IT201900005114A1 - Doping with 2D materials of photovoltaic multi-junction devices that include an absorber with perovskite structure - Google Patents

Description

TITLE

Doping with 2D materials of photovoltaic multi-junction devices that include an absorber with perovskite structure.

3. TECHNICAL FIELD

The invention reported here relates to tandem or multi-junction 2-terminal (2T) photovoltaic (PV) devices, in which one or more absorbing materials that make up the final structure of the device is a material with a perovskite crystalline structure and a chemical formula that includes cations organic or inorganic monovalent, divalent transition metals and halogens.

4. DESCRIPTION OF THE FIGURES

Figure 1 Schematic representation of a tandem solar cell comprising a front cell (2), a recombination layer (3), a rear cell (4) and the electrodes: front (1) and rear (5).

Figure 2 Schematic representation of all the layers of a perovskite solar cell in which graphene and / or related materials (GRM) are inserted inside the device. The structure consists of a conductive substrate (6), two charge-selective layers with opposite polarity (7) and (11), which can be doped (textured fill in the image) with GRM or 2D structure materials, and finally a counter electrode (13). Optional layers (dashed outlines) comprising GRMs or materials with 2D structure as inter-layer, (8) and (11), or as protective layers (12) can be added to this reference structure.

Figure 3 Schematic representation of the perovskite / silicon tandem solar cell presented in the example. The detailed description of all the layers is included in the “Example” section.

Figure 4 JV characteristic of two tandem cells including a graphene doped perovskite (PSC) cell (solid curve) and the non-doped reference cell (dashed curve).

5. BACKGROUND

The most promising strategy for increasing the performance of solar cells is based on multiple junctions (multi-junctions). To achieve this goal, the use of low-cost materials and processes is important.

A multi-junction solar cell is a photovoltaic technology that, unlike that with a single junction or single absorber, distributes the absorption of the electromagnetic spectrum available on different sub-cells.

Recently, M.A. Green showed that these silicon substrate-based multi-junction photovoltaic devices have the potential to achieve efficiencies of up to 42.5%. He pointed out that the optimal band-gap of the front cell is 1.7 eV for a tandem structure (with two sub-cells), nevertheless a cell with a smaller band-gap can be used if it has adequate transparency for the radiation with length d ' wave greater than its band-gap. (Green et al., 2013)

From an operational point of view, two types of tandem configurations can be distinguished: 2-terminal (2T) and 4-terminal (4T).

In a 2T architecture, commonly called monolithic tandem device, the realization of each sub-cell takes place directly above the underlying cell, which acts as a substrate. This requires process compatibility between all the technologies used. A further implication is that all junctions will be electrically connected in series.

The series connection requires that the same current flow throughout the structure and the final current generated by the device will be determined by the least performing sub-cell. For this reason, the optimization of the sub-cells by increasing the current, as often happens for single junction devices, is difficult to apply in 2T tandem PV devices.

The only way to further increase the tandem cell current is to simultaneously increase the currents generated in all sub-cells, avoiding parasitic optical losses and electrical losses. This condition is called current matching.

Despite this, the efficiency of 2T tandem devices can be improved by acting on two other parameters, namely the open circuit voltage (VOC) and the Fill Factor (FF).

Fig.1 shows a simple multi-junction with two sub-cells, which will be referred to below as front cell (2) and rear cell (4); between them is the recombination layer (3). The front contact (1) and the rear contact (5) are necessary for the collection of the electric charge and for its subsequent distribution to the load or to the connected circuit.

The tandem or multi-junction technology generally employs as back cell (4) devices based on crystalline silicon (c-Si) or hetero-junctions of amorphous and crystalline silicon (Si-HJ) or indium gallium copper selenide (CIGS) or indium-gallium arsenide (InGaAs) or copper zinc-tin sulphide (CZTS), but also other materials and alloys.

Most of the literature concerning front cell tandem PV devices (2) including materials with perovskite crystal structure are based on hetero-junction cells (4) of amorphous and crystalline silicon (a-Si: H / c-Si ) as recently shown by F. Sahli et al. (Sahli et al., 2018) or M. Jošt et al. (Jošt et al., 2018).

The main advantage linked to this technology, which has allowed to obtain the highest efficiency with single junction devices based on silicon are the following:

1) High potential linked to the energy band structure of thin film emitters;

2) Passivation of the silicon surface which can be obtained by introducing intermediate thin layers between crystalline silicon and doped amorphous silicon.

These characteristics are reflected in a higher VOC than that of conventional silicon-based technologies. This effect represents an advantage in a 2T tandem configuration in which the output voltage is equal to the sum of the voltages of the various sub-cells.

Perovskite-structured materials are ideal as absorbers in the front cell (2) due to their unusual properties. Their structure guarantees the possibility of inserting ions of different sizes and is particularly resilient to the formation of vacancies in the crystal lattice. Thanks to this versatility, they can be synthesized with an extreme variety of combinations of chemical elements.

Advances in the field of perovskite photovoltaic devices based on lead halide with organic and inorganic cations have been very rapid, with efficiencies reaching 23.7% on small area devices and 17% on modules with series connection, allowing these materials to compete with other thin-film photovoltaic technologies. A PSC exhibits higher internal potential and VOC relative to the band-gap than a crystalline silicon device under visible light illumination (400-800 nm).

Multiple architectures have been demonstrated and the most feasible for multi-junction applications is the planar structure made with low temperature solution fabrication techniques. In this structure the perovskite layer acts as a radiation absorber and is between two selective layers of the electric charge. Low-temperature processes are an essential aspect in the integration of perovskites on silicon hetero-junction cells (Si-HJ), since these cannot withstand temperatures above 200 ° C without incurring damage. The general procedures for manufacturing a tandem device are described, for example, in the patent WO2018221913A1, while as regards the perovskite devices, various processes related to their realization have been demonstrated in various patents, including US20150249170A1.

The versatility of perovskite as regards its chemical composition allows you to design materials with specific properties, including high thermal or chemical stability and various optical or electrical properties depending on the application. Recently, hybrid organic-inorganic halides with perovskite structure, such as, but not limited to, lead methylammonium iodide ((CH3NH3) PbI3 commonly referred to as MAPI or MALI in English languages) have been produced by substituting cations inorganic with iso-valent organic cations. Such materials have been widely used due to their simplicity of deposition during the manufacture of a solar cell.

Furthermore, the performance of the tandem device can be improved by adjusting the perovskite band-gap by inserting bromine (Br) or other halogens into the crystal structure. This strategy allows you to distribute the spectrum in such a way as to reach the condition of current "matching" and to increase the VOC of the entire tandem device. The distribution of the electromagnetic spectrum must not be symmetrical in each sub-cell, but must take into account the spectral response of the various sub-cells to the radiation, so as to ensure that the current generated in each cell is the same.

At the same time, the adjustment of the band-gap obtained by adding bromine generates the so-called Hoke effect, which consists in the segregation during illumination between phases rich in bromine and phases rich in iodine within the absorber material, this results in an increase in non-radiative recombinations assisted by trap states and, therefore, in the reduction of the performance of the device in operating conditions. This effect was addressed in 2015 by Eric T. Hoke et al. (Hoke et al., 2015)

The increase in VOC and FF of a multi-junction device can be achieved by passivating the interfaces with the addition of intermediate layers of thickness between 1 and 100 nm, however these additional layers could introduce optical losses due to reflections or unwanted absorptions. These optical losses could undermine the optimization of the current "matching" and, therefore, the performance of the entire device. In this regard, Graphene and other 2D materials have been used to increase the efficiency of perovskite devices.

The crystalline structure of graphite appears as a superposition of layers of graphene, held together by Van der Waals type bonds with an energy of about 2 eV / nm <2>, which makes the graphite easily exfoliated in the direction parallel to the crystalline plane, by means of the application of forces in the order of 300 nN / mm <2>.

Different techniques can be used to separate the crystalline planes: physical techniques of mechanical exfoliation and chemical techniques of intercalation of substances (inorganic acids, salts, solvents, etc.) which lead to the separation of the graphene planes. Further synthesis techniques consist of growth by chemical vapor deposition (CVD) on metal substrates or firing at high temperatures and vacuum heat treatments of silicon carbide (SiC).

Mechanical exfoliation consists in the application of a force on the surface of highly oriented graphite crystals (HOPG) to separate and reveal the crystalline planes until a single layer of graphene is obtained. Mechanical exfoliation is the simplest and most accessible technique for isolating graphene sheets as small as a few microns. However, it is a difficult method to industrialize.

Chemical exfoliation, on the other hand, makes it possible to obtain graphene, the compounds intercalated in it or graphene oxide, by means of simple chemical reactions or electro-chemical techniques. Specifically, the exfoliation of graphite oxide, a material that has the same lamellar structure as graphite, but in which some carbon atoms are bonded to oxygen in the form of hydroxyl or carbonyl groups (or more rarely epoxides and carboxyls) and in which the interlamellar distance is greater due to the steric hindrance of the oxygen. It has a strongly hydrophilic nature, which allows to obtain, through sonication or other techniques, the intercalation of water molecules and, therefore, a simple and almost complete exfoliation (about 90%) of the material in monoatomic oxide layers of graphene (GO).

However, GO is an insulating material in which the bonds with oxygen must be broken and therefore the carbon must be reduced in order to have properties similar to those of graphene. Both chemical reduction methods (using hydrazine, hydroquinone, sodium hydride and boron, or ascorbic acid), thermal methods, and methods based on UV radiation that produce materials with conductivity in the order of 100 S / cm. Despite this, the reduction processes often leave defects in the crystal structure that degrade the electronic properties.

Colloidal dispersions of graphene sheets were obtained by liquid phase exfoliation in organic solvents. The exfoliation, favored by sonication, occurs through the interaction of the solvent with the layers of graphene and is all the more effective the more the surface energy of the solvent is similar to that of graphene. There are several solvents that meet these characteristics, including N-methylpyrrolidone (NMP), N, N-dimethylacetamide (DMA), N, N-dimethylformamide (DMF), ��butyrolactone (GBL) etc. These methods have yields for obtaining mono-atomic layers that reach 50% by weight, moreover they are scalable and versatile.

The functionalizations can be different and depend on the type of use of the material within a photovoltaic device, for example graphene oxide functionalized with nitrogen (NGO) or with nano-particles of titania, or graphene oxide functionalized with lithium (GO-Li), or with chlorine (GO-Cl), or with porphyrins (GO-TPP), or with ethylene-dinitrobenzole (GO-EDNB), and others.

Specifically, GRMs have been used for the doping of selective contacts (Han et al., 2015; Luo et al., 2015; Umeyama et al., 2015) or as an intermediate layer between perovskite and selective layers (Wu et al., al., 2014; Yeo et al., 2014) in the form of graphene oxides (GO or rGO) obtaining an increase in the collection of charges at the respective electrodes and, therefore, an increase in efficiency. In the case of perovskite solar cells with a mesoporous structure (in which the photo-electrode is responsible for collecting negative charges), graphene oxide has been used as a dopant in the mesoporous layer in several publications.

Hans and collaborators (Acik & Darling, 2016) demonstrated the increase in performance in a device by creating a mesoporous layer using a graphene-mTiO2 nanocomposite synthesized from a graphite powder. The device obtained had a maximum efficiency of 14.5% using a concentration equal to 0.4% by weight of rGO in the mTiO2 solution.

Umeyama et al. (Umeyama et al., 2015) used rGO to doping both the compact layer of titania and the mesoporous layer, finding an optimal concentration of 0.15% and 0.015% by weight, respectively, for each layer.

Pure graphene nano-flakes were inserted into the cTiO2 solution by Wang and co-authors (Wang et al., 2014) with the aim of increasing the efficiency of flexible devices and obtaining a maximum of 15.6%. In this case, the mesoporous layer was made with alumina (Al2O3), a material whose process temperatures are more compatible with plastic substrates.

Graphene oxide in its reduced form (rGO) has recently been used as a dopant in the mesoporous layer of titania together with lithium salts (Li-TFSI) in mixed cation perovskite solar cells (Formamidinium FA and Methylammonium MA) and mixed anions (iodine and bromine). The authors achieved efficiencies greater than 19% (Cho et al., 2016).

In this context, Agresti et al. proposed an efficient PSC structure by introducing lithium neutralized graphene oxide (GO-Li) as an intermediate layer between the titania and the perovskite absorber. The effect of the introduction of the GO-Li layer was the increase of the short-circuit current (Jsc) and of the FF, positively influencing the electrical and thermal stability of the device, compared to the reference PSCs. In particular, a better extraction / injection of charges to the photo-electrode was observed thanks to the use of GO-Li, which allowed the passivation of the mTiO2 oxygen vacancies which act as reactive centers when the device is exposed to the humidity (Antonio Agresti, Pescetelli, Cinà, et al., 2016).

Furthermore, it has recently been reported that it is possible to improve the performance of a mesoporous device using graphene or other related 2D materials (GRM), integrating them into solution processes in PSC fabrication and easily reproducible on a large scale for the production of photovoltaic modules and large area devices. Graphene nano-flakes were inserted to dop the titania precursor solution for both the compact layer and the mesoporous, improving performance and stability. On the one hand, the introduction of graphene nano-flakes in c-TiO2 reduced the dark current, decreasing the series resistance and, therefore, improving the FF; on the other hand, both stationary and time-resolved photo-luminescence (PL) measurements revealed improved crystal morphology in graphene-doped titania, leading to improved VOC and improved electron extraction at the perovskite interface. / mTiO2 + G (A. Agresti et al., 2017; Antonio Agresti et al., 2017; Antonio Agresti, Pescetelli, Taheri, et al., 2016; Biccari et al., 2017; Bonaccorso et al., 2012; Taheri et al., 2018).

The addition of graphene has also improved the stability of the devices by preventing the diffusion of iodine from perovskite into the mesoporous layer of titania, delaying the degradation of the interface and increasing the life time of the devices (Busby et al., 2018).

A similar approach was followed by Cho et al. dispersing rGO in the crystal lattice of titania to improve the efficiency of the PSCs up to 19.54%. This strategy, combined with the treatment of mesoporous titania with lithium salts, improved the transport and injection of photo-excited charges (Cho et al., 2016).

In planar structures, the doping of tin oxide SnO2 was carried out by Zhao et al. incorporating graphene functionalized with naphthalene diimide in SnO2 to increase the hydrophobicity of the surface and form van der Waals interactions between surfactants and perovskite. This strategy made it possible to achieve efficiencies above 20% (Zhao Xiaojuan et al., 2018).

Regarding the doping of perovskite with 2D materials, nitrogen-doped rGO (N-rGO) was incorporated into the solution of the precursors of perovskite, which was then deposited by spin-coating on mesoporous titania. The authors of the research attributed the increase in performance (in terms of Jsc and FF) to the different morphology characterized by larger grains (and therefore reduced size of the grain boundaries) and to the greater thickness which led to a better absorption of radiation ( Hadadian et al., 2016).

The use of quantum-dot graphene (GQD) within the perovskite layer has been reported as a strategy for passivating the grain boundaries of perovskite, improving overall performance (Fang et al., 2017).

Graphene-related materials (GRMs) were introduced into the structure to improve charge collection at the counter electrode. In particular, GO has a work function of 4.95 eV that interfaces well with the HOMO (Highest Occupied Molecular Orbital) energy level of the perovskite (-5.4 eV) and the spiro-OMeTAD (-5.2 eV), allowing a good extraction of the charges positive (lacunae) at the perovskite / HSL interface (selective lacunae extraction layer). It can, therefore, be used both as an intermediate layer and as a dopant for HSL.

In the first case it has been shown that the application of GO as an intermediate layer between perovskite and HSL increases the performance of perovskite devices thanks to the passivation of the trap states at the interface through the Pb-O bonds that bind the GO to the perovskite . In addition, the presence of GO increases the wettability of the perovskite surface by favoring the adhesion of the spiro-OMeTAD through p-p interactions (Li et al., 2014).

The doping of HSL by p-type rGO was carried out with an acid solution based on iron iodide to replace the dopants commonly used in spiro-OMeTAD. These salts are, in fact, responsible for the rapid degradation of the active perovskite layer, shortening the life time of the devices in operating conditions. The devices obtained showed an efficiency decrease of less than 15% after 500 hours, compared to the decrease of 65% of the reference devices (Umeyama et al., 2015).

In addition to graphene and derivatives, other 2D structure materials based on dicalcogenide transition metals (TMDs) have been proposed as interlayer between perovskite and HSL (Ahn et al., 2016; Dasgupta, Chatterjee, & Pal, 2017; Huang et al. , 2017; Kim et al., 2016). Among these, Agresti et al. and Capasso et al. have used molybdenum disulfide (MoS2) as an intermediate layer for efficient extraction of gaps, leading to an increase in the efficiency and stability of the devices. The lowering of the VOC was compensated by the increase in the Jsc of the devices based on MoS2 (A. Agresti et al., 2017; Capasso et al., 2016).

The use of 2D materials in the front cell of a 2T tandem that includes perovskite as an absorber only concerns the use of graphene as an electrode of the PSC, as for example in the patent KR20190010197A.

In general, therefore, the use of graphene and GRMs has been widely addressed in the literature of the sector, but there is no systematic use of these materials to increase the FF and the VOC of the perovskite cells used as frontal cells of the tandem configuration. 2T, without affecting the transmittance of the same.

6. BRIEF DESCRIPTION OF THE INVENTION

The present invention allows to increase the performance of a multi-junction photovoltaic device by improving the Fill Factor (FF) and / or the open circuit voltage (VOC) through the use of graphene or other two-dimensional (2D) materials in one or more than the sub-cells which comprise a material with perovskite structure as an absorber of electromagnetic radiation. This strategy has the advantage of not modifying the transmittance of the aforementioned sub-cells to electromagnetic radiation and, therefore, does not alter the output current to the multi-junction device optimized according to the current "matching" condition in each sub-cell. .

The invention relates to the use of 2D materials such as graphene, graphene oxide, molybdenum disulfide and other related materials, as intermediate layers or as dopants in one or more layers constituting the device, with the aim of improving the charge transfer between the layers that make up the perovskite sub-cell, without altering its optical transmittance.

The objective is therefore achieved without introducing losses of electromagnetic radiation by reflection or parasitic absorption for wavelengths between 300 and 1200 nm.

Specifically, the proposed engineering concerns the interfaces of the material with perovskite structure with the selective layers of the electric charge (SL) by inserting further intermediate layers and / or doping the pre-existing SL and / or the active layer of perovskite, using graphene and / or related 2D materials (GRM).

The action on the GRM-based perovskite / SL interface is proposed as a strategy to increase FF and VOC by improving the charge transfer between the layers of the perovskite sub-cell (s). Specifically, graphene and / or other 2D materials are used as dopants in the layers constituting the device and / or as intermediate layers at the perovskite / SL interface. The 2D materials that can be used are for example graphene in its exfoliated or reduced form obtained from graphene oxide (GO), which in turn can be in the simple or functionalized form; other possibilities can be 2D materials such as molybdenum disulfide (MoS2) or tungsten disulfide (WS2); other materials other than those mentioned or even a combination of these can be used.

GRM production techniques can range from chemical exfoliation or be of the mechanical type starting from graphite in the liquid phase, but are not limited to those mentioned.

The proposed solution can be applied to any structure of perovskite devices: planar or mesoporous, direct (n-i-p) or inverted (p-i-n), on rigid or flexible substrates.

Unlike the pre-existing solutions, the GRM engineering proposed here constitutes a technological advancement to the extent that the improvement of the efficiency of the multi-junction PV device that includes perovskite absorbers passes through the increase of the FF and the VOC of the device while maintaining the structure previously optimized from the optical point of view and preserving this optimization in order to maintain the condition of the current “matching” in the 2T devices, with the possibility of reducing any hysteresis or instability phenomena in operating conditions.

The invention described here is easy to integrate into the production process of the devices and can be applied on a large scale and large area.

7. DETAILED DESCRIPTION

The invention proposed below allows to increase the electrical performance of a multi-junction photovoltaic device that includes a perovskite sub-cell, improving its Fill Factor (FF) and VOC, without limiting the transmittance of the various layers.

This purpose is achieved by a perovskite photovoltaic device by inserting graphene and / or other materials with two-dimensional (2D) structure into the different layers as dopants, or between them as intermediate layers.

The present invention is linked to the engineering of the device, acting on the interfaces between perovskite and the selective charge layers by inserting further interface layers and / or by doping the selective layers (SL) and / or the active layer itself. . The 2D materials that can be used are for example graphene in its exfoliated or reduced form obtained from graphene oxide (GO), which in turn can be in the simple or functionalized form; other possibilities can be 2D materials such as molybdenum disulfide (MoS2) or tungsten disulfide (WS2); other materials other than those mentioned or even a combination of these can be used. In the following, these materials will be referred to as GRM (Graphene Related Materials - materials similar to graphene) or more simply as 2D materials.

Fig. 2 shows a schematic and non-scale representation of a perovskite solar cell, to facilitate understanding of the structure, function and position of the various layers which will be referred to later. The structure in question is valid both for describing a single junction perovskite device and for a sub-cell of a multi-junction solar cell: the main difference will be dictated by the conductive substrate (6), which in the second case will consist of the others sub-cells constituting the multi-junction device. The dashed outlines delimit the optional layers in which the GRMs are inserted as intermediate or protective layers, while the textured fill indicates which layers can be doped by GRM.

The conductive substrate (6) in a single junction device can be any metal plate of any thickness, or silicon, glass, PET, PEN (and more) covered with a conductive material (tin doped indium oxide (ITO), fluorine doped tin oxide (FTO), aluminum doped zinc oxide (AZO), indium doped zinc oxide (IZO), copper nanowires (CuNW), silver nanowires (AgNW) etc. or a combination of these) that improves the electrical properties in order to have a resistance to the lateral transport of the charge (sheet resistance) ranging from 0 to 10 <3> Ω / □.

In a multi-junction device, the conductive substrate consists of the underlying sub-cells that make up the device and terminated with a conductive material (ITO, FTO, AZO, IZO, AgNW, CuNW etc. or a combination of these), or a junction tunnel (p-n or n-p), and which has a resistance to the vertical transport of the charge ranging from 0 to 10 <2> Ω and a thickness between 0 and 900 nm in order to minimize the optical losses by reflection or by absorption and electrical losses due to the Joule effect.

The charge-selective layer (7) can be a hole-selective layer (HSL), or an electron-selective layer (ESL), according to the type of structure chosen, respectively inverted or directed.

The materials that can be used for an HSL can be inorganic semiconductors such as copper thiocyanate (CuSCN), copper iodide (CuI), nickel oxide (NiO) and others, or organic materials such as Poly [bis (4-phenyl ) (2,4,6-trimethylphenyl) amine (PTAA), Poly (N, N'-bis-4-butylphenyl-N, N'-bisphenyl) benzidine (PolyTPD), 2,2 ', 7,7'- Tetrakis [N, N-di (4-methoxyphenyl) amino] -9,9'-spirobifluorene (spiro-OMeTAD or spiro-MeOTAD) and others.

The materials that can be used for ESL can be inorganic semiconductors such as titanium oxide (TiO2), tin oxide (SnO2), aluminum oxide (Al2O3) and others, or organic materials such as fullerenes (C60, PCBM and others), PEIE and others.

The absorber material, or active layer (9) is a material with a perovskite crystal structure and chemical formula ABX3 in which: A is a cation such as Methylammonium (MA), Formamidinium (FA), Cesium (Cs), Rubidium (Rb), Potassium (K) or other monovalent cations, or a combination thereof; site B is a divalent metal such as Lead (Pb), Tin (Sn) or others, or a combination of these; while C is an anion, more commonly a halogen such as Iodine (I), Bromine (Br), Chlorine (Cl), but it is not limited to these and can be a combination of them.

The other selective charge layer (11) will have opposite polarity with respect to the other (8): it will be an HSL in a direct structure (n-i-p) or an ESL in an inverted structure (p-i-n).

The materials used are similar to those previously mentioned, but are not limited to them.

The counter electrode (13) can be either a metal such as gold (Au), silver (Ag), copper (Cu) and others, or a transparent conductive oxide such as tin doped indium oxide (ITO ), fluorine doped tin oxide (FTO), aluminum doped zinc oxide (AZO), indium doped zinc oxide (IZO) or others.

An optional layer of graphene or other GRMs or any 2D material can be added as an intermediate layer (8) and / or (10) between the active layer and one of the filler selective layers (7) and / or (11). At the same time, or alternatively, GRMs could be used as protective layers (12) between a selective layer (11) and the counter electrode (13). At the same time, or alternatively, the 2D materials can be used as dopants or additives to the filler selective layers (7) and / or (11) and / or in the active layer (9). The doping or addition can take place in one or more layers (7), (9) and (11) at the same time and with different GRMs or 2D materials.

The different layers of the perovskite device can be deposited with different techniques depending on the architecture adopted:

In a direct mesoporous structure, the first selective charge layer (7) is an ESL, with a thickness ranging from 10 to 400 nanometers, commonly made with a compact layer of TiO2, TiOx, ZnO, Cs2CO3, or others, deposited with techniques different, such as screen printing, spray pyrolysis, sputtering, electroplating, anodizing, CVD, thermal or electron beam evaporation, pulsed laser deposition and others.

A mesoporous layer follows (hence the name of mesoporous structure) generally with a thickness in the range from 50 nm to 500 nm obtained by depositing a colloidal paste containing nanoparticles (for example of Al2O3, TiO2, ZrO2, SiO2 or others), solvents and binders organic. Generally, it is deposited with various techniques (screen printing, blade-coating, spin coating, slot-die coating, ink-jet and more).

The mesoporous layer is subsequently sintered with a thermal process with temperatures between 50 ° C and 600 ° C for a time between 10 and 3 00 minutes, necessary to evaporate the organic binders and to create the electromechanical bonds between the nanoparticles.

Alternatively, for plastic substrates, the mesoporous layer is made using a low temperature process, such as UV sintering with a radiation source comprising wavelengths between 200 nm and 350 nm, or with laser sintering.

Subsequently, this ESL is wetted by a perovskite solution containing its precursors, i.e. halogenated salts of one or more organic cations such as MA and / or FA and / or Cs or others and one or more inorganic compounds such as PbI2, PbCl2, PbBr2 , but not only.

This solution is deposited with various techniques, such as spin-coating, spray, physical vapor deposition, thermal evaporation, immersion, precipitation with anti-solvent and others).

The final crystalline perovskite is obtained by one or more thermal passages at a temperature generally between 50 ° C and 200 ° C for a time between 5 minutes and 120 minutes.

The thermal crystallization process could also be preceded by, simultaneous with, followed by, or alternative to an auxiliary step such as vacuum steps or compressed gas assisted crystallization

The processes of formation and crystallization of perovskite can also be different from these (for example the thermal evaporation of the precursors).

On the active layer (9), consisting of perovskite, a further selective charge layer (11) of opposite polarity with respect to the previous one is deposited (therefore an HSL in this case), consisting of P3HT, Spiro-OMETAD, PTAA, PEDOT: PSS , MoS2, CuSCN or others. This is deposited for spin coating, spray, inkjet, slot die and others.

The next layer (13) constitutes the counter-electrode of the photovoltaic cell and can be a conductive transparent oxide (ITO, FTO, AZO, IZO or other TCO films), or a metal, and can be deposited with any technique among those previously cited.

Alternatively, the direct planar cell is typically composed of an electron selective layer (ESL) which is generally deposited with a thickness in the range of 10 to 400 nanometers. It usually consists of a compact layer of TiO2, SnO2, ZnO2, PCBM and others, by means of techniques, such as screen printing, spray-pyrolysis, spin-coating, sputtering, electroplating, anodizing, CVD, Thermal or electron beam evaporation, LPD or other.

Subsequently, this ESL is wetted by a perovskite solution containing its precursors, i.e. halogenated salts of one or more organic cations such as MA and / or FA and / or Cs or others and one or more inorganic compounds such as PbI2, PbCl2, PbBr2 , but not only.

This solution is deposited with various techniques, such as spin-coating, spray, physical vapor deposition, thermal evaporation, immersion, precipitation with anti-solvent and others).

The final crystalline perovskite is obtained by one or more thermal passages at a temperature generally between 50 ° C and 200 ° C for a time between 5 minutes and 120 minutes.

The thermal crystallization process could also be preceded by, simultaneous with, followed by, or alternative to an auxiliary step such as vacuum steps or compressed gas assisted crystallization

The processes of formation and crystallization of perovskite can also be different from these (for example the thermal evaporation of the precursors).

On the active layer (9), consisting of perovskite, a further selective charge layer (11) of opposite polarity with respect to the previous one is deposited (therefore an HSL in this case), consisting of P3HT, Spiro-OMETAD, PTAA, PEDOT: PSS , MoS2, CuSCN or others. This is deposited by spin coating, spray, inkjet, slot die, atomic layer deposition (ALD) and others.

The next layer (13) constitutes the counter-electrode of the photovoltaic cell and can be a conductive transparent oxide (ITO, FTO, AZO, IZO or other TCO films), or a metal, and can be deposited with any technique among those previously cited.

The inverted planar perovskite solar cell has a structure where the hole selective layer (HSL) is first deposited on the substrate (7). The cell is generally composed of an HSL consisting of different materials such as PEDOT: PSS, Cu: NiOx, PolyTPD, VB-DAAF and others and deposited with different techniques (with spin coating, spray, ink jet, slot die coating and other).

Subsequently, this ESL is wetted by a perovskite solution containing its precursors, i.e. halogenated salts of one or more organic cations such as MA and / or FA and / or Cs or others and one or more inorganic compounds such as PbI2, PbCl2, PbBr2 , but not only.

This solution is deposited with various techniques, such as spin-coating, spray, physical vapor deposition, thermal evaporation, immersion, precipitation with anti-solvent and others).

The final crystalline perovskite is obtained by one or more thermal passages at a temperature generally between 50 ° C and 200 ° C for a time between 5 minutes and 120 minutes.

The thermal crystallization process could also be preceded by, simultaneous with, followed by, or alternative to an auxiliary step such as vacuum steps or compressed gas assisted crystallization

The processes of formation and crystallization of perovskite can also be different from these (for example the thermal evaporation of the precursors).

On the active layer (9), consisting of perovskite, a further selective charge layer (11) of opposite polarity with respect to the previous one is deposited (therefore a HSL in this case), consisting of PCBM, C60, BCP, PCBC fullerene, ZnO, but not only. This is deposited by spin coating, spray, inkjet, slot die, thermal evaporation, atomic layer deposition (ALD) and others.

The next layer (13) constitutes the counter-electrode of the photovoltaic cell and can be a conductive transparent oxide (ITO, FTO, AZO, IZO or other TCO films), or a metal, and can be deposited with any technique among those previously cited.

Numerous variations can be made to these basic structures with the insertion of graphene or GRM, such as graphene oxide (GO), reduced graphene oxide (rGO), or functionalized with lithium salts (GO-Li), or other 2D materials such as molybdenum disulfide (MoS2), or tungsten (WS2) and others. These can be inserted into the various layers that make up the device, such as the mesoporous layer, the active layer of perovskite, the HSL and the ESL, but not only, both as a dopant for the various layers, and as an additional or replacement intermediate layer.

The doping can be carried out by adding to the precursor of the layer to be doped a percentage by weight or by volume (between 0% and 15%) of the solution containing the GRMs.

The insertion of GRM as intermediate layers, on the other hand, can take place, depending on the desired structure, directly depositing them starting from the solution containing the 2D materials by means of spin-coating, spray, blade-coating, inkjet or others, or through techniques such as CVD, evaporation, ALD or others.

8. EXAMPLES

As an example, Figure 3 shows the use of graphene in order to increase the performance of a mechanically coupled 2T perovskite / silicon tandem solar cell. In order to increase the performance of the silicon cell, the tandem cell is obtained by coupling a semitransparent perovskite cell with a heterojunction silicon cell (22).

The strategy adopted is to combine the advantages of other conventional architectures, such as: the high voltage of the monolithic tandem given by the sum of the voltages obtained by the two cells; the possibility of manufacturing the two cells independently as happens for a 4T tandem while respecting process compatibility. In this example, the structure chosen for the perovskite top-cell was the mesoscopic one as it represents the most performing and most used structure in literature.

The structure of the perovskite device is shown in Figure 3 and consists of a glass substrate (14) where a layer of fluorine-doped tin oxide is used as a transparent photoelectrode (15). The electron selective layer consists of a compact titanium dioxide (TiO2) layer (16) and a mesoporous TiO2 layer (17). These two layers (16) and (17) were doped with graphene in order to demonstrate the effectiveness of the proposed strategy. The active layer (18) is a perovskite (CsFAMA) Pb (IBr) 3 based on mixed cations and anions. The lacunae selective layer (19) is the doped PTAA polymer. The rear contact (20) of the perovskite cell consists of a layer of ITO deposited by RF-sputtering. The bottom cell is a heterojunction silicon cell with a silver metal grid (21). The series connection is obtained by pressing the rear electrode (20) of the PSC cell on the grid (21) of the heterojunction silicon bottom cell (22). The resulting structure of the tandem device represents an easy approach for coupling between a solution-processed perovskite device and high-performance silicon cells provided with surface texturing, overcoming the previously mentioned limitations.

In this specific example, the perovskite top cell features a graphene-based photoelectrode (PE) obtained by doping both compact (c-TiO2) and mesoporous (m-TiO2) TiO2 layers with a nano-flake ink of graphene capable of increasing the stability and performance of the PSC cell.

It has been shown that the insertion of graphene flakes in the c-TiO2 solution reduces the dark current of the cell leading to the reduction of the series resistance and possibly to an increase in the FF of the device.

The results obtained by Steady state and PL time resolved photoluminescence (TRPL) reveal a better morphology of the perovskite crystals when it is grown inside the graphene-doped m-TiO2 layer (mTiO2 + G) leading to i) an increase in voltage a open circuit (VOC); ii) a better injection of electrons at the mTiO2 + G / perovskite interface.

The stability of the device is also increased thanks to the addition of graphene demonstrating to prevent the diffusion of iodine in the m-TiO2 layer by delaying the degradation of the mTiO2 / perovskite interface and finally increasing the life of the device.

The manufacture of the semi-transparent perovskite devices was carried out according to the following procedure:

Glass / FTO substrates with 12 Ω / □ resistance purchased from Pilkington undergo a cleaning process by successive sonication steps in acetone, ethanol and isopropanol for 10, 5 and 10 minutes respectively, in order to remove any organic residue from the surface .

The silver contacts are screen printed directly on the FTO to form the n-type contacts for the cell while preserving the anode from subsequent layer deposition.

The compact layer of TiO2 (c-TiO2) is deposited by spray-pyrolysis at 450 ° C starting from a precursor solution consisting of titanium diisopropoxide bis (acetylacetonate), acetylacetone and ethanol in a volume ratio 3: 2: 45. Optimizing the number of spray cycles was key to ensuring a 40nm thick layer of c-TiO2. In the case of the graphene doped c-TiO2 layer, an adequate amount of graphene ink was added to the TiO2 precursor solution optimizing the resulting PSC cell in terms of PCE.

The mesoporous layer for electron transport (m-TiO2) is deposited by spin coating at 3000 rpm for 15 seconds using a paste of titania Dyesol 30 NRD nanoparticles diluted with ethanol in a 1: 5 mass ratio. The solution thus prepared was doped by adding 1% by volume of ink based on graphene nanoflakes and subsequently mixed for two hours.

The preparation of the perovskite precursor solution consists of proportionally weighing PbI2, FAI, PbBr2, MABr and CsI in a single vial, followed by the addition of a mixture of DMF: DMSO solvents (4: 1), in order to obtain a solution at a concentration of 1.2 M.

The deposition was carried out in a nitrogen glove-box by spin coating the said solution with a program consisting of two phases at 1000 rpm and 6000 rpm for 10 and 20 seconds respectively. At 5 seconds before the end of the second phase, the chlorobenzene, used as an anti-solvent, is deposited on the sample.

The hole selective layer (PTAA) was prepared by dissolving 10 mg of material in 1 ml of toluene and deposited by spin coating at 3000 rpm for 20 seconds.

The silicon heterojunction solar cell is based on a high quality n-type 1-5 ohm * cm Si substrate, cleaned by standard RCA procedure and textured with a treatment called KOH wet processing.

The amorphous silicon layers were deposited through PECVD RF reactors. The intrinsic a-Si: H-based thin-film passivation layers were then deposited on both sides of the silicon wafer, while the hole and electron collectors were formed from p-type a-Si: H thin films - and n- by adding respectively PH3 and B2H6 as precursor gases to SiH4. The TCO films were deposited on doped a-Si: H layers followed by the deposition of metal grids at low temperature (<200 ° C) by screen printing.

The following table shows the best device performance with PCE of 26.3% over an active area of 1.43 cm <2>. The result shows an improvement compared to the reference cell without the use of any 2D material in the layers which has an efficiency of 24.2% for the best device (Figure 4).

A marked improvement in VOC and FF can be observed for the graphene-doped device, while the current is not affected at all by the doping strategy.

Claims (4)

CLAIMS The present invention, in specific application to multi-junction photovoltaic devices consisting of one or more absorbers with perovskite crystalline structure, consists of the following claims in order to increase the output voltage and Fill Factor of the aforementioned devices: 1) A doping process for one or more selective charge layers of the perovskite solar cell constituting a tandem photovoltaic device using 2D materials (graphene, GRM or others) which consists of the following steps: to. dispersion of the 2D material or materials in a solvent; b. dissolution of the precursor of the material constituting the selective filler layer by means of a solvent in which the solubility limit of the same is greater than 1 mg / mL, with subsequent homogenization of the solution by mechanical stirring for a time greater than 1 minute; c. the addition in solution of materials with 2D structure from the dispersion obtained in point a. in a proportion between 0.1% and 50% by weight or by volume and further homogenization for a time exceeding 1 minute; d. deposition of the resulting solution to form the selective filler layer by printing or solution processing techniques; And. heat treatment of the film obtained with a temperature between 20 ° C and 500 ° C for a time between 1 minute and 3 hours. 2) The process described in claim 1) which provides in step b. the use of a dispersion in a solvent of nanoparticles or nanocrystals or nano-flakes of the material constituting the selective filler layer. 3) The process described in claim 1) or 2) which involves adding the 2D materials in solution or dispersion directly to step b. 4) A procedure according to what is described in the preceding claims and which provides for the addition of an optical UV treatment, visible light and / or IR, with an emission peak at a wavelength between 150 nm and 1 mm, subsequently, previously or in place of the heat treatment of step e. of claim 1); 5) The deposition of a passivation or protection layer consisting of graphene or GRM or any 2D material such as layer (8) and / or (10) and / or (12) with reference to the drawing in figure 2 according to the following procedure: to. Dispersion of the 2D material or materials in a solvent; b. Deposition by printing or process techniques from solution of the dispersion in point a .; c. Heat treatment at a temperature between 20 ° C and 200 ° C of the obtained film.