.

In this embodiment, spectrally selective mirror 103 is configured to reflect light emitted from the luminescent material 102 of the photo luminescent layer 101 and to allow light with shorter wavelengths to pass. Incident solar light passes through the spectrally selective mirror 103, typically a multi-layer optical film, and then most of the light which has higher energy than the photo luminescence will be converted to light of XPL in the NIR region by the QDs 102. Light of wavelengths shorter than XC, the detection wavelength of photovoltaic components 90 arranged within photo voltaic layer 104 may theoretically be absorbed, and the spectrally selective mirror 103 is therefore preferably transparent to such light, i.e. to light of λ< XC. At wavelengths towards the UV, each photon is highly energetic, but they are also scarce, since the sun acts as a block body radiator from which there is little emission in this part of the spectrum. The spectrally selective mirror 103 may therefore have a limited degree of transparency below the visible wavelength range. Light of wavelengths between XPL and XC will not be absorbed by the photo luminescent layer 101, but may be absorbed in by the aforementioned solar cells, and the spectrally selective mirror 103 may therefore be transmissive also to such light. However, dependent on inter alia how narrow the photo luminescence emission peak is, how close XPL is to XC, and how much the reflectivity varies dependent on angle of incidence, the spectrally selective mirror 103 may in certain embodiments be configured to be substantially reflective to light in that range. Light of wavelengths longer than XC will not be absorbed in the cells 21, and whether or not the spectrally selective mirror 103 is made reflective or transparent in this region may be determined based on other factors. For the specific peak wavelength of luminescence XPL, though, the spectrally selective mirror 103 is preferably highly reflective. This way, substantially all light emitted by fluorescence in the photo luminescent layer 101 that is directed, scattered or reflected upwards against the spectrally selective mirror 103, will be reflected back. Also, as noted above, by carefully designing the luminescent material 102 of the photo luminescent layer 101, and the spectrally selective mirror 103, to be optimized for a wavelength of 950 nm, very little light of the useful part of the solar spectrum is prevented from entering the panel. It may be noted that the cut-on/cut-off optimum of the spectrally selective mirror 103 has a complex dependence of all components in the solar panel 2 as well as on the sun spectrum, and the inclination of the solar panel 2 towards the sun.

Preferably, the spectrally selective mirror 103 is optically matched to the photo luminescent layer 101. This way, Fresnel losses are minimized. Furthermore, the spectrally selective mirror 103 preferably also adheres to the photo luminescent layer 101. Examples of multi-layer optical films (MOF), usable for realizing the spectrally selective mirror 103, may include a 3M type GBO birefringent polymer multilayer as described by Andrew Ouderkirk et. al., "Giant Birefringent Optics in Multilayer Polymer Mirrors," Science 287, pp. 2451-2456, (2000) or e.g. be configured as a rugate filter, such as the design disclosed in Fig. 5 of "Combination of angular selective photonic structure and concentrating solar cell system" by Hohn et al, presented at the 27th European PV Solar Energy Conference and Exhibition, 24-28 September 2012 in Frankfurt, Germany. Another example of such a design is shown in Fig. 4A of Ouderkirk. This drawing shows a reflectivity profile for a spectrally selective mirror 103 having a cut-on wavelength at 830 nm for 9=0-degree angle of incidence (AOI). The spectrum shifts towards lower wavelengths as the angle of incidence Θ increases. As can be gathered from the drawing, the reflectivity of the spectrally selective mirror 103 is well over 95% at XPL=950 nm, and even over 99% at least around θ=0. However, the transmittance is over 90% in the visible range, where most of the useful solar radiation to be detected in a Si solar cell is emitted. The spectrally reflective mirror 103 is preferably configured to have a so-called 'stopband' with a limited spectral range. In one example, the spectrally reflective mirror is configured to reflect light from 830 nm to 990 nm, which includes photons having photo luminescent wavelength XPL. The spectrally reflective mirror is transmissive to photons having wavelengths outside this band. A typical feature of most types of spectrally selective mirrors is that the stopband has a dependence on the angle of incidence of the light onto the spectrally selective mirror (for example, see figure 4, AOI 0 vs AOI 40 - the spectrally selective reflective mirror is designed for photo luminescent light with a peak at 910 nm). The stopband is shifted down in wavelength for an increase in angle of incidence. In order to design an omnidirectional spectrally selective mirror, a stopband that is much broader than the photo luminescent spectra is needed in order for light of all angles of incidence to be reflected (e.g. 0 to 40 degrees). An important difference between a rugate filter based mirror and a quintic layer mirror is that qiantic mirrors exhibit additional higher order stopbands (as can be seen in figure 4A). These higher order stopbands may reflect light useful for conversion. However, a rugate mirror can be designed to only have one stopband. A rugate filter based mirror is preferably used when designing a conversion layer with a very long wavelength e.g. 950 for the photo luminescent light. Therefore, in one embodiment of the invention, the spectrally reflective mirror 103 is a rugate mirror.

A significant problem in silicon solar panels of the prior art is that they are heated up by the light that is not converted to electricity as well as by the resistive losses in the panel due to low voltages and high currents. In one embodiment of the invention, the amount of heat generated from photons having energy higher than the band gap hitting the PV is further reduced. In this embodiment, the top of the PV cell is completely covered by the anode/cathode thereby stopping light that would hit the PV cell directly without being converted by the luminescent layer. Therefore, the anode/cathode forms reflective component 96. If the top of the PV cell is not covered completely, the cell would be heated almost as much as in a normal silicon solar panel and would in addition be heated by the concentrated converted light thereby raising the temperature in the cell and reducing the efficiency further.

In one preferred embodiment, the printed connector on the mounting layer onto which the PV cell is mounted is larger than the PV cell itself to form a "cap" blocking light of non- converted wavelengths hitting the PV cell directly. The larger mounting pad also lowers the tolerance required when mounting the PV cell which leads to lower cost of the solar panel. In an alternative embodiment, an additional reflective component is an additional reflective layer placed over the photovoltaic components 90 to mask the photovoltaic components 90 from unconverted photons incident on the solar panel 2.

In the above embodiments, light can only be coupled into the side surfaces of the PV cell so that the part that is not converted is prohibited from hitting the PV-cells directly by top contact layers by an additional reflective component. The additional reflective component covers and protects at least the full PV cell area and reflects the light back to the QD layer for a second chance of conversion. The NIR photons created in this process will hereafter be reflected between the selective mirror and the full spectrum mirror (also typically a multilayer optical film). The design of these mirrors is critical.

In one embodiment, photo voltaic layer 104 also comprises a scattering layer at the back of the PV-cell, typically implemented by a Ti02 coating. The efficiency of the scattering layer is very important as a photon is likely to interact with the scattering layer multiple times. Since it is easier to control the scattering of a medium when the scattering particles are not interacting with each other it may be preferable to move the scattering function to within the bulk material by infusing the bulk material with the scattering material. It is then possible to make a highly efficient specular reflector to handle the light that is not trapped in TIR. In a preferred embodiment, the scattering material 204 comprises air bubbles.

Using typical Lambertian scatterers such as barium sulphide or titanium dioxide there is always some loss of energy. One way of solving this is to infuse the bulk material with gas bubbles that act as scatterers and have a multilayer mirror to trap the light. Typically, one would use a neutral gas such as nitrogen. This also has the added advantage of lowering the effective absorption of the bulk material.

The photo luminescent layer 101 is preferably configured to emit photo luminescent light, or in other words down-convert light incident upon it into light, of one or more wavelengths XPL, adapted for absorption by solar cells for conversion into electrical energy.

In an embodiment, efficient spectral concentration, or light conversion, is realized by means of including a layer of quantum dots (QDs) in the photo luminescent layer 101, due to their stable nature as compared to dyes. QDs are well described in the art of nanophysics, and so are several known properties. One specific optical feature of QDs is the emission of photons under excitation, and the wavelength of the emitted light. One photon absorbed by a QD will yield luminescence, in terms of fluorescence. Due to the quantum confinement effect, QDs of the same material, but with different sizes, can emit light of different wavelengths. The larger the dot, the lower the energy of the emitted light. As indicated by its name, a QD is a nano-sized crystal e.g. made of semiconductor materials, small enough to display quantum mechanical properties. Typical QDs may be made from binary alloys such as cadmium selenide or cadmium sulphide (ll-VI elements), indium arsenide or indium phosphide (lll-V elements), and lead selenide (IV-VI elements), or made from ternary alloys such as cadmium selenide sulphide. It is possible to grow a shell of another material with a different band gap around the core QD region, so-called core-shell structures, e.g. with cadmium selenide in the core and zinc sulphide in the shell.

One of the two main advantages with modern QDs is the high External Quantum Efficiency (EQE) achieved. The physical mechanisms behind this high EQE may involve multi- exciton/photon generation processes wherein e.g. one absorbed photon of energy E is converted into more than one luminescent photon, e.g. two photons having half the energy of the absorbed photon (E/2) at an efficiency of e.g. 95%, see e.g. Chapters 9 & 103 of Quantum Dot Solar Cells Eds. Wu & Wang by Springer.

The QDs may be of core, core/shell or giant core/shell type, typically with a surrounding polymeric. In an embodiment, the QDs are of a core/shell structure, which are suitable for infusion in a carrier material, e.g. a PET film, and still keep its high quantum efficiency.

Alternatives to the carrier, or matrix, material may include PMMA (Poly(methyl methacrylate)), PET, epoxy resins etc. For stability reasons, the luminescent material 102 normally needs to be well encapsulated from the environment. The luminescent material may be designed with additives in order to reduce degradation during shelf-time, application, and embedding, as well as to prolong the lifetime of the finished QD-film. This can be achieved by using a protective environment such as e.g. dried nitrogen. More preferably, luminescent material 102 is encapsulated in a polymer. Another option for the photo luminescent layer 101 is to have a diffusion barrier (e.g. a dielectric layer) on each side of the layer to maintain the function of the luminescent material 102, which may be adversely affected by moisture and oxygen. The diffusion barriers can of course be put elsewhere in the stack but an advantage of putting it on the photo luminescent layer 101 itself is that the photo luminescent layer 101 can then be produced in one location and shipped to another place for assembly. Typical diffusion barriers can be dielectric coatings but many other options exist. As one example, PTFE (Polytetrafluoroethylene) of a suitable quality can act as a diffusion barrier, e.g. CYTOP ^® , which is an amorphous fluoropolymer. In one preferred embodiment, the luminescent material 102 is printed onto a thin PTFE film and then coated with another layer of PTFE so that the luminescent material 102 is sealed within a PTFE structure protecting it from the environment while maintaining high optical clarity and good mechanical properties. The general function of incorporation of QDs 102 suspended in a polymer film has been suggested by Nanosys Inc., together with 3M, though for a quite different application. They provide a QD film (QDEF - Quantum Dot Enhancement Film) which replaces a traditional diffuser film of a backlight unit. In their solution, blue LEDs are used to inject light into a backlight light guide, and part of the blue light is then shifted to emit green and red in the QDEF to provide tri-chromatic white light. As photo luminescent light typically has isotropical scattering, some of the light travelling towards the top surface at angles from the normal of the top surface less than required for total internal reflection may leave the top surface of the module resulting in optical loss. In one embodiment, the refractive index of the carrier material is between 1.5 and 2. This advantageously reduces the amount of light lost through the top surface, especially if the layers below the carrier material layer have higher refractive indexes than the carrier material.

In one embodiment, solar panel 2 is coated with a barrier coating of SiO-Si02, MgF2. By proper selection of layer thickness, this barrier coating will also serve as an anti-reflection coating as well as improving scratch resistance. In one embodiment, the optical thickness of the layer is typically ¼ wavelength of the light to be transmitted.

In the embodiments disclosed herein, the core of the QDs, configured in sizes to emit at a suitable wavelength XPL with respect to a predetermined solar cell type. Where more than one type of solar cells is employed, or if they comprise more than one junction, QDs of different sizes may be included in the luminescent material, and potentially also of different materials. Going forward, reference will mainly be made to embodiments configured for use with single junction solar cells, and hence a single peak emission wavelength XPL for the photo luminescence.

As mentioned, the luminescent material 102 of the photo luminescent layer 101 is configured to emit photo luminescent light of an energy that is greater than the band gap of a predetermined solar cell type. Preferably, the QDs of the photo luminescent layer 101 are configured to emit light at a peak wavelength XPL in the near infrared region (NIR). In one non-limiting embodiment, the solar panel 2 is configured to operate with single junction Si cells with a band gap corresponding to a wavelength XC of about 1.1 μιη. In an embodiment, the photo luminescent layer 101 is configured to emit light at an emission peak of 950 +/- 50 nm.

One benefit of having a multitude of small individual PV cells as compared to a normal silicon solar panel is that utilization of circular mother wafers may be increased since the wasted area can be reduced. For standard silicon panel PV-cells each cell covers a certain area which is typically determined by a tradeoff between maximizing the size of the PV-cells, maximizing the fill factor and minimizing silicon "loss" related to the shape of the silicon rod that has to be sliced into a PV-cell.

An anti-reflective (AR) coating 15 is an optional layer that can be placed on the front surface 11 to reduce Fresnel reflections off the front surface. The AR coating 15 can be made in one single layer or multiple layers depending on the desired reduction in front reflectance, and the range of incident angles over which the cell will operate. For an embodiment in which the upper protective layer 14 is constituted of a perfluorinated polymer with a refractive index around 1.3, a very good choice of material for the AR coating would be one with refractive index around 1.15. Such a combination would reduce front reflections significantly. Other implementations of AR coatings such as quintic or simpler versions of refractive index gradient dielectric coatings may be especially well suited for roll to roll processes e.g. by simply varying the concentration of oxygen in the machine direction of an evaporation stage. The Top AR coating 15 can also act as a diffusion barrier to protect the QD material 102 from moisture and oxidation, if the photo luminescent layer 101 itself does not include this function.

The upper protective layer 14 may or may not be optically matched to the spectrally selective mirror 103. In one embodiment, the upper protective layer 14 is unattached to the spectrally selective mirror 103. This way it may be easier to replace a damaged upper protective layer to boost output. In such an embodiment, the lower surface of the upper protective 14 may also be covered with an AR coating.

Typically, the silicon wafer will be prepared in a similar manner as for a normal silicon solar panel but will then be singulated in the same way that integrated circuits are singulated out from wafers. This is normally done either by sawing or laser cutting, but other methods exist.

The singulated PV cells can be mounted using standard pick and place equipment, taking the singulated PV cell from the foil as is customary in normal IC mounting.

Two different subassemblies will be created that are joined in two final assembly steps. These two subassemblies/layers are designated Upper layers and Bulk.

The Upper layers consist of the AR layer, the Upper protective layer, the Selective mirror and the QD layer. All of these layers can be produced separately. Alternatively, the selective mirror and the AR layer can be created with the upper protective layer as a base material. It is also possible to deposit the QD layer onto the selective mirror directly. If the layers are produced separately they are typically attached to each other in a lamination process with an optically clear adhesive as form of attachment.

The Bulk consists of the mounting layer, the bulk layer, the PV cells, the conductive grids and the scattering or reflective layer.

Upper layer production

Step 1 - AR layer added on upper protective layer

An AR layer is added on top of the upper protective layer. This can be done batch wise or roll- to-roll. An example if the upper protective layer is a PTFE film it can be beneficial to add a single layer of refractive index 1.15 or close to 1.15 to minimize the reflection losses.

Step 2 - Add selective mirror to upper protective layer

The selective mirror is added to the bottom of the upper protective layer. Either the selective mirror is pre-produced. In this case they are joined by lamination with an OCA. Or, optionally the upper protective layer is used as the base for the selective mirror and the selective mirror layers is added either in a batch process or in a roll-to-roll process.

Step 3 - Add luminescent layer to selective mirror

The luminescent layer is added to the bottom of the selective mirror. Either the QD layer is pre-produced in a film. In this case they are joined by lamination with an OCA. Or, optionally the luminescent material is coated onto the selective mirror directly and then encapsulated for protection.

BULK PRODUCTION

Step 1 - Apply electrical transfer grid onto lower bulk

A grid made of conductive material (e.g. copper, silver, carbon nano-tubes, nano-silver etc) is applied (with e.g. ink jet printing) in a suitable grid onto a carrier. The carrier is made of an optically transparent non-conductive material. The wires are printed so that they overlap with areas that are designated as mounting points for the PV cell elements. Preferably the grid is printed so that in the designated mounting areas for PV cells there is a slightly larger conductive point so that the current can flow more easily into and out of the PV cell as well as it being possible to optically distinguish the designated mounting points. These points are the electrical interface points towards the PV cell.

PRODUCTION OF DOUBLE BULK WITHOUT FILLER Three different subassemblies will be created that are joined in two final assembly steps. These three subassemblies/layers are designated: upper layers, bulk and lower bulk.

Step 1 - Apply electrical transfer grid onto lower bulk

A grid made of conductive material (e.g. copper, silver, carbon nano-tubes, nano-silver etc) is applied (with e.g. ink jet printing) in a suitable grid onto a carrier. The carrier is made of an optically transparent non-conductive material. The wires are printed so that they overlap with areas that are designated as mounting points for the PV cell elements. Preferably the grid is printed so that in the designated mounting areas for PV cells there is a slightly larger conductive point so that the current can flow more easily into and out of the PV cell as well as it being possible to optically distinguish the designated mounting points. These points are the electrical interface points towards the PV cell.

Step 2 - Apply insulating layer on top of electrical transfer grid.

A layer of non-conductive material is applied on top of the conductive grid applied in step 1. This non-conductive layer can also be applied using e.g. ink-jet printing or any other method that can apply a layer of controlled thickness in specific areas onto a substrate. This second layer is only applied in the areas that are not designated as electrical interface points towards the PV cell (as described above). This second layer allows for both optical as well as electrical interface towards the PV cell. The PV cell will be in optical contact with the carrier via the optically transparent insulating layer while still being in electrical contact with the electrical transfer grid.

Step 3 - Place PV cells onto designated mounting points

The PV cells are in the next step placed onto the designated mounting points. As the designated mounting points are optically distinguishable it is possible to use standard pick- and-place machinery to place the PV cells onto the designated mounting points. Preferably the optically transparent material applied in step 2 has adhesive properties (such as an OCA, optically clear adhesive) so that the PV cells will stay in place even though the carrier substrate may move (such as in a roll-to-roll process).

Step 4 - Join upper and lower carrier

A second carrier substrate has been created using step 1 and 2 but without mounting the PV cells as described in step 3. The two carriers (one with PV cells mounted and one without) are aligned in a roll-to-roll process so that the designated mounting points of the second carrier are aligned with the top surface of the PV cell. Thereby the PV cells are placed with one side on the designated mounting point of the first carrier and the other side on the designated mounting point of the second carrier.

Thereafter the two carriers are joined together using a suitable combination of temperature and pressure so that the two carriers and their optically transparent insulation layers come in permanent optical contact with the PV cells sandwiched in between the two carriers. The total thickness of the carrier sandwich is constant which means that material will have been slightly rearranged to allow for the non-deformable PV cells to fit within the structure.

Step 5 - QD layer joined with carrier stack

In step 5 the pre-created QD layer film (where the QDs are suitably dispersed within a film material) is joined with the carrier stack created using steps 1-4. These are joined either using a suitable combination of heat and pressure or using an OCA (optically clear adhesive). Note that the heat and pressure must be controlled to not damage the QD layer.

Step 6 - Selective mirror joined with QD/carrier stack

In step 6 the pre-created selective mirror is joined with the carrier stack created using steps 1-5. These are joined either using a suitable combination of heat and pressure or using an OCA. Note that the heat and pressure cannot be high enough so that the QD layer and/or the selective mirror are damaged in the process.

Step 7 - Apply diffusing layer onto stack

In step 7 a diffuse layer (e.g. titanium dioxide based paint) is applied onto the back of the stack (i.e. on the bottom side of the lower bulk). It is necessary that air and moisture cannot pass through this diffusing layer as it would in the end degrade the QD performance.

Step 8 (optional) - Apply protective layer onto top of stack

In an optional 8th step it is possible to apply a protective layer onto the selective mirror (it is also possible that this is already included in the selective mirror in which case this step is obsolete). This protective layer could be a coating with anti-reflective properties but the main function is to block air/moisture from passing into the stack and degrading the QD performance.

Step 9 - The sides of the module are made reflective or diffusely scattering in order to achieve a homogeneous internal light distribution within the complete module.

It is possible to envision a process flow where all of the above steps are produced in a single production line with a continuous flow of material (i.e. the carriers are continuous). INTERCONNECTS

Another advantage of using distributed PV cells is the increased flexibility to choose how to connect the cells i.e. in series or in parallel or any combination thereof. In an embodiment shown in figure 5a, photovoltaic components 90 are connected in a component group 201 connected in serial via internal interconnects 202. Group interconnects 203 allow connection to component groups. External interconnects 204 allow connection to between modules or to converters.

In one embodiment shown in figure 5b, a solar panel with the size 1.2 times 2.4 meters may be placed in a hexagonal pattern using 1 mm2 PV cells with a total fill factor of l/64th of the panel. This provides approximately 45000 PV cells on a single panel. Assuming that the cells each generate approximately 0.5V, the panel voltage may be controlled by modifying the degree of parallel to serial connected components. Normal silicon solar panels typically have 36 cells and they are all in series to get an output voltage of approximately 18-36V. A higher voltage gives lower resistive losses in the transmission grid inside the panel and allows for using grid material with lower conductivity (usually silver) thereby reducing cost while keeping electrical losses to a minimum. If the solar panel is designed to have some predetermined variation of the internal light levels, different parts of the module may have different electrical layouts. Standard PV module cells are normally large enough to experience single cell shading problems and thus requiring shunt diodes in order to protect the cells from breaking. The most severe shading situation occur when a cell in a long series of cells without a shunt diode is partly shaded. The problem of shading has been discussed by several authors and is well known within the PV community (e.g. Konrad Mertens Photovoltaics Fundamentals, Technology and Practice, Wiley 2014 Chapter 6.1) With this invention, it is possible to e.g. make "localized hexagonal component groups of serially connected cells where the component groups are then parallel connected. This may significantly reduce the problem with shading that affects standard silicon solar cells since there is no or little risk of a single cell having 25-50% shading (which is worse is depending of the cell characteristics) of a single cell, in a serially connected group to become completely shaded. In a standard silicon solar panel a shadow that partly covers one of the cells will result not only in that current contribution to disappear but also cause the rest of the illuminated cells of the module in series with the shaded cell to drive a back current that will be converted into heat impacting not only the output of the panel but potentially damaging the cell and thus the whole module permanently, unless protective diodes are used.

If the same shadow covers the same region of a solar panel with locally serially connected PV cells it may be observed that no single serially coupled cell is significantly differently shaded than its serial neighbours. This means that there will not be the same output drop to the series group or the same potential damage to the now similarly shaded cells.

It will be advantageous to make the serial groups localized as a given shadow will be less likely to induce illumination variations within the entire serially coupled cells. Other patterns are of course possible and anyone skilled in the art will be able to see the advantages of different patterns.

PV CELL PATTERNS

One preferred pattern for the serially connected PV cells is a hexagonal pattern. This gives the shortest mean path for a given converted photon to a PV cell where it can be converted to electricity and at the same time as reduced the difference in illumination between cells of the same serial chain. Another pattern, such as that shown in figure 6, may be a rectangular pattern.

LOWER PROTECTIVE LAYER/HEAT SINK

To protect the system from the environment it may be desirable to add a lower protective layer. This layer can also act as a heat sink for waste heat. Materials that are resistant to the environment as well as having high heat conductivity are all good choices. An example of such a material could be copper. It is also possible to shape the lower protective layer into a shape that allows it to dissipate heat better, such as typical cooling flanges.

As described in item 5 a vital part of the invention is that it is possible to roll-to-roll produce the solar cell. This is needed to get the cost of the solar cell as low as possible. Below is a description of one typical embodiment of the production process.

This process flow assumes that the selective mirror and the QD layer have already been produced in a roll-to-roll process but not necessarily in the same place as the final assembly of the module.

This process flow also assumes that the PV cells have been sliced into suitably large pieces before reaching the final assembly stage. It also assumes that the PV cells have been suitably doped and that the anode and cathode layers have been put in place.