Methods and apparatus for identifying the presence of shunts in monolithically integrated multi-junction solar cells

Abstract

Methods and apparatus are presented for identifying the presence of shunts in a first sub- cell of a multi-junction solar cell having monolithically integrated first and second sub-cells. The multi-junction cell is illuminated with light suitable for generating luminescence from the second sub-cell, while generating substantially no excess charge carriers in the first sub-cell. First and second images of luminescence generated from the second sub-cell with the multi-junction solar cell under different load conditions are acquired and compared to identify the presence of shunts in the first sub-cell.

Description

Methods and Apparatus for the Presence of Shunts in Monolithically Integrated Multi-Junction Solar Cells Field of the Invention

[0001] The present invention relates to methods and apparatus for identifying the presence of shunts in monolithically integrated multi-junction solar cells. However it will be appreciated that the invention is not limited to this particular field of use. Related Applications

[0002] The present application claims priority from Australian Provisional Patent Application No 2023902264 filed on 14 July 2023, the contents of which are incorporated herein by reference. Background of the Invention

[0003] Any discussion of the prior art throughout this specification should in no way be considered as an admission that such prior art is widely known or forms part of the common general knowledge in the field.

[0004] Solar power is becoming an increasingly significant part of the global power generation mix, with the vast majority of solar modules currently in use being based on crystalline silicon solar cells. While this technology is extremely well developed, alternatives are being sought because single-junction silicon solar cells have a maximum theoretical efficiency of about 29% under non-concentrated sunlight in terrestrial applications, limited in part by their single-junction nature with the excess energy of significantly above band-gap photons being lost as heat.

[0005] Higher theoretical efficiencies are offered by so-called multi-junction solar cells formed from a stack of two or more sub-cells having different band-gaps, arranged with the highest band-gap sub-cell at the top and the lowest band-gap sub-cell at the bottom. Higher energy photons, e.g. blue light, are absorbed in the top sub-cell, with lower energy photons, e.g. near infrared light, passing through the top sub-cell for absorption in a lower sub-cell. Multi-junction cells having two sub-cells are also known as tandem cells. In some multi-junction solar cell designs, generally known as four-terminal or 4T designs,the constituent sub-cells are formed with their own terminals and mechanically stacked, while in two-terminal or 2T designs the constituent sub-cells are monolithically integrated. Both designs have advantages and disadvantages, with 4T designs generally felt to be easier to manufacture and to have fewer design constraints, but lower efficiency because of the additional conductive layers. Multi-junction solar cells with so-called three-terminal or 3T designs are also known. These are similar to 2T designs in that the constituent sub-cells are monolithically integrated, but with a third terminal that may for example be on the back surface or between two sub-cells.

[0006] Many combinations of materials for multi-junction cells have been studied in recent years, with one of the most promising being the combination of perovskite on silicon, reviewed for example in S. Akhil et al ‘Review on perovskite silicon tandem solar cells: Status and prospects 2T, 3T and 4T for real world conditions’ Materials & Design 211 (2021) 110138 and Y. Cheng et al ‘Perovskite / Si tandem solar cells: Fundamentals, advances, challenges, and novel applications’ SusMat 1 (2021) 324-344. Perovskites are a class of organic-inorganic hybrid metallic halide materials, such as methyl ammonium lead iodide, with many favourable optoelectronic properties such as high absorption coefficient, long carrier diffusion length and high carrier lifetime, as well as band-gaps in the 1.6 to 1.8 eV range that are well-suited for partnering with the 1.1 eV band-gap of crystalline silicon. Compared to single-junction silicon solar cells, perovskite-silicon tandem cells have a much higher theoretical efficiency limit of about 43% under non- concentrated sunlight in terrestrial applications, which makes perovskite-silicon a very promising candidate to continue the reduction in the cost of electricity generated by solar modules.

[0007] Perovskites are also amenable to low-cost solution-based processing and can be deposited onto silicon substrates by several techniques such as spin coating, slot coating, spray coating and inkjet printing. While such deposition techniques are convenient for the manufacture of monolithically integrated (2T) multi-junction cells, they have a tendency to leave pinholes or voids, especially if the underlying surface is textured for improved light trapping as is commonly the case with silicon cells. Such deposition imperfections can lead to localised defects known as shunts in the completed perovskite sub-cell. Shuntsare short-circuit paths that reduce the efficiency of the multi-junction cell, as modelled by C. Blaga et al ‘Palliating the efficiency loss due to shunting in perovskite / silicon tandem solar cells through modifying the resistive properties of the recombination junction’ Sustainable Energy Fuels 5 (2021) 2036-2045. Shunts are also known to occur in crystalline silicon solar cells, e.g. from crystalline inclusions or process-induced breaks in the pn-junction.

[0008] In single-junction solar cells the presence or severity of shunts can generally be inferred from thermal imaging or current-voltage measurements, with techniques such as lock-in thermography and laser beam induced current (LBIC) traditionally used to determine the location of shunts, see for example US patent No 7,989,729 entitled ‘Detecting and repairing defects of photovoltaic devices’. As noted by Blaga et al, however, the current-voltage effect of shunts on multi-junction cells can differ significantly from their effect on single-junction cells. Furthermore, simple extension of LBIC to multi-junction cells may not result in the ability to determine the location of shunts, as discussed in B. Jeco et al, ‘Laser Beam Induced Current (LBIC) Mapping of InGaP / GaAs / Ge Triple Junction Solar Cells with Luminescence Coupling’, 43rdIEEE Photovoltaic Specialists Conference, Portland, Oregon, pp. 1229-1234 (2016). It would be desirable to develop alternative techniques for identifying the presence or determining the location of shunts in multi-junction solar cells.

[0009] Unless the context clearly requires otherwise, throughout the description and the claims the words ‘comprising’, ‘comprises’ and the like are to be construed in an inclusive sense as opposed to an exclusive or exhaustive sense. That is, they are to be construed in the sense of ‘including, but not limited to’. Similarly, unless the context clearly requires otherwise, the word ‘or’ is to be construed in an inclusive sense rather than an exhaustive sense. That is, the expression ‘A or B’ is to be construed as meaning ‘A, or B, or both A and B’.Summary of the Invention

[0010] In accordance with a first aspect of the present invention there is provided a method for identifying the presence of shunts in a first sub-cell of a multi-junction solar cell having monolithically integrated first and second sub-cells, the method comprising: (a) illuminating the multi-junction solar cell with light suitable for generating luminescence from the second sub-cell, while generating substantially no excess charge carriers in the first sub-cell; (b) acquiring a first image of luminescence generated from the second sub-cell, with the multi-junction solar cell under a first load condition; (c) acquiring a second image of luminescence generated from the second sub-cell, with the multi-junction solar cell under a second load condition; and (d) identifying shunts in the first sub-cell based on a comparison of intensities in the first and second images of luminescence.

[0011] Preferably, the first image of luminescence is acquired with the multi-junction solar cell under open-circuit conditions. The second image of luminescence is preferably acquired with the multi-junction solar cell under short-circuit or near short-circuit conditions.

[0012] In certain embodiments the comparison of intensities includes calculating a difference image from the first and second images of luminescence.

[0013] In certain embodiments an absorber of the first sub-cell has a larger band-gap than an absorber of the second sub-cell. Preferably, the energy of photons in the light is less than the band-gap of the absorber of the first sub-cell. In certain embodiments the absorber of the first sub-cell includes a perovskite material. In certain embodiments, the absorber of the second sub-cell includes one of either silicon or cadmium telluride or perovskite or a combination of one or more of copper, indium, gallium, selenium or sulphur.

[0014] In certain embodiments, the method further comprises:(e) determining the shunts in the first sub-cell based on spatial variations in intensity in the first image of luminescence or in the second image of luminescence.

[0015] In certain embodiments, the method further comprises: (e) determining the location of shunts in the first sub-cell based on spatial variations in intensity in the difference image.

[0016] In certain embodiments, the method for identifying the presence of shunts in a first sub-cell of a multi-junction solar cell having monolithically integrated first and second sub-cells may comprise: (a) acquiring a first image of luminescence generated from the first sub-cell while illuminating the multi-junction solar cell with light suitable for generating luminescence from the first sub-cell but not from the second sub-cell; (b) acquiring a second image of luminescence generated from the first sub-cell while illuminating the multi-junction solar cell with light suitable for generating luminescence from the first sub-cell and from the second sub-cell; and (c) identifying the presence of shunts in the first sub-cell based on a comparison of intensities in the first and second images of luminescence.

[0017] In accordance with a second aspect of the present invention there is provided an apparatus for identifying the presence of shunts in a first sub-cell of a multi-junction solar cell having monolithically integrated first and second sub-cells, the apparatus comprising: a light source for illuminating the multi-junction solar cell with light suitable for generating luminescence from the second sub-cell, while generating substantially no excess charge carriers in the first sub-cell; a system for applying different load conditions to the multi-junction solar cell; and an image capture device for acquiring first and second images of luminescence generated from the second sub-cell, with the multi-junction solar cell under different load conditions.

[0018] Preferably, the apparatus further a computer for identifying the presence of shunts in the first sub-cell based on a comparison of intensities in the first and second images of luminescence. In certain embodiments the computer is configured to compare intensities in the first and second images of luminescence by calculating a difference image from the first and second images of luminescence. In certain embodiments, the computer is configured to determine the location of shunts in the first sub-cell based on spatial variations in intensity in the first image of luminescence or in the second image of luminescence. In certain embodiments, the computer is configured to determine the location of shunts in the first sub-cell based on spatial variations in intensity in the difference image.

[0019] Preferably, the apparatus is configured to acquire the first image of luminescence with the multi-junction solar cell under open-circuit conditions. The apparatus is preferably configured to acquire the second image of luminescence with the multi- junction solar cell under short-circuit or near short-circuit conditions.

[0020] In certain embodiments the apparatus is configured to identify the presence of shunts in the first sub-cell of a multi-junction solar cell in which an absorber of the first sub-cell has a larger band-gap than an absorber of the second sub-cell. The light source is preferably configured to illuminate the multi-junction solar cell with light having photon energy less than the band-gap of the absorber of the first sub-cell. In certain embodiments the apparatus is configured to identify the presence of shunts in the first sub-cell of a multi-junction solar cell in which the absorber of the first sub-cell includes a perovskite material. The absorber of the second sub-cell may include one of either silicon or cadmium telluride or perovskite or a combination of one or more of copper, indium, gallium, selenium or sulphur.

[0021] In accordance with a third aspect of the present invention there is provided a method for estimating local shunt resistance in a first sub-cell of a multi-junction solar cell having monolithically integrated first and second sub-cells, the method comprising:(a) illuminating a selected the second sub-cell with light suitable for generating excess charge carriers in the second sub-cell, while generating substantially no excess charge carriers in the first sub-cell; (b) applying a voltage to the multi-junction solar cell while the selected region of the second sub-cell is being illuminated; (c) measuring the current induced in the multi-junction solar cell by the illumination, for at least two different values of the applied voltage that are less than the open-circuit voltage of the illuminated region of the second sub-cell; and (d) determining, from the dependence of the measured current on the applied voltage, an estimate of the local shunt resistance in an area of the first sub-cell laterally proximate to the illuminated region of the second sub-cell.

[0022] In certain embodiments the selected region of the second sub-cell is laterally proximate to the position of a suspected shunt in the first sub-cell. In certain embodiments an absorber of the first sub-cell has a larger band-gap than an absorber of the second sub- cell. Preferably, the energy of photons in the light is less than the band-gap of an absorber of the first sub-cell. In certain embodiments the absorber of the first sub-cell includes a perovskite material. The absorber of the second sub-cell may include one of either silicon or cadmium telluride or perovskite or a combination of one or more of copper, indium, gallium, selenium or sulphur.

[0023] In accordance with a fourth aspect of the present invention there is provided an apparatus for estimating local shunt resistance in a first sub-cell of a multi-junction solar cell having monolithically integrated first and second sub-cells, the apparatus comprising: an illumination system for illuminating a selected region of the second sub-cell with light suitable for generating excess charge carriers in the second sub-cell, while generating substantially no excess charge carriers in the first sub-cell; a variable voltage source for applying a voltage to the multi-junction solar cell while the selected region of the second sub-cell is being illuminated; a current measurement device for measuring the current induced in the multi- junction solar cell by the illumination, for at least two different values of the appliedvoltage that are less than the open- of the illuminated region of the second sub-cell; and a computer for determining, from the dependence of the measured current on the applied voltage, an estimate of the local shunt resistance in an area of the first sub-cell laterally proximate to the illuminated region of the second sub-cell.

[0024] In certain embodiments the apparatus is configured to illuminate a selected region of the second sub-cell that is laterally proximate to the position of a suspected shunt in the first sub-cell. In certain embodiments the apparatus is configured to estimate local shunt resistance in the first sub-cell of a multi-junction solar cell in which an absorber of the first sub-cell has a larger band-gap than an absorber of the second sub-cell. The illumination system is preferably configured to illuminate the selected region of the second sub-cell with light having photon energy less than the band-gap of an absorber of the first sub-cell. In certain embodiments the apparatus is configured to estimate local shunt resistance in the first sub-cell of a multi-junction solar cell in which the absorber of the first sub-cell includes a perovskite material. The absorber of the second sub-cell may include one of either silicon or cadmium telluride or perovskite or a combination of one or more of copper, indium, gallium, selenium or sulphur.

[0025] In accordance with a fifth aspect of the present invention there is provided a method for identifying the presence of shunts in a first sub-cell of a multi-junction solar cell having monolithically integrated first and second sub-cells, the method comprising: (a) sequentially illuminating a plurality of selected regions of the second sub-cell with light suitable for generating excess charge carriers in the second sub-cell, while generating substantially no excess charge carriers in the first sub-cell; (b) applying to the multi-junction solar cell, while the selected regions of the second sub-cell are being illuminated, a voltage that is less than the open-circuit voltage of the illuminated regions of the second sub-cell; (c) measuring, at each of the selected regions of the second sub-cell, the current induced in the multi-junction solar cell by the illumination; and (d) identifying shunts in the first sub-cell based on variations in the measured current.

[0026] Preferably, the method further estimating the severity of shunts in the first sub-cell based on the magnitude of the variations in the measured current.

[0027] In certain embodiments an absorber of the first sub-cell has a larger band-gap than an absorber of the second sub-cell. Preferably, the energy of photons in the light is less than the band-gap of the absorber of the first sub-cell. In certain embodiments the absorber of the first sub-cell includes a perovskite material. The absorber of the second sub-cell may include one of either silicon or cadmium telluride or perovskite or a combination of one or more of copper, indium, gallium, selenium or sulphur.

[0028] In accordance with a sixth aspect of the present invention there is provided an apparatus for identifying the presence of shunts in a first sub-cell of a multi-junction solar cell having monolithically integrated first and second sub-cells, the apparatus comprising: an illumination system for sequentially illuminating a plurality of selected regions of the second sub-cell with light suitable for generating excess charge carriers in the second sub-cell, while generating substantially no excess charge carriers in the first sub- cell; a voltage source for applying to the multi-junction solar cell, while the selected regions of the second sub-cell are being illuminated, a voltage that is less than the open- circuit voltage of the illuminated regions of the second sub-cell; a current measurement device for measuring, at each of the selected regions of the second sub-cell, the current induced in the multi-junction solar cell by the illumination; and a computer for identifying shunts in the first sub-cell based on variations in the measured current.

[0029] Preferably, the computer is configured to estimate the severity of shunts in the first sub-cell based on the magnitude of the variations in the measured current.

[0030] In certain embodiments the apparatus is configured to identify the presence of shunts in the first sub-cell of a multi-junction solar cell in which an absorber of the first sub-cell has a larger band-gap than an absorber of the second sub-cell. The illumination system is preferably configured to illuminate the selected regions of the second sub-cellwith light having photon energy less than band-gap of the absorber of the first sub-cell. In certain embodiments the apparatus is configured to identify shunts in the first sub-cell of a multi-junction solar cell in which the absorber of the first sub-cell includes a perovskite material. The absorber of the second sub-cell may include one of either silicon or cadmium telluride or perovskite or a combination of one or more of copper, indium, gallium, selenium or sulphur.

[0031] In accordance with a seventh aspect of the present invention there is provided a non-transitory computer readable medium with an executable program code configured to implement the method according to any of the first, third or fifth aspects, or to operate the apparatus according to any of the second, fourth or sixth aspects. Brief Description of the Drawings

[0032] Benefits and advantages of the present invention will become apparent to those skilled in the art to which this invention relates from the subsequent description of exemplary embodiments and the appended claims, taken in conjunction with the accompanying drawings, in which: Figure 1 illustrates in schematic form a simplified cross-sectional view of a monolithically integrated multi-junction solar cell having upper and lower sub- cells; Figure 2 illustrates in schematic form a simplified cross-sectional view of a monolithically integrated multi-junction solar cell having an upper sub-cell comprising a perovskite material and a lower sub-cell comprising crystalline silicon, with various defects in the upper sub-cell; Figure 3 depicts in schematic form an apparatus for identifying the presence of shunts in the upper sub-cell of the multi-junction solar cell of Figure 2, according to a first aspect of the present invention; Figures 4A and 4B show in schematic form the multi-junction solar cell of Figure 2 illuminated with light that is above band-gap for the lower sub-cell but below band-gap for the upper sub-cell, while under open-circuit conditions and short- circuit conditions respectively;Figures 5A and 5B show of luminescence generated from the lower sub-cell of the multi-junction solar cell of Figure 2 under open-circuit conditions and short-circuit conditions respectively, assuming no lateral transport of photo-generated charge carriers; Figure 5C shows a difference image obtained by subtracting the image of Figure 5B from the image of Figure 5A; Figure 6A shows a simulated image of luminescence generated from the upper sub-cell of the multi-junction cell of Figure 2 at open-circuit, assuming no lateral transport of photo-generated charge carriers; Figure 6B shows a ‘corrected’ version of the image of Figure 6A, in which the effects of shunts identified from Figure 5C have been removed; Figure 7 depicts, in schematic form, an apparatus for identifying the presence of shunts in the lower sub-cell of a multi-junction solar cell, according to the first aspect of the present invention; Figure 8 depicts, in schematic form, an apparatus for estimating local shunt resistance in the upper sub-cell of the multi-junction solar cell of Figure 2, according to a second aspect of the present invention; Figure 9 shows four selected regions of a multi-junction solar cell at which the local shunt resistance is to be estimated; and Figure 10 shows simulated plots of measured current versus applied voltage at the selected regions shown in Figure 9, obtained using the apparatus of Figure 8. Detailed Description

[0033] Preferred embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings. Although the embodiments of the invention will be described with reference to perovskite substrates, it will be appreciated that other types of substrates are possible. Perovskite / perovskite, perovskite / CdTe and perovskite / CI(G)S are all upcoming tandem technologies which may be applicable for the present invention. Some embodiments of the present invention may be suitable for all types of tandems as long as the right / suitable illumination and bias conditions are used.

[0034] Figure 1 shows, in schematic a simplified cross-sectional view of a monolithically integrated multi-junction solar cell 100 having an upper sub-cell 102 and a lower sub-cell 104 separated by a recombination junction 106 that connects the two sub- cells in series, as well as upper and lower laterally conductive layers 108, 110 and metal electrodes 112, 114. The upper sub-cell 102 comprises an absorber or base 116 and an emitter layer 118 and likewise the lower sub-cell 104 comprises an absorber or base 120 and an emitter layer 122. The terms ‘upper’ and ‘lower’ are defined with respect to the intended direction of solar irradiation 124, with the absorber 116 of the upper sub-cell 102 having a larger band-gap than the absorber 120 of the lower sub-cell 104. Higher energy photons 126 in the solar irradiation 124, e.g. blue light, are absorbed in the upper sub- cell 102, while lower energy photons 128 that are below band-gap for the upper sub-cell, e.g. near-infrared light, pass through the upper sub-cell for absorption in the lower sub- cell 104. Some multi-junction solar cell designs have one or more additional sub-cells below the lower sub-cell 104, for absorption of photons of successively lower energy. Excess charge carriers generated by absorption of photons in the absorbers 116, 120 migrate to the upper or lower laterally conductive layers 108, 110 for extraction via the metal electrodes 112, 114. The recombination junction 106 preferably has high transparency for the lower energy photons 128, as well as low lateral conductivity to reduce the impact of shunts in either of the sub-cells 102, 104.

[0035] As mentioned previously, the combination of perovskite on silicon is a class of multi-junction solar cell that has received considerable attention in recent years. The principles of the present invention will be described in detail for this type of multi- junction solar cell, but the principles are generally applicable to monolithically integrated multi-junction solar cells having two or more sub-cells. Accordingly, Figure 2 shows, in schematic form, a simplified cross-sectional view of a monolithically integrated multi- junction solar cell 200 having an upper sub-cell 202 comprising a perovskite absorber 216 deposited on a lower sub-cell 204 comprising a crystalline n-type silicon absorber 220 and a p-type crystalline silicon emitter layer 222, with the recombination junction 206 comprising a highly doped n++ / p++tunnel junction for example. The compositions of the absorber 216 and emitter layer 218 of the perovskite sub-cell 202 may be selected from alarge range of compositions that have over the years, as discussed in the previously mentioned reviews by Cheng et al and Akhil et al for example. Those skilled in the art will understand that actual perovskite / silicon multi-junction cell structures may have several additional layers besides those shown in the simplified cross-sectional view of Figure 2.

[0036] A first aspect of the present invention concerns a method for identifying the presence of shunts in a first sub-cell 102 of a multi-junction solar cell 100 having monolithically integrated first and second sub-cells 102, 104. Throughout this specification, the terms “identifying the presence of shuts” is intended to mean identifying that one or more shunts exist in a sub-cell of a multi-junction solar cell. In some embodiments, identifying the presence of shunts may also comprise identifying an approximate or precise location of one or more shunts in a sub-cell of a multi-junction solar cell. The method makes use of the phenomenon that when excess charge carriers are substantially generated only in the second sub-cell 104 and the multi-junction cell 100 is under short-circuit, near short-circuit or other current extraction conditions, current leakage through shunts in the first sub-cell 102 will reduce the voltage in the second sub- cell 104. Since the intensity of band-to-band luminescence caused by radiative recombination of excess charge carriers depends exponentially on the local voltage, the local voltage reduction caused by shunts in the first sub-cell 102 will manifest as a reduction in luminescence emitted from the second sub-cell 104. If there is negligible lateral transport of photo-generated charge carriers, e.g. through the emitter layer 122 of the second sub-cell 104 or through the recombination junction 106, the voltage reductions will be localised to areas of the second sub-cell above or below, i.e. laterally proximate to, shunts in the first sub-cell. In this case, the location of shunts in the first sub-cell 102 may be determined based on spatial variations in luminescence emitted from the second sub- cell 104. The shunt-related luminescence intensity variations will become less localised as lateral transport of photo-generated charge carriers becomes more significant, but it will still be possible to identify the presence of shunts in the first sub-cell 102 based on spatial variations of luminescence when the multi-junction cell 100 is under short-circuit, near short-circuit or other current extraction conditions. Furthermore, it will be appreciated thatthe effect of lateral carrier transport corrected by applying known noise filtering methods, for example.

[0037] It will be appreciated that embodiments of the invention are applicable when the multi-junction cell is not under complete short-circuit conditions. A short-circuit condition is defined as a bias voltage of exactly 0 V. A near short-circuit condition is defined by a bias voltage of anywhere between -0.5 V and 0.5 V. A disadvantage of imaging at short circuit is that when effects such as S-shaped IV curves (which occur due to energy barriers between functional layers in the sub-cell) occur close to short-circuit conditions (0V), the image no longer represents the impact of shunts only, but rather also of the S-shaped IV curve. This can cause false identification of shunts in the device. By measuring near short- circuit (rather than at short circuit), the voltage range where such S-shaped effects occur can be avoided and the image then truly reflects the areas of shunting rather than areas where S-shaped effects occur.

[0038] Voltage reductions associated with shunting in the first sub-cell 102 are not expected to occur when the multi-junction cell 100 is under open-circuit conditions, and there may even be a small voltage increase in areas above or below a shunt in the first sub- cell 102, especially if the illumination generates a small number of excess charge carriers in the first sub-cell.

[0039] This luminescence-based method will firstly be described for the case where the first sub-cell, i.e. the sub-cell in which we are looking for shunts, is above the second sub- cell and therefore has the larger band-gap. Accordingly, the perovskite sub-cell 202 shown in Figure 2 includes a number of shunts 230 that may for example be caused by pinholes or voids in the absorber 216. The perovskite sub-cell 202 may also have other types of defects such as regions of low carrier lifetime material 232 caused for example by impurities or grain boundaries, as well as series resistance defects 234 such as areas of high contact resistance between the emitter layer 218 and the upper laterally conductive layer 208.

[0040] Figure 3 depicts in schematic form an apparatus 301 for identifying the presence of shunts 230 in the upper sub-cell 202 of a monolithically integrated multi-junction solarcell 200. For the example of a multi- cell 200 comprising a perovskite upper sub- cell 202 and a silicon lower sub-cell 204, the apparatus 301 comprises: a light source 303 for broad area illumination of the multi-junction cell 200 with light 305 suitable for generating luminescence 307 from the silicon sub-cell 204; a system 309 for applying different load conditions to the multi-junction cell 200; and an image capture device 311 for acquiring two or more images of luminescence 307 generated from the silicon sub- cell 204 under different load conditions. In preferred embodiments the apparatus 301 also comprises a computer 313 equipped with suitable machine readable code for identifying the presence of shunts 230 in the perovskite sub-cell 202 based on a comparison of intensities in the two or more images of luminescence 307, e.g. by calculating a difference image. Alternatively, a user may identify the presence of shunts in the perovskite sub-cell 202 by visual comparison of intensities in two or more images of luminescence 307 generated from the silicon sub-cell 204 under different load conditions. If present, the computer 313 may also be equipped with suitable machine-readable code for controlling the light source 303 or the system 309 for applying different load conditions to the multi- junction cell 200.

[0041] Several other components may be present in the apparatus 301, such as optical power elements for shaping the illumination 305 or collecting the luminescence 307, a short-pass filter between the light source 303 and the cell 200 for preventing long wavelength tail radiation in the light 305 from reaching the image capture device 311, or a long-pass or band-pass filter between the cell 200 and the image capture device for preventing reflected illumination from reaching the image capture device. In this example embodiment the image capture device 311 should be sensitive to the band-to-band luminescence emission from crystalline silicon and may for example be a silicon or InGaAs CCD camera.

[0042] The illumination conditions should be chosen such that substantially no excess charge carriers are generated in the perovskite sub-cell 202. Since it will generally be convenient to illuminate the multi-junction cell 200 from the upper, sun-facing side as shown in Figure 3, the required illumination conditions can be provided with a light source 303 emitting light 305 that is above band-gap for the silicon sub-cell 204 but belowband-gap for the perovskite absorber is, the photons in the light 305 preferably have energy greater than the band-gap of the silicon sub-cell 204 but less than the band- gap of the perovskite absorber 216. If the perovskite and silicon materials have band-gaps of 1.8 eV (690 nm) and 1.1 eV (1130 nm) respectively, a light source 303 emitting light 305 somewhere in the range 800 to 1000 nm, such as an 808 nm or 915 nm laser diode, would be a suitable choice. The multi-junction cell 200 could alternatively be illuminated from the lower side if the lower laterally conductive layer 210 is transparent to the illumination. While it would again be preferable for the illumination to be above band- gap only for the silicon sub-cell 204, in this case the illumination could also be above band-gap for the perovskite absorber 216 provided the absorption of the illumination in the silicon sub-cell is sufficiently high such that substantially no excess charge carriers are generated in the perovskite absorber. In certain embodiments the condition of ‘substantially no excess charge carriers being generated’ in a first sub-cell 202 may be satisfied if the total excess carrier generation rate across the thickness of the first sub-cell is less than 5%, more preferably less than 0.1%, of that across the thickness of the second sub-cell 204.

[0043] As shown in Figure 3 the luminescence 307 generated from the silicon sub-cell 204 is collected with an image capture device 311 positioned on the upper, sun-facing side of the multi-junction cell 200. Alternatively, the image capture device 311 may be positioned to collect luminescence emitted from the lower side of the cell 200 if the lower laterally conductive layer 210 is transparent to the luminescence.

[0044] In a preferred embodiment first and second images of luminescence are acquired with the multi-junction cell 200 under open-circuit and short-circuit or near short-circuit conditions respectively, since this pair of load conditions will generally provide a large luminescence intensity differential around any shunts 230. In this embodiment the system 309 for applying different load conditions may be as simple as a breakable connection 315 between the metal electrodes 212, 214. In general, however, any pair of different load conditions, e.g. applied voltages equal to 25% open-circuit voltage and 75% open-circuit voltage, applied for example with a system 309 in the form of a variable power supply, will produce an intensity differential between a pair of luminescence images. It will beappreciated that the load condition of junction cell 200 could also be altered by varying the current injection level, for example. In preferred embodiments the comparison of intensities comprises calculating a difference image, e.g. by subtracting the second image from the first image, although the two images may be compared by other means such as by calculating intensity ratios, or visually.

[0045] Figure 4A shows in schematic form a monolithically integrated perovskite / silicon multi-junction cell 200 illuminated, while under open-circuit conditions, with light 305 that is above band-gap for the silicon sub-cell 204 but below band-gap for the perovskite absorber 216. Radiative recombination of excess charge carriers generated in the silicon sub-cell 204 by the light 305 produces band-to-band luminescence 307 that can be captured with an image capture device (not shown). Figure 5A shows a simulated image 500 of luminescence collected from the multi-junction cell 200 under open-circuit conditions and under the assumption that there is no lateral transport of photo-generated charge carriers. In this particular simulation the silicon sub-cell 204 was under 0.3 Suns illumination of above band-gap light from a monochromatic 808 nm light source 303 and the perovskite sub-cell 202 was under 0.001 Suns illumination of above band-gap light from stray ambient light, so that the simulated luminescence shown in Figure 5A is substantially generated from the silicon sub-cell 204. Any luminescence that happens to be generated from the perovskite sub-cell 202 will generally be in a different wavelength range and could be blocked with suitable filtering. Several spatially resolved intensity variations can be seen in Figure 5A, including some relatively high brightness features indicated by arrows 502.

[0046] Figure 4B shows in schematic form the same perovskite / silicon multi-junction cell 200 illuminated 305 in the same manner as in Figure 4A, but this time with the metal electrodes 212, 214 connected 400 to provide short-circuit conditions. It will be appreciated that embodiments of the invention may work when the electrodes are connected to provide near short-circuit conditions. Provided there is little lateral transport of photo-generated charge carriers, and noting that the illumination 305 generates substantially no excess charge carriers in the perovskite absorber 216, the silicon sub-cell 204 will be under open circuit conditions everywhere except in regions 436 underneath orproximate to shunts 230 in the perovskite cell 202. Shorting of the perovskite sub-cell 202 associated with the shunts 230 causes leakage of current from the underlying regions 436 of the silicon sub-cell 204, as represented by the current path 404 through the upper emitter layer 218 or upper laterally conductive layer 208, the external short-circuit 400 and the lower laterally conductive layer 210. This current leakage reduces the local voltage at, and therefore the amount of luminescence 307 emitted from, the localised areas 436. A simulated image 504 of luminescence collected from the multi-junction cell 200 under these short-circuit conditions is shown in Figure 5B. Several spatially resolved intensity variations can be seen, including some relatively low brightness features indicated by arrows 506.

[0047] Subtracting the image of Figure 5B from the image of Figure 5A yields the difference image 508 shown in Figure 5C, showing three bright regions indicated by arrows 510 in positions corresponding to the intensity variations indicated by the arrows 502, 506 in Figures 5A and 5B and that can be identified with the approximate location of shunts 230 in the perovskite sub-cell 202. Intensity features that appear the same in the open-circuit image 500 and the short-circuit image 504, such as the relatively dark area across the lower left corner of Figures 5A and 5B, are presumably due to other features in the silicon sub-cell 204 known to affect its luminescence emission, such as areas of low carrier lifetime material.

[0048] The localisation of shunt-related intensity variations in the difference image of Figure 5C, or in luminescence images acquired under different load conditions, will be reduced to a greater or lesser extent depending on the extent of lateral transport of photo- generated charge carriers, but the presence of shunts in the perovskite sub-cell 202 can still be identified based on a comparison of intensities in the luminescence images acquired under different load conditions. For example the presence of shunts in the perovskite sub-cell 202 will cause an overall decrease in intensity of a luminescence image acquired under short-circuit conditions compared to a luminescence image acquired under open-circuit or near short-circuit conditions.

[0049] Returning to Figure 4B, we note other types of defects that may be present in the perovskite sub-cell 202, such as regions of low carrier lifetime material 232 and series resistance defects 234, are not expected to cause current leakage from the underlying silicon sub-cell 204. Consequently, the luminescence-based method is expected to be highly selective for identifying the presence of shunts 230 in the perovskite sub-cell 202. Optionally, and with reference to Figure 3, the presence of other types of defects 232, 234 in the perovskite sub-cell 202 can be investigated by illuminating the multi-junction cell 200 with light 305 that is above band-gap for the perovskite absorber 216. This time band-to-band luminescence 307 is generated from the perovskite sub-cell 202, for capture by the image capture device 311 and optional analysis in a computer 313. Provided the incident illumination 305 is sufficiently absorbed by the perovskite sub-cell 202, the underlying silicon sub-cell 204 will make substantially no contribution to the measured luminescence signal, and in any event luminescence from the silicon sub-cell could be blocked with a suitable filter in front of the image capture device 311.

[0050] Figure 6A shows a simulated image 600 of luminescence generated from the multi-junction cell 200 of Figure 2 at open-circuit and with the perovskite sub-cell 202 and silicon sub-cell 204 respectively illuminated with 0.3 Suns and 0.001 Suns of light that is above band-gap for each sub-cell. The image 600 includes three low intensity regions 602 that are known from the previously obtained images 5A, 5B and 5C to be caused by shunts in the perovskite sub-cell 202. These intensity-depression regions can be removed by image processing techniques, e.g. by ‘adding back’ an appropriately scaled version of Figure 5C, to obtain a ‘corrected’ open-circuit luminescence image 604 shown in Figure 6B, in which the remaining intensity variations may be due to carrier lifetime- related defects 232 in the perovskite sub-cell 202.

[0051] We now consider the case where the first sub-cell, i.e. the sub-cell in which we are looking for shunts, is below the second sub-cell and therefore has the smaller band-gap. For consistency we will again use the example of a monolithically integrated perovskite / silicon multi-junction cell 200, so that in this case the first sub-cell is the silicon sub-cell 204. Although shunts are generally less likely to occur in the silicon sub- cell than in the perovskite sub-cell, in other types of multi-junction cells, such as cellshaving two (or more) perovskite sub- compositions, shunts may be highly likely to occur in the lower, smaller band-gap sub-cell.

[0052] Figure 7 depicts in schematic form an apparatus 701 for identifying the presence of shunts 730 in the lower, silicon sub-cell 204 of a monolithically integrated perovskite / silicon multi-junction cell 200. Similar to the apparatus 301 depicted in Figure 3, the apparatus 701 comprises a light source 703, a system 709 for applying different load conditions to the multi-junction cell 200 and an image capture device 711, and optionally a computer 713. In this case the light source 703 is selected to provide broad area illumination with light 705 suitable for generating luminescence 707 from the perovskite sub-cell 202 and the image capture device 711 is selected to be sensitive to band-to-band luminescence 707 emitted from the perovskite sub-cell 202. For example if the perovskite absorber 216 has a band-gap of 1.8 eV (690 nm) the light source 703 may be a 540 nm or 490 nm laser diode and the image capture device 711 may be a silicon CCD camera. Provided the absorption of the illumination 705 in the perovskite absorber 216 is sufficiently high, substantially no excess charge carriers will be generated in the silicon sub-cell 204 by the illumination 705.

[0053] For the shunt-detecting method the image capture device 711 acquires two or more images of luminescence 707 generated from the perovskite sub-cell 202 under different load conditions imposed by the system 709, for example open-circuit and short-circuit or near short-circuit conditions. In preferred embodiments a computer 713 equipped with suitable machine-readable code identifies the presence of shunts 730 in the silicon sub-cell 204 based on a comparison of intensities in the two or more images of luminescence 707, e.g. by calculating a difference image. Alternatively, a user may identify the presence of shunts 730 by visual comparison of two or more images of luminescence 707 generated from the perovskite sub-cell 202 under different load conditions. If present, the computer 713 may also be equipped with suitable machine-readable code for controlling the light source 703 or the system 709 for applying different load conditions to the multi-junction cell 200. Similar to the previously described embodiment, the presence of shunts 730 in the silicon sub-cell 204 can be identified by their tendency to leak photo-generated current from laterally proximate regions 736 in the perovskite sub-cell 202. Other types ofdefects in the silicon sub-cell 204, such lifetime defects 732 and series resistance defects 734, will not cause current leakage but could be investigated, if required, in one or more separate measurements in which luminescence is generated from the silicon sub-cell.

[0054] Although the method and apparatus for identifying the presence of shunts in a first sub-cell of a multi-junction solar cell have been described for the case where the multi- junction solar cell has only two monolithically integrated sub-cells, they are not so limited. For example if the first sub-cell, i.e. the sub-cell being investigated for shunts, is the uppermost of three or more monolithically integrated sub-cells, the multi-junction cell could be illuminated with light that is above band-gap for all sub-cells except the first sub- cell, thereby generating excess charge carriers in all sub-cells except for the first sub-cell. The presence of shunts in the first sub-cell could then be identified based on a comparison of intensities in two or more images of luminescence generated from one of the underlying sub-cells, preferably from the immediately underlying sub-cell, with the multi-junction cell under different load conditions. Alternatively, the presence of shunts in the lowermost of three or more monolithically integrated sub-cells could be identified if the multi-junction cell is illuminated with light that contains no wavelengths that pass through to the lowermost sub-cell, thereby generating excess charge carriers in all of the overlying sub-cells but not in the lowermost sub-cell. In yet another example, the presence of shunts in the middle sub-cell of a multi-junction solar cell having three monolithically integrated sub-cells could be identified if the multi-junction cell is illuminated with two wavelength bands selected to generate excess charge carriers in the uppermost and lowermost sub- cells but not in the middle sub-cell.

[0055] A second aspect of the present invention concerns a method for estimating local shunt resistance in a first sub-cell of a multi-junction solar cell having monolithically integrated first and second sub-cells. This method will firstly be described for the case where the first sub-cell is above the second sub-cell and therefore has the larger band-gap. In an example embodiment the first and second sub-cells are composed of a perovskite material and silicon respectively.

[0056] Figure 8 depicts in schematic apparatus 801 for estimating local shunt resistance in the upper, perovskite sub-cell 202 of a monolithically integrated perovskite / silicon multi-junction cell 200. The apparatus 801 comprises: an illumination system 815 for illuminating a selected region 817-1 of the silicon sub-cell 204 with light 805 suitable for generating excess charge carriers in the silicon sub-cell, while generating substantially no excess charge carriers in the perovskite sub-cell 202; a variable voltage source 819 for applying a controllable voltage to the multi-junction cell 200 while the selected region 817-1 is being illuminated; a current measurement device 821 such as an ammeter for measuring current induced in the multi-junction cell 200 by the illumination, for at least two different values of the applied voltage; and a computer 813 for determining, from the dependence of the measured current on the applied voltage, an estimate of the local shunt resistance in an area 823 of the perovskite sub-cell 202 laterally proximate to the illuminated region 817-1 of the silicon sub-cell 204. The computer 813 will in general be equipped with suitable machine readable code for moving the illumination 805 relative to the multi-junction cell 200 so as to illuminate different selected regions 817-1, 817-2, 8173 of the silicon sub-cell 204, controlling the variable voltage source 819 and interrogating the current measurement device 821, as well as for determining estimates of local shunt resistance in the perovskite sub-cell 202. In the illustrated embodiment the illumination system 815 includes a light source 823 such as a laser emitting a collimated beam 825 and a beam steering mirror 827 rotatable in two axes to scan the illumination 805 across the multi-junction cell 200 in two dimensions, although many other suitable illumination systems such as a digital projector or a light source combined with a MEMS device or a spatial light modulator could be employed. Alternatively, the on-sample position of the illumination 805 may be scanned by movement of the multi-junction cell 200 in two dimensions. The illumination system 815 may also include one or more optical power elements for controlling the on-sample spot size of the illumination 805, which may for example be around 2 mm in diameter. In Figure 8 the variable voltage source 819 and current measurement device 821 are shown as separate components, while in other embodiments they may be combined into a single system such as a variable power supply.

[0057] For the present method the to the multi-junction solar cell 200 by the variable voltage source 819 should be less than the open-circuit voltage of the illuminated region 817-1 of the silicon sub-cell 204 under the present illumination conditions 805. To ensure that the illumination 805 generates substantially no excess charge carriers in the perovskite sub-cell 202, the light source 823 of the illumination system 815 is preferably selected to emit photons that are below band-gap for the perovskite material.

[0058] Excess charge carriers generated in the selected illuminated region 817-1 of the silicon sub-cell 204 will produce a local voltage Vlocalacross the pn-junction of the silicon sub-cell. When the selected illuminated region 817-1 is laterally proximate to a shunt 230-1 in the overlying perovskite sub-cell 202 as shown in Figure 8, the local voltage will drive a current I through the shunt equal to: I = (Vlocal – Vapplied) / (Rshunt + Rseries) (1) where Vapplied is the voltage applied to the multi-junction cell 200 by the variable voltage source 819, Rshuntis the shunt resistance of the shunt 230-1 and Rseriesis the combined series resistance of all other parts of the current path. Measurements of the current I at two or more values of Vapplied will provide, as the negative of the inverse of the slope of I versus Vapplied, a value for (Rshunt+ Rseries). If Rseriesis relatively small, say around a few ^ or less, the current will be limited by the shunt, so that the negative inverse slope will provide a good estimate for Rshunt. Even if Rseriesis not small, the negative inverse slope of I versus Vapplied will at least provide an upper bound for Rshunt, which is still considered to be an estimate for Rshunt. In cases where there is significant lateral transport of photo- generated carriers, the estimate for Rshuntmay be affected by carriers generated in areas proximate to the shunt in the selected illuminated region 817-1. Accordingly, the accuracy of the estimate for Rshuntwill be higher when the lateral transport of photo-generated carriers is low.

[0059] In a preferred embodiment the locations of shunts 230-1, 230-2 in the perovskite sub-cell 202 will already have been ascertained by the qualitative method previously described with reference to Figures 3 to 5C. The shunt location information can then beused to guide the illumination system illuminate corresponding laterally proximate locations 817-1, 817-2 in the silicon sub-cell 204 for estimation of local shunt resistance. In other embodiments the illumination 805 may be raster scanned across the multi- junction cell 200 with synchronous modulation or sweeping of the applied voltage to obtain a map of shunt resistance across the perovskite sub-cell 202.

[0060] Figure 9 shows four images similar to the simulated shunt-identifying difference image of Figure 5C, each with a numbered circled region indicating the position of a selected illuminated region 817 of the underlying silicon sub-cell 204. In each image the greyscale represents the magnitude of the shunt conductance in units of mΩ-1cm-2. Figure 10 shows plots 1001, 1002, 1003 and 1004 of measured current I as a function of Vapplied at the corresponding locations 1, 2, 3 and 4 of Figure 9. Figure 10 also shows equations for each plot, calculated by linear regression. Plots 1001, 1002 and 1003 were obtained at the locations of previously identified shunts, while plot 1004 was obtained at a location, such as the selected region 817-3 in Figure 8, where there are no shunts in the corresponding area 825 of the perovskite sub-cell 202. The negative inverse slopes of plots 1 to 4 respectively provide values of approximately 800 ^, 503 ^, 962 ^ and 16.1 k ^ for the combined shunt and series resistance (Rshunt + Rseries) at the locations 1 to 4. If Rseries is of order a few ^ or less as will usually be the case, the Rshunt term will be dominant at each location, so that the negative inverse slope provides a close estimate for Rshunt.

[0061] The present method may also be used to estimate local shunt resistance in a first sub-cell of a monolithically integrated multi-junction solar cell for the case where the first sub-cell is below the second sub-cell and therefore has the smaller band-gap. In this case the illumination source 815 is configured to illuminate selected regions of the second, upper sub-cell 202 with light 805 suitable for generating excess charge carriers in that sub- cell, while generating substantially no excess charge carriers in the first, lower sub-cell 204, and the voltage applied by the variable voltage source 819 should be kept below the open-circuit voltage of the illuminated region of the second, upper sub-cell 202 under the present illumination conditions.

[0062] As with the previously described and apparatus for identifying shunts, the method and apparatus for estimating local shunt resistance in a first sub-cell of a multi- junction solar cell are not limited to multi-junction solar cells having only two monolithically integrated sub-cells. Provided the illumination conditions are chosen to generate excess charge carriers in all of the sub-cells except for the sub-cell being investigated for shunt resistance, the method and apparatus can also be applied to multi- junction solar cells having three or more monolithically integrated sub-cells.

[0063] The apparatus 801 shown in Figure 8 can also be used in a qualitative mapping mode for identifying shunts in a first sub-cell of a multi-junction solar cell having monolithically integrated first and second sub-cells, according to a third aspect of the present invention. This aspect will be described for the case where the first sub-cell is above the second sub-cell and therefore has the larger band-gap, with the first and second sub-cells for example composed of a perovskite material and silicon respectively.

[0064] For this qualitative mapping mode the illumination system 815 is configured to sequentially illuminate a plurality of selected regions 817-1, 817-2, 817-3 of the second, silicon sub-cell 204 with light 805 suitable for generating excess charge carriers in the second sub-cell 204, while generating substantially no excess charge carriers in the first, perovskite sub-cell 202. The voltage source 819 is configured to apply to the multi- junction solar cell 200, while the selected regions 817-1, 817-2, 817-3 of the second sub- cell 204 are being illuminated, a voltage that is less than the open-circuit voltage of the illuminated regions 817-1, 817-2, 817-3 of the second sub-cell 204. The current measurement device 821 is configured to measure, at each of the selected regions 817- 1,817-2, 817-3 of the second sub-cell 204, the current induced in the multi-junction solar cell 200 by the illumination 805, and the computer 813 is configured to identify shunts in the first sub-cell 202 based on variations in the measured current. Since the induced current measured at a given illuminated region will generally be inversely proportional to the combined shunt and series resistance (Rshunt+ Rseries) according to equation (1), higher currents will be measured when illuminating regions 817-1, 817-2 of the silicon sub- cell 204 that are laterally proximate to shunts 230-1, 230-2 in the first sub-cell 202. Furthermore, the magnitude of variations in the measured current enables estimation ofthe severity of shunts, i.e. the inverse of resistance, in the first sub-cell 202. This enables a qualitative map of shunt resistance to be obtained by measuring current I at constant Vapplied while the illumination 805 is scanned across the multi-junction cell 200. While this approach has some similarities to LBIC, it differs in that the illumination 805 is not generating excess charge carriers in the sub-cell 202 of interest for the shunt investigation, but in the other sub-cell 204.

[0065] As with the previously described methods, this qualitative shunt mapping method can also be applied to monolithically integrated multi-junction solar cells in which the first sub-cell, i.e. the sub-cell of interest, is below the second sub-cell and therefore has the smaller band-gap, as well as to multi-junction solar cells having three or more monolithically integrated sub-cells.

[0066] Although the present invention has been described with particular reference to certain preferred embodiments thereof, variations and modifications of the present invention can be effected within the spirit and scope of the following claims.

Claims

We Claim:

1. A method for identifying the presence of shunts in a first sub-cell of a multi- junction solar cell having monolithically integrated first and second sub-cells, the method comprising: (a) illuminating the multi-junction solar cell with light suitable for generating luminescence from the second sub-cell, while generating substantially no excess charge carriers in the first sub-cell; (b) acquiring a first image of luminescence generated from the second sub-cell, with the multi-junction solar cell under a first load condition; (c) acquiring a second image of luminescence generated from the second sub-cell, with the multi-junction solar cell under a second load condition; and (d) identifying the presence of shunts in the first sub-cell based on a comparison of intensities in the first and second images of luminescence.

2. The method according to claim 1, wherein the first image of luminescence is acquired with the multi-junction solar cell under open-circuit conditions.

3. The method according to claim 1 or claim 2, wherein the second image of luminescence is acquired with the multi-junction solar cell under short-circuit or near short-circuit conditions.

4. The method according to any one of the previous claims, wherein the comparison of intensities includes calculating a difference image from the first and second images of luminescence.

5. The method according to any one of the previous claims, wherein an absorber of the first sub-cell has a larger band-gap than an absorber of the second sub-cell.

6. The method according to claim 5, wherein the energy of photons in the light is less than the band-gap of the absorber of the first sub-cell.

7. The method according to claim 5 or 6, wherein the absorber of the first sub-cell includes a perovskite material and the absorber of the second sub-cell includes one ofeither silicon or cadmium telluride or or a combination of one or more of copper, indium, gallium, selenium or sulphur.

8. The method according to any one of claims 1 to 3, further comprising: (e) determining the location of shunts in the first sub-cell based on spatial variations in intensity in the first image of luminescence or in the second image of luminescence.

9. The method according to claim 4, further comprising: (e) determining the location of shunts in the first sub-cell based on spatial variations in intensity in the difference image.

10. An apparatus for identifying the presence of shunts in a first sub-cell of a multi- junction solar cell having monolithically integrated first and second sub-cells, the apparatus comprising: a light source for illuminating the multi-junction solar cell with light suitable for generating luminescence from the second sub-cell, while generating substantially no excess charge carriers in the first sub-cell; a system for applying different load conditions to the multi-junction solar cell; and an image capture device for acquiring first and second images of luminescence generated from the second sub-cell, with the multi-junction solar cell under different load conditions.

11. The apparatus according to claim 10, further comprising a computer for identifying the presence of shunts in the first sub-cell based on a comparison of intensities in the first and second images of luminescence.

12. The apparatus according to claim 11, wherein the computer is configured to determine the location of shunts in the first sub-cell based on spatial variations in intensity in the first image of luminescence or in the second image of luminescence.

13. The apparatus according to claim 11 or claim 12, wherein the computer is configured to compare intensities in the first and second images of luminescence by calculating a difference image from the first and second images of luminescence.

14. The apparatus according to wherein the computer is configured to determine the location of shunts in the first sub-cell based on spatial variations in intensity in the difference image.

15. The apparatus according to any one of claims 10 to 14, wherein the apparatus is configured to acquire the first image of luminescence with the multi-junction solar cell under open-circuit conditions.

16. The apparatus according to any one of claims 10 to 14, wherein the apparatus is configured to acquire the second image of luminescence with the multi-junction solar cell under short-circuit or near short-circuit conditions.

17. The apparatus according to any one of claims 10 to 16, wherein the apparatus is configured to identify the presence of shunts in the first sub-cell of a multi-junction solar cell in which an absorber of the first sub-cell has a larger band-gap than an absorber of the second sub-cell.

18. The apparatus according to claim 17, wherein the light source is configured to illuminate the multi-junction solar cell with light having photon energy less than the band- gap of the absorber of the first sub-cell.

19. The apparatus according to claim 17 or claim 18, wherein the apparatus is configured to identify the presence of shunts in the first sub-cell of a multi-junction solar cell in which the absorber of the first sub-cell includes a perovskite material and the absorber of the second sub-cell includes one of either silicon or cadmium telluride or perovskite or a combination of one or more of copper, indium, gallium, selenium or sulphur.

20. A method for estimating local shunt resistance in a first sub-cell of a multi- junction solar cell having monolithically integrated first and second sub-cells, the method comprising: (a) illuminating a selected region of the second sub-cell with light suitable for generating excess charge carriers in the second sub-cell, while generating substantially no excess charge carriers in the first sub-cell;(b) applying a voltage to the solar cell while the selected region of the second sub-cell is being illuminated; (c) measuring the current induced in the multi-junction solar cell by the illumination, for at least two different values of the applied voltage that are less than the open-circuit voltage of the illuminated region of the second sub-cell; and (d) determining, from the dependence of the measured current on the applied voltage, an estimate of the local shunt resistance in an area of the first sub-cell laterally proximate to the illuminated region of the second sub-cell.

21. The method according to claim 20, wherein the selected region of the second sub- cell is laterally proximate to the position of a suspected shunt in the first sub-cell.

22. The method according to claim 20 or claim 21, wherein an absorber of the first sub-cell has a larger band-gap than an absorber of the second sub-cell.

23. The method according to claim 22, wherein the energy of photons in the light is less than the band-gap of an absorber of the first sub-cell.

24. The method according to claim 22 or claim 23, wherein the absorber of the first sub-cell includes a perovskite material and the absorber of the second sub-cell includes one of either silicon or cadmium telluride or perovskite or a combination of one or more of copper, indium, gallium, selenium or sulphur.

25. An apparatus for estimating local shunt resistance in a first sub-cell of a multi- junction solar cell having monolithically integrated first and second sub-cells, the apparatus comprising: an illumination system for illuminating a selected region of the second sub-cell with light suitable for generating excess charge carriers in the second sub-cell, while generating substantially no excess charge carriers in the first sub-cell; a variable voltage source for applying a voltage to the multi-junction solar cell while the selected region of the second sub-cell is being illuminated; a current measurement device for measuring the current induced in the multi- junction solar cell by the illumination, for at least two different values of the appliedvoltage that are less than the open- of the illuminated region of the second sub-cell; and a computer for determining, from the dependence of the measured current on the applied voltage, an estimate of the local shunt resistance in an area of the first sub-cell laterally proximate to the illuminated region of the second sub-cell.

26. The apparatus according to claim 25, wherein the apparatus is configured to illuminate a selected region of the second sub-cell that is laterally proximate to the position of a suspected shunt in the first sub-cell.

27. The apparatus according to claim 25 or claim 26, wherein the apparatus is configured to estimate local shunt resistance in the first sub-cell of a multi-junction solar cell in which an absorber of the first sub-cell has a larger band-gap than an absorber of the second sub-cell.

28. The apparatus according to claim 27, wherein the illumination system is configured to illuminate the selected region of the second sub-cell with light having photon energy less than the band-gap of an absorber of the first sub-cell.

29. The apparatus according to claim 27 or claim 28, wherein the apparatus is configured to estimate local shunt resistance in the first sub-cell of a multi-junction solar cell in which the absorber of the first sub-cell includes a perovskite material and the absorber of the second sub-cell includes one of either silicon or cadmium telluride or perovskite or a combination of one or more of copper, indium, gallium, selenium or sulphur.

30. A method for identifying shunts in a first sub-cell of a multi-junction solar cell having monolithically integrated first and second sub-cells, the method comprising: (a) sequentially illuminating a plurality of selected regions of the second sub-cell with light suitable for generating excess charge carriers in the second sub-cell, while generating substantially no excess charge carriers in the first sub-cell;(b) applying to the multi- cell, while the selected regions of the second sub-cell are being illuminated, a voltage that is less than the open-circuit voltage of the illuminated regions of the second sub-cell; (c) measuring, at each of the selected regions of the second sub-cell, the current induced in the multi-junction solar cell by the illumination; and (d) identifying shunts in the first sub-cell based on variations in the measured current.

31. The method according to claim 30, further comprising estimating the severity of shunts in the first sub-cell based on the magnitude of the variations in the measured current.

32. The method according to claim 30 or claim 31, wherein an absorber of the first sub-cell has a larger band-gap than an absorber of the second sub-cell.

33. The method according to claim 32, wherein the energy of photons in the light is less than the band-gap of the absorber of the first sub-cell.

34. The method according to claim 32 or claim 33, wherein the absorber of the first sub-cell includes a perovskite material and the absorber of the second sub-cell includes one of either silicon or cadmium telluride or perovskite or a combination of one or more of copper, indium, gallium, selenium or sulphur.

35. An apparatus for identifying shunts in a first sub-cell of a multi-junction solar cell having monolithically integrated first and second sub-cells, the apparatus comprising: an illumination system for sequentially illuminating a plurality of selected regions of the second sub-cell with light suitable for generating excess charge carriers in the second sub-cell, while generating substantially no excess charge carriers in the first sub- cell; a voltage source for applying to the multi-junction solar cell, while the selected regions of the second sub-cell are being illuminated, a voltage that is less than the open- circuit voltage of the illuminated regions of the second sub-cell;a current measurement device for at each of the selected regions of the second sub-cell, the current induced in the multi-junction solar cell by the illumination; and a computer for identifying shunts in the first sub-cell based on variations in the measured current.

36. The apparatus according to claim 35, wherein the computer is configured to estimate the severity of shunts in the first sub-cell based on the magnitude of the variations in the measured current.

37. The apparatus according to claim 35 or claim 36, wherein the apparatus is configured to identify shunts in the first sub-cell of a multi-junction solar cell in which an absorber of the first sub-cell has a larger band-gap than an absorber of the second sub-cell.

38. The apparatus according to claim 37, wherein the illumination system is configured to illuminate the selected regions of the second sub-cell with light having photon energy less than the band-gap of the absorber of the first sub-cell.

39. The apparatus according to claim 37 or claim 38, wherein the apparatus is configured to identify shunts in the first sub-cell of a multi-junction solar cell in which the absorber of the first sub-cell includes a perovskite material and the absorber of the second sub-cell includes one of either silicon or cadmium telluride or perovskite or a combination of one or more of copper, indium, gallium, selenium or sulphur.

40. A non-transitory computer readable medium with an executable program code configured to implement the method according to any one of claims 1 to 9, 20 to 24 or 30 to 34, or to operate the apparatus according to any one of claims 10 to 19, 25 to 29 or 35 to 39.

Images