WO2016012274A1

WO2016012274A1 - Organic-inorganic tandem solar cell

Info

Publication number: WO2016012274A1
Application number: PCT/EP2015/065826
Authority: WO
Inventors: Felix Eickemeyer; Daniel WALDMANN; Wilfried HERMES; Peter Erk; Maraike Ahlf; Michaela Agari; Maximilian HEMGESBERG
Original assignee: Basf Se
Priority date: 2014-07-21
Filing date: 2015-07-10
Publication date: 2016-01-28
Also published as: TW201607092A

Abstract

The present invention relates to organic-inorganic tandem solar cells. In particular the present invention relates to solar cells comprising the following layers in the following order: a metal layer (10), a n-doped crystalline silicon layer (20), a p-doped crystalline silicon layer (30), an electron-conducting layer (40), a perovskite layer (50), a hole-conducting layer (60), and a transparent conductor layer (70).

Description

Organic-inorganic Tandem Solar Cell

The present invention is in the field of organic-inorganic tandem solar cells. The present invention further relates to a process for producing these solar cells and their use for the generation of electrical energy.

Solar cells play an important role for the environmentally benign power generation as they do not depend on fossil resources upon operation. In order to make solar cells even more attractive from an economic point of view, the cost per unit of electrical energy generated needs to be reduced. This can either be achieved by improving the solar-to-electrical power conversion efficiency or by reducing the production cost of the solar cells. In conventional solar cells this efficiency is limited by the band gap of the absorber. For large band gaps only those photons are absorbed with an energy exceeding the band gap thereby discarding the energy of the other photons. For small band gaps of the absorber, the amount of energy exceeding the band gap of high-energy photons is lost. Tandem solar cells provide a solution to overcome this dilemma by combining a solar cell containing a high-band-gap absorber with a solar cell containing a low- band-gap absorber. The former absorbs the high-energy photons with low energy loss, the latter absorbs the remaining photons. Tandem solar cells are known. WO 2014 / 045 021 A1 discloses a tandem solar cell in which a complex silicon-based solar cell is combined with a perovskite-sensitized solar cell.

It was an objective of the present invention to provide a solar cell which can be made with relatively little effort and which has at the same time a high solar-to-electrical power conversion effi- ciency.

These objectives were achieved by a solar cell comprising the following layers in the following order a metal layer (10),

a n-doped crystalline silicon layer (20),

a p-doped crystalline silicon layer (30),

an electron-conducting layer (40),

a perovskite layer (50),

a hole-conducting layer (60), and

a transparent conductor layer (70).

The present invention further relate to a process for preparing the solar cell according to the present invention comprising

(A) preparing a metal layer (10) with an n-doped crystalline silicon layer (20) and a p-doped crystalline silicon layer (30) and

(B) sequentially depositing on top of the p-doped silicon layer (30) a charge recombination layer (35) if present, an electron-conducting layer (40), a porous layer (45) if present, a perov- skite layer (50), a hole-conducting layer (60), and a transparent conductor layer (70).

The present invention further relates to the use of the solar cell according to the present inven- tion for the generation of electrical energy.

The present invention further relates to a solar panel comprising the solar cell according to the present invention. Preferred embodiments of the present invention can be found in the description and the claims. Combinations of different embodiments fall within the scope of the present invention.

The solar cell according to the present invention comprises several layers which are generally arranged in a stack as shown in figure 1 . Layer in the context of the present invention refers to a thin structure with an arbitrary surface. It may be flat, but in most cases it is very rough. A layer can even form an interpenetrating network with an adjacent layer to increase its contact area to the adjacent layers. A layer can have holes of various sizes and shapes. Unless indicated otherwise, a layer typically covers at least 50 % of the underlying surface, preferably at least 70 %, more preferably at least 90 %, in particular complete or essentially complete. The thickness of a layer can be measured by transmission electron microscopy.

According to the present invention the solar cell comprises a metal layer (10). Suitable metals have a high work function, preferably at least 4.0 eV, more preferably at least 4.25 eV, in particular at least 4.5 eV. The work function of a metal is typically measured by X-ray photoemission spectroscopy (XPS) as described in ISO 13424 EN (Surface chemical analysis - X-ray photoe- lectron spectroscopy - Reporting of results of thin-film analysis; October 2013). Examples for such metals are Co, Rh, Ir, Ni, Pd, Pt, Cu, Ag, Au, or alloys comprising these such as Ti/Pd/Ag. Ag or Ag alloys are preferred, in particular Ag or Ti/Pd/Ag. The metal layer (10) is usually just thick enough to ensure electrical conductivity, such as 1 to 100 μηη, preferably 10 to 25 μηη.

The solar cell according to the present invention further comprises a n-doped crystalline silicon layer (20) and a p-doped crystalline silicon layer (30). Crystalline silicon can mean polycrystal- line or single-crystalline, preferably single-crystalline. According to the present invention the n-doped crystalline silicon layer (20) typically contains 10¹³ to 10¹⁸ atoms/cm³ of elements of group 15 in the periodic table of the elements as dopant, preferably 10¹⁴ to 10¹⁶ atoms/cm³. The dopant concentration can for example be measured by the four-probe method described by D. Schroder in "Semiconductor material and device characterization", 3^rd edition, IEEE Press, 2006. Preferred elements are P, As, Sb, in particular P. Preferably, the concentration of dopant in the n-doped crystalline silicon layer (20) is higher on the surface facing the metal layer (10) relative to the remaining part of the n-doped crystalline silicon layer (20), in particular 10¹⁸ to 10²¹ atoms/cm³. This higher dopant concentration typically extends 0.05 to 1 μηι from said surface into the n-doped crystalline silicon layer (20), preferably 0.1 to 0.6 μηη, in particular 0.2 to 0.4 μηη.

The n-doped crystalline silicon layer (20) has preferably a resistivity of 0.1 to 20 Ω-cm, in partic- ular 1 to 7 Ω-cm. The surface of the n-doped crystalline silicon layer (20) facing the metal layer (10) has preferably a sheet resistance of 10 to 200 Ω/sq, preferably 50 to 120 Ω/sq, in particular 70 to 90 Ω/sq. The sheet resistance can be measured by four probe resistivity measurements as described for example by Smits in The Bell System Technical Journal, 1958, page 71 1 -718. The n-doped crystalline silicon layer (20) preferably has a layer thickness of 10 to 500 μηη, more preferably of 20 to 300 μηη, in particular 50 to 200 μηη.

Preferably the solar cell comprises a silicon nitride layer (15) between the metal layer (10) and the n-doped crystalline silicon layer (20). The silicon nitride layer (15) usually has a thickness of 1 to 500 nm, preferably 10 to 200 nm, in particular 50 to 100 nm.

According to the present invention the p-doped crystalline silicon layer (30) typically contains 10¹⁶ to 10²¹ atoms/cm³ of elements of group 13 in the periodic table of the elements as dopant, preferably 10¹⁹ to 5-10²⁰ atoms/cm³. Preferred elements are B, Al, Ga, In, in particular B. Usually, the p-doped crystalline silicon layer (30) has a surface resistivity of 10 to 200 Ω/sq, preferably 30 to 100 Ω/sq, in particular 50 to 70 Ω/sq. The thickness of the p-doped crystalline silicon layer (30) is preferably 0.1 to 10 μηη, more preferably 0.2 to 3 μηη, in particular 0.3 to 1 μηη.

Preferably the solar cell further comprises a charge recombination layer (35) between the p- doped crystalline silicon layer (30) and the electron-conducting layer (40). Suitable materials for the charge recombination layer (35) are electrically conductive which usually means that their resistivity is not more than 0.1 Ω-cm, preferably not more than 10^"2 Ω-cm, in particular not more than 10^"3 Ω-cm. Suitable materials include metals like Ti, Cr, Fe, Co, Ni or Cu; alloys like Fe/C, Fe/Cr/V or Cr/Mn; carbonaceous materials like graphite, carbon nanotubes, graphene; conductive polymers like poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate) (PEDOT-PSS).

Preferably, the charge recombination layer (35) comprises highly doped semiconductors. Even more preferably, the charge recombination layer (35) comprises a combination of highly n- doped and highly p-doped semiconductors wherein the highly p-doped semiconductor is primarily located at or near the surface facing the p-doped silicon layer (30) and the highly n- doped semiconductor is primarily located at or near the surface facing the electron-conducting layer (40). Highly doped in the present context usually means that the semiconductor has a high density of free charge carriers, such as 10¹⁸ to 10²¹ cm-³. Doping can be achieved by including elements other than making up the semiconductor or by intrinsic defects like vacancies in the crystal lattice. Doping can also be achieved by a combination of defects and other elements in the semiconductor. Preferably, the charge recombination layer (35) comprises highly p-doped silicon, in particular highly boron-doped silicon, and n-type metal oxides, in particular fluorine- doped tin oxide (FTO), indium-doped tin oxide (ITO) or aluminum-doped zinc oxide (AZO). The charge recombination layer (35) should be optically transparent. Transparent in the context of the present invention means that light of 550 nm wavelength is transmitted to at least 50 %, preferably at least 70 %, in particular at least 85 %. If the materials used for the charge recombination layer (35) have a high extinction coefficient such as metals, it is advantageous if the charge recombination layer (35) contains holes. These holes can be isolated such as in a grid or contact each other. The holes can contact each other in such a way that the charge recombination layer (35) becomes discontinuous. An example for such a discontinuous layer is an array of isolated droplets. An array of isolated metal droplets is preferred. The distance between different parts of the material making up the charge recombination layer (35) should be small enough to ensure that either holes or electrons can reach the charge recombination layer (35). Preferably any point in the surface contacted by the charge recombination layer (35) is at maximum at a distance of 10 μηη away from the next part of the material making up the charge recombination layer (35), more preferably at maximum 5 μηη, in particular at maximum 1 μηη. According to the present invention the solar cell further comprises an electron-conducting layer (40). Suitable materials include semi-conductive metal oxides including oxides of titanium, tin, zinc, iron, tungsten, zirconium, hafnium, strontium, indium, cerium, yttrium, lanthanum, vanadium, caesium, niobium or tantalum. Further, composite semiconductors such as M¹ _xM²yM³ _zOt may be used, wherein M¹ , M² and M³ are independent of each other a metal atom, O is an oxy- gen atom, and x, y, z and t are numbers including 0 which are chosen such that a non-charged molecule is formed. Examples are T1O2, Sn02, Fe203, WO3, ZnO, Nb20s, SrTi03, Ta20s, CS2O, zinc stannate, complex oxides such as barium titanate, binary and ternary iron oxides, and indium gallium zinc oxide (IGZO). Preferred materials are T1O2, SnC"2, ZnO, in particular T1O2. The metal oxides can further be doped. The electron-conducting layer (40) has preferably a high surface area, it is for example porous.

Preferred methods for producing the semi-conductive metal oxides are sol-gel methods described for example in Materia, Vol. 35, No. 9, Page 1012 to 1018 (1996). The method developed by Degussa Company, which comprises preparing oxides by subjecting chlorides to a high temperature hydrolysis, is also preferred.

In the case of using titanium oxide as the semi-conductive metal oxides, the above-mentioned sol-gel methods, gel-sol methods, high temperature hydrolysis methods are preferably used. Other preferred sol-gel methods are those described in Barbe et al., Journal of American Ce- ramie Society, Vol. 80, No. 12, Page 3157 to 3171 (1997).

According to the present invention the solar cell further comprises a perovskite layer (50) which comprises perovskite absorbers. Perovskite absorbers are typically compounds of the general formula (I): AMX3. A stands for an alkali metal such as Li, Na, K, Rb, Cs; or an ammonium ion in which one or more hydrogen atoms may be exchanged by alkyl or acyl groups. Ammonium ions in which one or more hydrogen atoms are exchanged by alkyl groups include monoalkylammo- nium ions, dialkylammonium ions, trialkylammonium ions, tetraalkylammonium ions. Preferably, the alkyl group or groups are independent of each other Ci to C6 alkyl groups, in particular methyl or ethyl. Ammonium ions in which one or more hydrogen atoms are exchanged by acyl groups include amidinium ions and N-alkylamidinium, preferably amidinium ions. Preferably, the amidinium ion is derived from a Ci to C6 carboxamide, in particular from formamide or acetam- ide. Preferably A is Cs or an ion comprising a positively charged nitrogen atom.

In general formula (I) M stands for a divalent metal atom, preferably for Pb or Sn. X stands for halogens, in particular CI, Br, I. X in compounds of general formula (I) can contain all the same or different halogens. Specific examples for perovskite absorbers include methyl ammonium lead halogenides, such methyl ammonium lead iodide (CHsNHsPb ) or CHsNHsPbBr ; forma- dinium lead halogenides like formamidinium lead iodide (HC(NH2)Pbl3), formamidinium lead bromide (HC(NH2)PbBr3) or formamidinium lead chloride iodide (HC(NH2)PbCl2l); or cesium tin iodide (CsSn ). Methyl ammonium lead halogenides and formamidinium lead halogenides are preferred.

It is further possible that the perovskite layer (50) according to the present invention comprises in addition to the perovskite absorber other materials. In this case the perovskite layer often comprises a mixture of a perovskite absorber and another material. The other material is often in an amorphous state. Preferred examples are alkylammonium lead halogenide or alkylammo- nium tin halogenide as described above mixed with additional alkylammonium halogenide.

Sometimes this mixture is represented by one combined formula, for example Ai+iMX3+i, wherein A, M and X have the same meaning as above and i is number larger than zero and preferably smaller than 6, more preferably, i is a number in the range of 3 to 5. Two specific examples are (CHsNH^sPbls.sCls.s or (CHsNHs^Snls.sC .s.

Furthermore, the solar cell may further comprise a porous layer (45) of an insulating metal oxide between the electron-conducting layer (40) and the perovskite layer (50). The pores of the porous layer (45) are filled with the material making up the perovskite layer (50). Suitable metal oxides are for example AI2O3, S1O2, Zr02, or MgO.

According to the present invention the solar cell further comprises a hole-conducting layer (60) comprising a hole-conducting material. Hole-conducting materials can be inorganic or organic.

Inorganic hole-transporting materials preferably contain a Cu(l) species such as Cul, CuSCN, CulnSe₂, Cu(ln,Ga)Se₂, CuGaSe₂, Cu₂0, CuS, CuGaS₂, CulnS₂, CuAISe₂. Cul and CuSCN are preferred. Other inorganic hole-transport materials are GaP, NiO, CoO, FeO, B12O3, M0O2, Cr₂0₃.

Organic hole-transporting materials include p-type semiconducting organic conjugated polymers and reversibly oxidizable low molecular weight materials. Examples for conjugated polymers are polyacetylene; polyphenylene; polyaniline; polytoluidine; poly(triphenylamine); polypyrrole; poly- thiophene such as poly(3-hexylthiophene), poly(3,4-ethylenedioxythiophene) or poly(4-undecyl- 2,2'-biothiophene); polyarylenevinylene such as poly(p-phenylenevinylene) or polythienylene- vinylene; poly(N-vinylcarbazole); donor-acceptor copolymers of benzotriazole, benzothiadiazole, pyridathiadiazole, diketopyrrolopyrrol, thienothiophene or thieno-pyrrole-2,6-dione with arylenes and heteroarylenes.

Examples for reversibly oxidizable low molecular weight materials are aromatic amines as described e.g. in WO 2010 / 094 636; triphenylenes disclosed e.g. in JP 1 1 176 489; oligothio- phene compounds disclosed e.g. in Journal of the American Chemical Society 120 (1998) 664- 672; oligoselenophene compounds. It is possible to use one single organic hole-transport mate- rial or mixtures of two or more organic hole-transport materials.

Preferred organic hole-transporting materials are spirobifluorenes (see for example

US 2006 / 0 049 397), more preferred are asymmetric spirobifluorenes as disclosed in

WO 2014 / 037 847. A particularly preferred spirobifluorene is 2,2',7,7'-tetrakis(N,N-di-p- methoxyphenyl-amine)9,9'-spirobifluorene.

Preferably the hole-conducting layer (60) further comprises a dopant to increase the hole conductivity. Examples for such dopants are N(PhBr)3SbCl6, silver bis-(trifluoromethylsulfonyl)imide, metal oxides such as V2O5, or copper complexes as disclosed for example in

WO 201 1 / 033 023. Furthermore, the hole conducting layer (60) may contain additives, e.g. lithium bis-(trifluoromethylsulfonyl)imide, hydroxamates or 4-feri-butylpyridine.

According to the present invention the solar cell further comprises a transparent conductor layer (70) which comprises a transparent conductor. Suitable transparent conductors include trans- parent conductive oxides (TCO). These include Sn02 which can be doped with In, F, CI, As or Sb; ZnO which can be doped with Al, Ga or In; Ιη2θ3 which can be doped with Ge, Sn, Pb, As, or Sb. Indium-doped Sn02 (ITO) or fluorine-doped Sn02 (FTO) are preferred. If metals such as Ag or Au are used, the transparent conductor layer (70) has to be very thin, such as 1 to 100 nm, preferably 5 to 20 nm. Alternatively, the transparent conductor layer (70) can have holes which let the light shine through. This is for example realized in a metal grid, in which the mesh size is not more than 10 μηη, in particular not more than 1 μηη. Metal grids are preferred. It is further possible that nanowires are used which touch each other such that all wires are electrically connected but still leave a large percentage of the underlying surface unshaded. This approach is described by Yang et al. in ACS Applied Materials & Interfaces 3 (201 1 ) 4075-4084. Other suitable transparent conductors are conductive polymers, in particular poly(3,4- ethylenedioxythiophene) poly(styrenesulfonate) (PEDOT-PSS). PEDOT-PSS is preferred. It is further possible to combine several transparent conductors such as ITO or FTO in combination with PEDOT-PSS. Usually, the thickness of the transparent conductor layer (70) is 3 nm to 10 μηη, preferably 100 nm to 1 μηη. Preferably, the transparent conductor layer (70) comprises carbonaceous materials, in particular carbon nanotubes or graphene. Preferably the solar cell is encapsulated to make it more robust against mechanical stress or weather conditions. The encapsulation can be made of glass such as low-cost soda glass of high strength or non-alkali glass from which no alkaline elution occurs. Alternatively, a transparent polymer film may be used such as tetraacetyl cellulose (TAC), polyethylene terephthalate (PET), polyethylene naphthalate (PEN), syndiotactic polystyrene (SPS), polyphenylenesulfide (PPS), polycarbonate (PC), polyarylate (PAr), polysulfone (PSF), polyestersulfone (PES), poly- imide (PI), polyetherimide (PEI), polycycloolefin such as polynorbornene, or brominated phe- noxy resin. Glass is preferred. In general, the incident light hits the solar cell from the side of the transparent conductor layer (70). In this way the light first reaches the perowskite layer, which often preferentially absorbs light of the shorter wavelengths, and subsequently reaches the crystalline silicon layers which absorb the remaining light. The present invention also relates to a process for producing the solar cell according to the invention. This process comprises preparing a metal layer (10) with an n-doped crystalline silicon layer (20) and a p-doped crystalline silicon layer (30). This is usually done by starting from a silicon wafer, preferably an n-doped silicon wafer, in particular phosphor-doped silicon. Ranjan et al. describe how these are made in Computers & Chemical Engineering 35 (201 1 ) 1439- 1453. Preferably single-crystal silicon wafers are used which are made by the Czochralski process. The wafer is p-doped on one side by exposing it to a compound containing a group 13 element in the gaseous state. Examples of such compound include B2H6, BC , BBr3, AIH3, preferably BBr3. If the concentration of dopant in the n-doped crystalline silicon layer (20) is higher on the surface facing the metal layer (10) relative to the remaining part of the n-doped crystal- line silicon layer (20), the respective side of the wafer is exposed to a compound containing a group 15 element in the gaseous state. Examples of such compounds include PH3, PCI3, POCI3, AsHs, AsCIs, SbHs, SbCI₃, preferably POCI₃.

If a silicon nitride layer (15) is present this layer is usually made by chemical vapor deposition. Typical precursors are SihU and N2 or N2O.

The metal layer (10) can be made by applying metal-plating or vapor-depositing such as physical vapor deposition or chemical vapor to deposit the metal directly onto the layer it shall contact. Alternatively, the metal layer can be made by applying a metal-containing ink or paste onto the layer it shall contact and remove ingredients such as solvent by heating. Using metal- containing pastes is preferred, in particular using a silver-containing paste. The composition of such a metal paste is for example described in WO 201 1 / 026 769. Preferably, the application of the metal-containing ink or paste is done by a printing technique, in particular by screen printing.

The process according to the invention further comprises a sequentially depositing on top of the p-doped silicon layer (30) a charge recombination layer (35) if present, an electron-conducting layer (40), a porous layer (45) if present, a perovskite layer (50), a hole-conducting layer (60), and a transparent conductor layer (70). Preferably, the layers are sequentially deposited in the following order: p-doped silicon layer (30) a charge recombination layer (35) if present, an electron-conducting layer (40), a porous layer (45) if present, a perovskite layer (50), a hole- conducting layer (60), and a transparent conductor layer (70). Various processes for depositing these layers are available including vapor deposition processes. These include sublimation, physical vapor deposition, chemical vapor deposition, atomic layer deposition, or direct liquid injection. The material out of which the layer is formed is brought into the gaseous state and deposited onto the other layers. Preferably, the vapor processes are done under reduced pres- sure such as from 100 to 10^"8 mbar, more preferably from 10 to 10^"5 mbar, in particular 1 to 10 ² mbar.

Preferably the charge recombination layer (35) if present, the electron-conducting layer (40), the perovskite layer (50), the hole-conducting layer (60), and the transparent conductor layer (70) are deposited independent of each other by a wet-chemical coating process. These include spin-coating, spray-coating, dip-coating, drop-casting, doctor-blading, slot-die coating, 2D ink jet printing, gravure printing, offset printing, flexo printing, screen printing, or microcontact (wave) printing. After deposition the respective layer is preferably heated. Suitable temperatures are 100 to 600 °C, preferably 200 to 500 °C. Suitable time periods are 1 minute to 2 hours, preferably 10 minutes to 1 hour.

Description of the Figures

Figure 1 shows a schematic of the layer structure of the solar cell according to the present invention.

Figure 2 shows a scanning electron microscopy (SEM) image of a fractured solar cell.

Example

A two terminal organometal halide perovskite/ silicon tandem solar cell was fabricated according to the following steps:

Silicon solar cell:

A double-sided polished n-type <100> silicon wafer (phosphorous doped, POC thermal diffusion) was doped with boron in the front side (BBr3 diffusion) with a doping level of -7-10¹⁹ cm-³. The rear side of the wafers was passivated with a 75 nm-thick silicon nitride (SiN_x) film, which was subsequently opened by laser processing (Nd:YV04, 532 nm) with lines of about 50 μηη thick to access to the n-side of the silicon wafer. The rear contact was made by Ag, either due to Ag evaporation or by using a Ag paste.

Recombination layer:

Fluorine doped tin oxide (FTO) and indium doped tin oxide (ITO) were deposited by sputtering on the polished front side of the silicon solar cell.

To obtain a high quality interface between the front p-doped side of the silicon solar cell and the recombination layer of interest, two cleaning procedures were followed prior to the sputtering process to remove the native S1O2 layer grown at the silicon surface and possible contaminants from the diffusion process. The silicon solar cells were cleaned (i) in 5 wt-% HCI for 5 min prior to the FTO deposition, and (ii) 5 min in 5 wt-% HCI followed by 2 min in 2 wt-% HF prior to the ITO deposition. For the use of ITO and FTO very similar results were found. Blocking layer:

A T1O2 blocking layer was deposited by spray pyrolysis at 350 °C. 13 pulses consisting of 20 s spray time result in a thickness of -30 nm. The precursor solution consisted of 24.3 g titani- um(IV)diisopropoxide bis(acetylacetonate) (TAA) as a titanium source dissolved in 250 ml etha- nol.

Perovskite layer:

Mesoporous T1O2 scaffolds were spin coated onto the T1O2 blocking layer at 2500 rpm during 45 s (1 s acceleration) and subsequently sintered at 450 °C in a muffle furnace. The spin coating solution consisted of a molar ratio 1 :5 of 18-NRT T1O2 paste (Dyesol) in ethanol, which lead to a scaffold thickness of -250 nm.

A solution of methyl ammonium iodide (MAI) and lead chloride PbC with the molar ratio 3:1 was prepared inside a N2 glove-box. 397.4 mg MAI and 231.4 mg PbC were dissolved in 1 ml dimethylformamide (DMF) and stirred in a hot plate at 60 °C. The solution was then filtered through a 0.2 μηη PTFE filter to remove possible aggregates. The scaffold was pretreated in a plasma cleaning under O2 for 2 min to activate the surface of the scaffold. 100 μΙ of this solution was spun at 2000 rpm during 45 s onto the T1O2 scaffold layer. The resulting film was dried for 30 min on a drying furnace at 1 10 °C. The samples were left inside the furnace for cool down for additional 30 min.

Hole-conducting layer: A solution containing 60 mg 2,2',7,7'-Tetrakis-(N,N-di-4-methoxyphenylamino)-9,9'- spirobifluorene (spiro-OMeTAD) (purchased from Lumec), 1 1.6 μΙ tert-butyl pyridine (tBP) and 25.3 μΙ of a Li-TFSI solution (143.5 mg lithium-bis(trifluoromethanesulfonyl)imide (Li-TFSI) in 1 ml acetonitrile) were dissolved in 1 ml anhydrous chlorobenzene inside a N2 globe box. 100 μΙ of the spiro-OMeTAD solution were spun at 2500 rpm during 30 s. γ-Butyrolactone was used to clean the edges of the silicon solar cells from possible perovskite or spiro-OMeTAD contaminations.

Transparent conductor layer:

Gold thin films with a thickness of -14 nm were evaporated on top of the spiro-OMeTAD hole conductor layer. The top contacts consist of gold interdigitated finger electrodes of about 200 nm. Except for the T1O2 paste (purchased from Dyesol) and the spiro-OMeTAD (purchased from Lumec) all chemicals were purchased from Sigma Aldrich.

The performance of the tandem cell was determined by measuring their current-voltage (IV) curves with a source meter (Model 2400, Keithley Instruments) under the illumination of a solar simulator LS0106 with a Xenon arc lamp (LOT- Quantum Design). With the procedure described above a tandem cell with an open circuit voltage V_oc = 520 mV, a short circuit current Isc = 3.8 mA/cm², a fill facor FF = 24.5 % and an efficiency of 0.49 % was fabricated.

Figure 2 shows a scanning electron microscopy (SEM) image of a fractured solar cell exposing the different layers.

Claims

1 . A solar cell comprising the following layers in the following order a metal layer (10),

a n-doped crystalline silicon layer (20),

a p-doped crystalline silicon layer (30),

an electron-conducting layer (40),

a perovskite layer (50),

a hole-conducting layer (60), and

a transparent conductor layer (70).

2. The solar cell according to claim 1 , wherein a charge recombination layer (35) is further comprised between the p-doped crystalline silicon layer (30) and the electron-conducting layer (40).

3. The solar cell according to claim 2, wherein the charge recombination layer (35) comprises highly p-doped silicon and a n-type metal oxide. 4. The solar cell according to claim 2, wherein the charge recombination layer (35) comprises metal.

5. The solar cell according to any of the claims 1 to 5, wherein the electron-conducting layer (40) comprises T1O2, ZnO, or Sn02.

6. The solar cell according to any of the claims 1 to 6, wherein the perovskite layer (50) comprises a methyl ammonium lead halogenide or a formamidinium lead halogenide.

7. The solar cell according to any of the claims 1 to 7, wherein the hole-conducting layer (60) comprises copper iodide or copper thiocyanate.

8. The solar cell according to any of the claims 1 to 7, wherein the hole-conducting layer (60) comprises 2,2',7,7'-tetrakis(N,N-di-p-methoxyphenyl-amine)9,9'-spirobifluorene. 9. The solar cell according to any of the claims 1 to 9, wherein the transparent conductor layer (70) comprises poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate).

10. The solar cell according to any of the claims 1 to 9, wherein the transparent conductor layer (70) comprises indium-doped Sn02 or fluorine-doped Sn02.

1 1 . The solar cell according to any of the claims 1 to 9, wherein a silicon nitride layer (15) is further comprised between the metal layer (10) and the n-doped crystalline silicon layer (20).

A process for preparing the solar cell according to any of the claims 1 to 1 1 comprising

(A) preparing a metal layer (10) with an n-doped crystalline silicon layer (20) and a p- doped crystalline silicon layer (30) and

(B) sequentially depositing on top of the p-doped silicon layer (30) a charge recombina- tion layer (35) if present, an electron-conducting layer (40), a porous layer (45) if present, a perovskite layer (50), a hole-conducting layer (60), and a transparent conductor layer (70).

The process according to claim 12, wherein the charge recombination layer (35) if present, the electron-conducting layer (40), the perovskite layer (50), the hole-conducting layer (60), and the transparent conductor layer (70) are deposited independent of each other by a wet-chemical process.

Use of the solar cell according to any of the claims 1 to 1 1 for the generation of electrical energy.

15. A solar panel comprising the solar cell according to any of the claims 1 to 1 1.