WO2023026189A1

WO2023026189A1 - Phase stable perovskite film and method

Info

Publication number: WO2023026189A1
Application number: PCT/IB2022/057892
Authority: WO
Inventors: Erkan Aydin; Stefaan DE WOLF; Jiang Liu
Original assignee: King Abdullah University Of Science And Technology
Priority date: 2021-08-24
Filing date: 2022-08-23
Publication date: 2023-03-02

Abstract

A tandem solar cell (100) includes a first-type base (102) having first and second surfaces (102A, 102B), opposite to each other, a second-type layer (108) formed on the second surface (102B) of the first-type base (102), wherein the first-type is one of n- or p-type and the second-type is another of the n- or p-type, and the second-type layer (108) and the first-type base (102) form a first pn junction, and an additive-treated perovskite layer (124) formed on the first surface (102A) of the first-type base (102), wherein the additive-treated perovskite layer (124) and the first-type base (102) form a second pn junction. The additive-treated perovskite layer (124) has an organic additive (310) that includes carbon, hydrogen and nitrogen, and is distributed at borders between grains (Gi) of the perovskite layer. The additive-treated perovskite layer (124) includes a perovskite material that has a band gap equal to or larger than 1.65 eV.

Description

PHASE STABLE PEROVSKITE FILM AND METHOD

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Provisional Patent Application No. 63/236,414, filed on August 24, 2021 , entitled “PREPARATION OF ADDITIVE- ASSISTED PEROVSKITE FILMS FOR PHASE-STABLE DEVICES,” the disclosure of which is incorporated herein by reference in its entirety.

BACKGROUND

TECHNICAL FIELD

[0002] Embodiments of the subject matter disclosed herein generally relate to a device and method for increasing the phase stability of a perovskite based device, and more particularly, to adding carbazole as an additive to a perovskite film in an electronic device.

DISCUSSION OF THE BACKGROUND

[0003] Over the past decade, metal halide perovskites have triggered intense research activities in the photovoltaics (PVs) field due to their advantages. Their band gap tunability, processing flexibility, and excellent optoelectronic properties make them an ideal sub-cell candidate for tandem applications, specifically when paired with mainstream crystalline silicon (c-Si) PVs. Owing to their high absorption coefficient and low voltage loss, perovskite top cells can reduce the thermalization losses in the blue portion of the solar spectrum, relative to a narrower bandgap single junction c-Si cell, and enable power conversion efficiencies (PCEs) of tandem solar cells beyond the single-junction limits.

[0004] High-efficiency perovskite/silicon tandem devices can be realized by tuning the perovskite top-cell band gap (Eg) to a value of 1 .65-1 .70 eV, where the ideal value may vary considering the field operation temperature and possible parasitic optical losses in the device layer stack. Utilizing such perovskites as top-cell absorber, paired with silicon heterojunction (SHJ) bottom cells, perovskite/silicon tandem solar cells have been realized with PCE up to 29.5%, which is already well beyond the best-reported single-junction silicon heterojunction bottom cells.

[0005] To achieve a high-efficiency of such devices with long-term operational stability, implementing a phase-stable, wide-bandgap perovskites is required.

However, to date, wide-bandgap metal halide perovskites (/.e., where E_g >1 .65 eV) have been associated with considerable device phase-stability issues. Many efforts such as via compositional engineering, crystallinity control, and defect passivation have been devoted in the past few years to not only suppress the phase segregation issue of wide-bandgap perovskites, but also reduce the nonradiative recombination loss. For instance, increasing the Cs and reducing the Br concentrations, or adopting triple-halides in the perovskite for a target bandgap have been demonstrated as effective strategies to stabilize the perovskite layer under illumination. However, a high Cs concentration in the perovskite precursor solution usually results in a poor film morphology and deteriorated electronic properties, possibly due to induced stresses in the resulting films.

[0006] Several earlier reports show that improved crystallinity and reduced trap densities render perovskite films to be more stable against halide segregation under illumination. In addition, grain boundaries (GBs) accommodate a significant amount of charged defects due to the ionic nature of the perovskite film. The passivation of trap states via some specific additives inside the precursor, such as theophylline [1], imidazolium (C3H4N2) [2], piperidinium (C5H11N) [3] and some amine salts [4, 5], has been reported as an effective strategy to circumvent this issue. In these approaches, molecular passivation deactivates some defects via coordinate binding, and hence, reduces non-radiative recombination and improves photostability.

[0007] However, the existing methods do not address the long-term lack of stability for the wide-bandgap perovskites. Thus, there is a need for a new method and system that is capable of overcoming these problems and providing the necessary long-term phase stability.

BRIEF SUMMARY OF THE INVENTION

[0008] According to an embodiment, there is a tandem solar cell that includes a first-type base having first and second surfaces, opposite to each other, a second- type layer formed on the second surface of the first-type base, wherein the first-type is one of n- or p-type and the second-type is another of the n- or p-type, and the second-type layer and the first-type base form a first pn junction, and an additive- treated perovskite layer formed on the first surface of the first-type base, wherein the additive-treated perovskite layer and the first-type base form a second pn junction. The additive-treated perovskite layer has an organic additive that includes carbon, hydrogen and nitrogen, and is distributed at borders between grains (Gi) of the perovskite layer. The additive-treated perovskite layer includes a perovskite material that has a band gap equal to or larger than 1 .65 eV.

[0009] According to another embodiment, there is a single-junction device that includes a semiconductor base and an additive-treated perovskite layer formed on a surface of the semiconductor base, wherein the additive-treated perovskite layer and the semiconductor base form a pn junction. The additive-treated perovskite layer has an organic additive that includes carbon, hydrogen and nitrogen, and is distributed at borders between grains (Gi) of the perovskite layer. The additive-treated perovskite layer includes a perovskite material that has a band gap equal to or larger than 1 .65 eV.

[0010] According to yet another embodiment, there is a method for making a perovskite-based device, and the method includes providing a semiconductor substrate, mixing perovskite precursors with an organic additive to form a perovskite mixture, applying the perovskite mixture to the semiconductor substrate, and annealing the perovskite mixture to form an additive-treated perovskite layer. A mass of the organic additive is equal to or less than 1% of a total mass of the additive- treated perovskite layer. The organic additive includes carbon, hydrogen and nitrogen, which are distributed at borders between grains (Gi) of the perovskite layer, and the additive-treated perovskite layer includes a perovskite material that has a band gap equal to or larger than 1 .65 eV.

BRIEF DESCRIPTION OF THE DRAWINGS

[0011] For a more complete understanding of the present invention, reference is now made to the following descriptions taken in conjunction with the accompanying drawings, in which:

[0012] Figure 1 is a schematic diagram of a tandem solar cell having a carbazole-treated perovskite layer;

[0013] Figures 2A and 2B illustrate a flow chart of a method for manufacturing the tandem solar cell having the carbazole-treated perovskite layer;

[0014] Figure 3A illustrates perovskite precursors being mixed with carbazole for obtaining the carbazole-treated perovskite layer and Figure 3B illustrates the location of the carbazole, between the various grains of the perovskite material;

[0015] Figures 4A and 4B illustrate the evolution of time-dependent photoluminescence spectra under 10 sun-equivalent laser illumination for a control solar cell and an encapsulated carbazole-treated perovskite layer;

[0016] Figure 5 illustrates the evolution of steady-state PL spectra for the unencapsulated pristine film and carbazole-treated perovskite layer upon exposure to air;

[0017] Figure 6 illustrates the temperature-dependent conductivity of the pristine film and carbazole-treated perovskite layer;

[0018] Figure 7 illustrates Fourier transform infrared spectroscopy of pure carbazole and carbazole-Pbl2 films; [0019] Figure 8A illustrates time-resolved photoluminescence (TRPL) spectra of the wide-bandgap (1 .68 eV) perovskite films with or without carbazole on 2PACz- coated ITO glass substrate;

[0020] Figure 8B illustrates the electric carriers’ mobility as a function of excitation carrier density, extracted from Terahertz spectroscopy;

[0021] Figure 9 illustrates a single-junction semiconductor device having a carbazole-treated perovskite layer;

[0022] Figure 10A illustrates J-V curves of a 2PACz-based perovskite device with or without carbazole treatment;

[0023] Figure 10B is a table that illustrates the photovoltaic parameters of the devices considered with regard to Figure 10A;

[0024] Figure 10C illustrates the open circuit voltage V_oc evolution as a function of light intensities for the control and carbazole-treated perovskite devices;

[0025] Figure 11 A illustrates J-V curves of the tandem cell of Figure 1 with an aperture area of 1 .03 cm²;

[0026] Figure 11 B illustrates the J-V curves of a 3.8 cm² aperture area tandem device similar to that of Figure 1 ;

[0027] Figure 12 illustrates the outdoor performance of the tandem solar cell device in a hot desert climate for 43 days. J-V scans under forward direction were performed approximately every 10 min during the daytime. 1 sun indicates 100 mW/cm² solar irradiance. The daily maximum values of Voc and Pm are connected by dotted lines in an interpolated manner; and [0028] Figure 13 is a flow chart of a method for making a semiconductor device that includes a pn-junction made with an additive-treated perovskite material.

DETAILED DESCRIPTION OF THE INVENTION

[0029] The following description of the embodiments refers to the accompanying drawings. The same reference numbers in different drawings identify the same or similar elements. The following detailed description does not limit the invention. Instead, the scope of the invention is defined by the appended claims. The following embodiments are discussed, for simplicity, with regard to a specific perovskite material (e.g., Cso.o5FAo.8MAo.i5Pb(lo.75Bro.25)3) combined with carbazole (C12H9N) so that the carbazole is located at the borders between the grains in the perovskite material. However, the embodiments to be discussed next are not limited to this specific perovskite, but may be applied to other perovskite materials. In one embodiment, it is possible to use another organic material instead of the carbazole. [0030] Reference throughout the specification to “one embodiment” or “an embodiment” means that a particular feature, structure or characteristic described in connection with an embodiment is included in at least one embodiment of the subject matter disclosed. Thus, the appearance of the phrases “in one embodiment” or “in an embodiment” in various places throughout the specification is not necessarily referring to the same embodiment. Further, the particular features, structures or characteristics may be combined in any suitable manner in one or more embodiments.

[0031] According to an embodiment, the carbazole is used as an additive for perovskite precursors, and this additive can passivate deep charge traps between the grains of the perovskite material, enhance performance, and suppress phase segregation of wide-bandgap perovskites. Note that some similar additive molecules with this kind of nitrogen-containing heterocyclic, such as Imidazole, Piperidine, etc., are also suitable for improving the stability of the perovskite film. The organic, nitrogen-containing heterocyclic molecule interacts with halide anions of the perovskite through hydrogen bonds, immobilizes halide species and passivates trap states. Specifically, this happens at the grain boundaries since the carbazole molecules with a relatively large molecular radius cannot be incorporated into the perovskite lattice. Combining this enhancement with minimized optical losses, a solar cell (e.g., double-side textured perovskite/silicon tandems, to be discussed later) made based on this process shows a record-certified PCE of 28.2%. Equally advantageous, passivating the defects at the grain borders inhibits ion migration within the perovskite film and improves the long-term stability of the tandem device. The carbazole-treated perovskite layer within the tandem devices retain over 93% of their initial performance for more than 40 days when tested outdoors, which demonstrates the improved phase stability of the wide-bandgap perovskites.

[0032] A carbazole-treated tandem solar cell and a method for making it are now discussed in more detail with regard to the figures. Figure 1 shows the carbazole-treated tandem solar cell 100 having an n-doped float-zone (FZ) Si wafer 102 with a thickness of 260-280 pm, which was used for the Si bottom cell fabrication. The Si wafer 102 has a top surface 102A and a bottom surface 102B and was processed in step 200 (see Figures 2A and 2B) to have a double-side textured top and bottom surfaces with random distributed top and bottom pyramids 104A and 104B, which was obtained by using an alkaline solution. In other words, the top and bottom surfaces of the Si wafer 102 are not flat. Note that a flat sided Si wafer may be used instead of the textured one to obtain a tandem solar cell. The size of the pyramids 104A and 104B is controlled by adjusting the alkaline concentration and the process temperature. The wafer 102 was dipped into a hydrofluoric acid solution followed by a cleaning process, before being transferred into a plasma enhanced chemical vapor deposition (PECVD) cluster for amorphous silicon (a-Si) deposition. In step 202, a 8 nm intrinsic (i) layer 106 was grown on the top and bottom faces of the wafer 102, a 6 nm n-doped a-Si layer 110 was grown on the layer 106 on the top face, and a 13 nm p-doped a-Si layer 108 was grown on the layer 106 on the bottom face using the PECVD cluster tool. Then, in step 204, a 150 nm ITO layer 112 and a 250 nm Ag layer 114 were sputtered on the layer 108 on the backside of the wafer 102. A 15 nm ITO recombination layer 118 was sputtered on the layer 110 on the front side of the wafer 102 in step 206. In order to recover sputtering damage, an annealing step 208 was performed at 200 °C for 10 min. Then, a 2-PACz layer 120 was spin coated (as a hole transporting layer) onto the ITO layer 118 in step 212. The layers 102/106/108 form a first pn junction at the bottom surface of the device. While Figure 1 shows the base 102 to be n-type and the Si layer 108 to be p-type, in another embodiment is possible to reverse the type of semiconductor, e.g., layer 102 is p-type and layer 108 is n-type. In yet another embodiment, it is possible to user other semiconductor materials instead of Si.

[0033] The top pyramids 104A of the Si wafer 102 were subjected to the optional step 210, which takes place after step 208, to UV-Ozone treatment for 15 min before being transferred into the glovebox for further processing. One of two materials may be deposited above layer 118 to act as the hole transport layer 120, Poly[bis(4-phenyl)(2,4,6-trimethylphenyl)amine (PTAA) or [2-(9H-carbazol-9- yl)ethyl]phosphonic acid (2PACz). For the PTAA deposition, 2 mg/mL PTAA solution in anhydrous chlorobenzene (CB) was used. For 2PACz deposition, 1 mg/mL 2- PACz in ethanol was used. The PTAA or 2PACz material, which acts as the hole transport layer (HTL) 1120, was spin-coated in step 212 (other processes may be used for this step) on the ITO-coated substrate 102 at 5000 rpm for 50 s, followed by the step 208 of drying at 100 °C for 10 min.

[0034] A 1 .7 M Cso.o5FAo.8MAo.i5Pb(lo.75Bro.25)3 perovskite precursor solution was prepared in step 214 by dissolving a mixture of Formamidinium iodide (FAI), methylammonium bromide (MABr), Csl, Pbl2, and PbBr2 in a mixed solvent (e.g., anhydrous dimethylformamide (DMF)/ anhydrous dimethyl sulfoxide (DMSO) = 4:1). In step 216, a given amount of carbazole (equal to or less than 1% by weight of the total mass of the perovskite) was added into the perovskite solution to form a perovskite mixture. The perovskite mixture was then spin-coated in step 218 at 2000 rpm for 50 s on the hole transport layer 120, then followed with 7000 rpm for 10 s, to form the carbazole-treated perovskite layer 124, as illustrated in Figure 3A. Other methods may be used to form the carbazole-treated perovskite layer 124, for example, blade coating or any other solution-based method. Chlorobenzene may be dropped in the center of the substrate, 12 s before the end of the spin-coating process. After the rotation ceased, the substrate was immediately transferred in step 220 onto a hotplate of 100 °C and was annealed for 15 min to form the carbazole- treated perovskite layer 124. Note that the carbazole-treated perovskite layer 124 in its final form includes the carbazole substance/additive 310 concentrated at the boundaries 312 between its grain regions Gi, as schematically shown in Figure 3B, and not at the exterior surface 320 of the perovskite material. Also note that the layers 102/124 or 1 18/124 form a second pn junction on the top surface of the device. Because of the first and second pn junctions, the solar cell 100 is called a tandem solar cell.

[0035] After the perovskite deposition, 1 nm Li F 126 and 15 nm Ceo 128 were subsequently deposited by thermal evaporation in step 222. A 20 nm SnC>2 layer 130 was then deposited by atomic layer deposition (ALD) in step 224. The substrate 102 temperature was maintained at 100 °C during the ALD deposition with TDMASn precursor source at 80 °C and H2O source at 18 °C. A 70 nm indium zinc oxice (IZO) layer 132 was sputtered in step 226 on top of the SnC>2 layer 130 through a shadow mask. An Ag finger 134 with a thickness of about 350 nm was thermally evaporated in step 228 using a high precision shadow mask. Finally, a 120 nm MgF2 layer 136 was thermally evaporated in step 230 as an anti-reflection layer. This layer was formed directly on top of the IZO layer 132. The thicknesses of the LiF, Ceo, IZO and MgF2 layers can vary from the values given herein by up to 20%.

[0036] While the method discussed with regard to Figures 2A and 2B is directed to the tandem solar cell 100 shown in Figure 1 , a single-junction perovskite solar cell may also be used with the modified perovskite film 124. For this solar cell, an ITO glass is ultrasonically cleaned with detergent, deionized water, acetone, and isopropanol successively, and then blow-dried with compressed nitrogen. The substrate may be subjected to a UV-Ozone treatment for 15 min before any film deposition. For PTAA deposition, 2 mg/mL PTAA solution in CB was used. For 2PACz deposition, 1 mg/mL 2-PACz in ethanol was used. The PTAA or 2PACz as hole transport layer (HTL) was spin-coated on ITO-coated substrates at 5000 rpm for 50 s, followed by drying at 100 °C for 10 min. 1.7 M Cso.o5FAo.8MAo.i5Pb(lo.75Bro.25)3 perovskite precursor solution was prepared by dissolving a mixture of FAI, MABr, Csl, Pbl2, and PbBr2 in a mixed solvent (DMF/DMSO = 4:1 ). A certain amount of carbazole (less than 1% of the total mass of the mixture) was added into the perovskite solution. The perovskite films were spin-coated at 2000 rpm for 50 s then followed with 7000 rpm for 10 s. Chlorobenzene was dropped in the center of the substrates 12 s before the end of the spin-coating process. After the rotation ceased, the substrates were immediately transferred onto a hotplate of 100 °C and were annealed for 15 min. 1 nm LiF, 25 nm Ceo, 6 nm BCP and 100 nm Ag layers were thermally evaporated on perovskite films sequentially.

[0037] Various characteristics of the tandem solar cell 100 were measured as now discussed. Systematic photoluminescence (PL) measurements were performed to understand the role of the carbazole additive 310 on the halide segregation in the selected 1 .68 eV perovskite with the composition of Cso.o5FAo.8MAo.i5Pb(lo.75Bro.25)3. In one application, the selected perovskite material has a band gap equal to or larger than 1 .65 eV. The bandgap of the perovskite layer 124 was confirmed from a Tauc plot and the PL peak position. To exclude the influence of environmental factors such as humidity and oxygen, in one embodiment, the carbazole-treated perovskite layer 124 was encapsulated in a nitrogen glove box. Under 1 -sun equivalent illumination, no significant change in the PL peak position was observed within 15 min for both pristine and carbazole-treated perovskite layers. However, it is necessary to determine that these perovskite films will remain stable under long-term operation. To accelerate the degradation test, the light intensity was increased to 10 suns, as shown in Figures 4A and 4B. In this case, the pristine perovskite film (i.e. , no carbazole) shows an extra peak in its lower-energy region, which gradually increases with the irradiation time, indicating the formation of some light-induced I- rich trap states. Notably, the carbazole-treated layer (corresponding to Figure 4B) is much less affected, suggesting improved phase stability. Halide segregation of wide- bandgap perovskite films commonly occurs at grain boundaries, activated by strain. It is believed that carbazole molecules most probably stay at the grain boundaries due to their incompatible size to fit the perovskite lattice. Thus, it is believed that the introduction of carbazole additives to the perovskite precursor solution can hinder the migration of ions in the film, thereby suppressing halide segregation.

[0038] Next, the influence of ambient air on phase stability was investigated, and thus performed PL measurements on unencapsulated samples exposed to ambient air (RH -55%). As shown in Figure 5, exposure to humidity accelerates phase segregation, especially for the control (pristine) sample. When the pristine layer was exposed to ambient air under dark for about three days, the re-measured PL spectrum shows a significant red-shift, while the carbazole-treated layer was much less affected, implying enhanced phase-stability even under humid conditions. In addition, it was found that the synergistic effect of light and humidity will accelerate phase segregation. These results show that the illumination intensity, the environmental atmosphere and trap-passivation all alter the dynamics of halide segregation. In the case of superposition of multiple stressors, halide segregation occurs faster, but the addition of carbazole minimizes this behavior.

[0039] To gain insight into the activation energy (E_a) of ion migration in the perovskite layer, temperature-dependent conductivity measurements were carried out, as illustrated in Figure 6. The E_a for the pristine and carbazole-treated perovskite layers were calculated to be 0.38 and 0.77 eV in the activation region (>260 K), respectively. Theoretical calculations have predicted that the halide ions are the most mobile ions in the perovskite layers since they have the lowest activation energy. Thus, the measured E_a here can mainly be linked to that for halide migration. The obvious increase in E_a after carbazole treatment offers quantitative evidence for additive-suppressed ion migration, which may be attributed to the bonding effect of carbazole at the GBs.

[0040] Conductive atomic force microscopy (c-AFM) measurements were also performed to study the conductivity difference between the grains and the GBs, potentially induced by the carbazole treatment. The topographic and current maps collected for both samples indicate that the GBs have different conductivity compared to the grains. For the control sample, the GB region exhibits a higher dark conductivity than the grain surface, indicating a higher ion migration activity at the GBs, which is also consistent with previous results that GBs can act as a channel for ion migration. However, for the carbazole-treated layer 124, most of the GB regions show a lower current compared to their surrounding grain region, which could be attributed to the suppression of ion migration by the carbazole 310 at the GBs 312. [0041] To investigate the interaction between the carbazole and the perovskite materials, Fourier transform infrared (FTIR) spectroscopy was performed. In Figure 7, the absorption peaks of the pure carbazole material at around 3416, 1597 and 1449 cm'¹ could be attributed to N-H stretching, N-H related bending, and C-N stretching frequencies, respectively. For the carbazole-Pbl2 material, the C-N stretching feature did not show any major shift in frequency, while the N-H stretching and N-H related bending vibrations shifted to 3410 and 1602 cm'¹, respectively, indicating that carbazole could interact with the perovskite through N-H bonding.

[0042] In addition, density functional theory (DFT) calculations were performed on the carbazole/perovskite interfacial models to further understand the role of carbazole on passivating the perovskite surface, as well as the atomistic interaction between them. Based on the X-ray diffraction (XRD) patterns, three representative (100), (101 ) and (111 ) perovskite surfaces with different terminations were considered and estimated the binding affinities of carbazole with these surfaces (all the optimized perovskite slab structures before and after carbazole adsorption). It was found that the carbazole molecule is inclined to bind with surface I- ions via hydrogen bonding (/.e., N-H- ■■ I, 2.73 to 2.92 A), where such interactions result in binding energies (Eb) in the range of -0.4 to -0.7 eV for FAI-rich (100) and l-rich (1 11 ) surfaces, respectively. For the l-rich (101 ) surface, a shorter hydrogen bond length (2.60 A) as well as a larger Eb value (-2.26 eV) for the carbazole adsorption is found, suggesting that the carbazole molecules can stabilize the perovskite surface through hydrogen bonding interactions with halide ions, preventing halide ion migration from the surface into the bulk. [0043] To demonstrate the passivation effect of this additive material, time- resolved photoluminescence (TRPL) measurements were carried out as illustrated in Figure 8A. The perovskite films with carbazole additive on both PTAA and 2PACz- based substrates exhibit a prolonged average PL decay time T_ave, which improved from 910 to 1403 ns in the case of 2PACz. These results indicate that nonradiative recombination is significantly suppressed upon the addition of carbazole, which is expected to benefit the open-circuit voltage ( V_oc) and fill factor (FF) of the tandem device. The electronic properties of the perovskite film 124 were further investigated upon carbazole additive via contactless Terahertz (THz) spectroscopy, as illustrated in Figure 8B. The extracted mobility generally decreases with increasing carrier density. At the lowest photoinduced carrier densities (~1 .6x10¹⁸ cm^-3), the pristine film shows a charge mobility of around 22.1 cm² V'¹ s’¹; this value increases to 25.8 for 0.25 mol%, 24.6 cm² V'¹ s^-1 for 1 mol% sample. Combining the charge lifetime rave from TRPL results, it is possible to calculate the charge diffusion length L to be 7.19 and 9.64 pm for the pristine and passivated films, respectively, which are noticeably longer than the thickness of the absorber in the devices. The improved carrier diffusion length in the perovskite film, which is the result of largely suppressed trap density, can benefit charge collection.

[0044] In one embodiment, the role of the carbazole additive concentration on the performance of a single-junction perovskite device was determined. For this, a p- i-n device 900 (see Figure 9) with glass 902, ITO 904, 2PACz 906, 1 .68 eV perovskite 124, LiF 908, Ceo 910, BCP 912, and Ag 914 layer configuration has been fabricated. Note that depending on the selection of the substrate layer 902 (which can be n- or p-type Si or other materials instead of glass), the device 900 may be any device that includes a perovskite film (may be p- or n-type doped material), for example, perovskite solar cells, light-emitting diodes (LEDs), sensing devices (e.g., photodetectors), switching devices, memories, and energy storage.

[0045] To obtain micrometer thick perovskites, similar to those used on the textured Si bottom cells in the tandem solar cell 100, concentrated perovskite precursors (1 .7 M) were used, which results in a -900 nm thick perovskite layer 124 on a planar substrate. Such thick films provided a test platform to optimize the carbazole additives for subsequent application to tandem devices. The best singlejunction current-voltage (J- V) curves under forward and reverse scans are shown in Figure 10A and the photovoltaic parameters are listed in the table in Figure 10B. The 2PACz-based control device shows a PCE of 19.2% with a l/_oc of 1.21 V, and a FFof 0.79. With carbazole treatment, the device exhibited a l/oc of 1.22 V, a short-circuit current density (J_sc) of 20.6 mA-crm², and a FFof 0.81 , yielding a PCE of 20.2%. The statistical PCE results demonstrate the reproducibility of the performance improvement after a small amount of carbazole (under 1% of the total mass of the perovskite layer) addition was used. The improvement mainly comes from l/_oc and FF, which are associated with the suppressed non-radiative recombination. The built-in potential change upon carbazole addition has been evaluated by the Mott- Schottky analysis. Compared with the control device, the carbazole-treated device shows a shift toward higher cutoff voltage, indicating a higher flat-band potential, which is also consistent with improved l/oc after carbazole treatment. [0046] To further demonstrate the minimized carrier recombination upon carbazole treatment, the dependence of the V_oc on the light intensity was obtained, as shown in Figure 10C. From the semi-logarithmic plot, it is possible to find a linear n.kT relationship, following the expression slope = —log_we, where n is the diode quality factor. The control and carbazole-treated devices exhibit an n value of 1 .86 and 1 .74, respectively. The lower n value implies that trap-assisted recombination was suppressed in the carbazole-treated device. At the device level, the reduced defect density at the GBs via temperature-dependent admittance spectroscopy measurements was observed. At 300 K, the pristine device shows a peak trap density of 1 .32x10¹⁷ cm^-3 eV^-1 at 0.44 eV, while the passivated device has a smaller peak intensity of 1.23x10¹⁷ cm^-3 eV^-1 at 0.39 eV.

[0047] Absolute photoluminescence (PL) imaging was carried out on the two perovskite devices (traditional single-junction device without carbazole and the device 900) under 1 sun light illumination. The absolute PL image could be converted into quasi-Fermi level splitting (QFLS) and Urbach energy (E_u) mapping. The carbazole-passivated device 900 exhibits a slightly higher and narrower QFLS distribution than the control device, with their statistical mean value of 1 .20 and 1.19 eV, respectively. The QFLS directly reflects the internal voltage of the complete devices, and it is indeed remarkably close to the V_oc obtained from the 1 -sun J-V curves. Ultraviolet photoelectron spectroscopy (UPS) results confirmed that the work function (WF) shifted down from 5.07 to 5.22 eV upon carbazole treatment, which could enable a higher internal voltage and more effective electron transfer at the interface between ETL and perovskite layer. In addition, the Eu distribution map with a mean value of 16.2 meV was also obtained for the control sample (not shown), while the mean value for the carbazole-passivated sample is about 15.8 meV (not shown). The relatively larger QFLS and smaller E_u for carbazole-passivated device 900 provide further evidence for the suppressed non-radiative recombination.

[0048] Similar tests to the single-junction device 900 were performed on the carbazole-treated 1.68 eV perovskites on double-side textured silicon heterojunction bottom cells in the form of the tandem device 100. In the tandem device 100, the conformal coated 2PACz layer 120 formed on the 20 nm ITO layer 118 acts as the recombination junction. Figure 11 A shows the J- V characteristic of the perovskite/Si tandem device 100 under an aperture area of 1 .03 cm². The inset displays the photovoltaic parameters for the device 100, with a PCE of 28.9% and negligible J-V hysteresis. From MPP measurements with the 3-point perturbation method, a stabilized PCE of 28.6% was obtained, with a I/MPP of 1 .58 V. The reproducibility of the processes for obtaining the tandem device 100 was verified, finding a narrow statistical distribution. The corresponding external quantum efficiency (EQE) spectra for this tandem device 100 were obtained. The integrated photocurrent density was calculated to be 19.70 and 19.95 mA/cm², for perovskite and silicon subcells, respectively, which are in good agreement with the J_sc values from the -Vscan curve.

[0049] The tandem device 100’s capabilities are commensurate with the state- of-the-art tandem devices in literature and represents the record-certified PCE for double-textured perovskite/Si tandem devices. The tandem device 100 shows a relatively high J_sc value, by taking the advantage of improved light management in double-side textured bottom cell structures. The relatively large refractive index difference between the Si and perovskite layer generally causes a considerable reflectance from the overall device. Applying a random-pyramid textured structure at the top side of the Si bottom cell increases the optical light path and reduces the reflectance loss. To demonstrate the superiority of the double-textured structure in terms of light-harvesting, the EQE and reflectance of the perovskite subcell on planar and textured Si substrates were compared. It was found that the textured structure significantly reduces the light reflectance in the response region of the c-Si subcell. A certain percentage of photons at the band edge of the perovskite layer (wavelength of 580 to 720 nm) still could transmit through the perovskite cell to reach the Si bottom cell. With the highly crystalline micrometer thick perovskite, the photon utilization of the perovskite subcell near the band edge was also greatly improved. [0050] The scalability potential of the tandem device 100 was shown on a larger substrate with an aperture area of 3.8 cm². A best PCE of 27.1% under reverse scan was achieved, as shown in Figure 11 B. The l/_oc of this device reached 1 .88 V, even slightly higher than that of the ~1 cm² area device. Based on the respective performance results of perovskite and silicon single-junction devices, the l/oc contribution from perovskite and Si subcell may be ~1 .19 and -0.69 V, respectively. The EQE spectra (not shown) from three different spots show that the integrated sc is within a range of ±0.08 mA/cm². All these findings confirm the uniformity of the device 100 on larger substrates. The absolute -3% FFand 0.5 mA/cm² Jsc losses between -1 cm² and -3.8 cm² devices are due to the relatively large series resistance of evaporated silver contacts, and additional shading losses of the silver fingers.

[0051] To evaluate the effect of the carbazole additive on the stability of the tandem device 100, a series of tests on encapsulated devices were performed under different environmental conditions, including outdoor and continuous light soaking tests, as well as 85 °C/85% IEC 61215:2016 (also called ISOS-D3) damp heat tests. Specifically, 43 days outdoor testing in a hot and sunny desert climate includes all quadrants of the operational conditions such as continuous operation under illumination and heat, and heat cycles from day to night. The daily peak device temperature can reach above 45 °C and the light intensity on a sunny day can reach 0.95 suns (measured by a pyranometer) at noon. These conditions represent an ideal test platform to understand the real-world behavior of the tandem device 100 at a targeted operational location, such as for large solar energy plants. For better comparison, the daily performance evolution of the pristine device (dashed lines 1210 and 1212) is superimposed on that of the carbazole-treated device (filled areas) as shown in Figure 12.

[0052] On the sub-graph of the short-circuit current (/_sc), the two devices show an almost coinciding trend. However, the daily peak l/_oc line 1220 and the power output (Pm) line 1222 of the tandem device 100 gradually diverge from the control device with the passing of time. The carbazole-treated device 100 exhibits superior stability compared with the control device. For the carbazole-treated device 100, on the first day (23-Nov-2020), the peak P_m value and the corresponding light intensity was 18.2 mW/cm² and 0.937 sun, yielding a PCE of 19.4%. After 38 days of outdoor testing, the PCE retention could reach almost 98%. In contrast, the control device shows a faster decay in performance, with the power output decreasing from 17.7 mW/cm² initially to 13.0 mW/cm² and the PCE from 19% to 14.6% after 40 days. Depending on the intensity and distribution of the daily spectrum, the daily PCE or retention shows some fluctuations. Overall, after 40 days of outdoor testing, the performance retention reaches over 93% for the carbazole-treated device 100, while it is only about 77% for the control device. The performance degradation in the control device was mainly from the decrease in l/_oc, possibly indicating the degradation in this field test was from perovskite itself. These results demonstrate that the carbazole additive can improve the stability of tandem devices under light and heat stress.

[0053] Next, the inventors carried out the operational stability evaluation of tandem devices in a controlled lab environment under continuous Xenon-lamp illumination with MPP tracking. After 250 h, the reference device showed about 8% performance loss. In contrast, the carbazole-treated device 100 retained nearly 100% of the initial performance. The main difference between the two devices is in the initial few hours of the aging process, which may be associated with a rapid increase in the density of trap states for the control device. The bias during the aging process may cause ion migration and promote the generation of some trap defects, which especially occurred in the initial aging process. In the subsequent aging process, the trap state density may be favorably stabilized at the GBs or at the interfaces, and accordingly, the performance would be relatively stable. Therefore, it is concluded that the defect formation rate in the reference device was faster than that in the carbazole-treated device 100.

[0054] To demonstrate the tolerance of the encapsulated devices to high temperature and high humidity environments, a damp heat test at 85 °C with 85% relative humidity was performed. Notably, after nearly 500 hours, the l/_oc and _sc in the tandem device 100 were quite stable, and the PCE retention could reach up to 87%. The performance degradation mainly was found in the FF, which could be associated with degraded top contact electrode. The highly stable l/_oc implies that the perovskite layer itself did not degrade after 500 h of damp heat testing. It was concluded that as the testing time progressed, the degradation in the damp heat test should be due to moisture invasion rather than inherent thermal decomposition.

[0055] After the long-term stability testing, the PL mappings (not shown) on two tandem devices were obtained. For the control device, significant phase segregation could be observed in the range of hundreds of microns, possibly implying irreversible degradation. Based on the PL spectra (not shown), some narrow bandgap (<1 .68 eV) regions appeared in a large areal portion; these regions show a stronger PL intensity because the iodine-rich phase primarily acts as a recombination trap. In contrast, such phase segregation was not observed in the carbazole-treated devices after outdoor testing, as evidenced by the uniform distribution of PL spectra. These results confirm that the carbazole treatment suppressed phase segregation, thus improving the long-term operational stability of the tandem devices. [0056] The carbazole-treated perovskite devices discussed above can be made with a low-cost method to improve the phase-stable wide-bandgap perovskite films, especially for wide-bandgap perovskite film. The method can be adopted to the any perovskite absorber processing without a need to adjust the original process and without the increasing in process complexity. The carbazole additive provides improved electronic properties by passivating the interfacial and surface defects, and improves the phase stability simultaneously

[0057] According to an embodiment, a method for making a perovskite-based device 100, 900 is now discussed with regard to Figure 13. The method includes a step 1300 of providing a semiconductor substrate 102, 902, a step 1302 of mixing perovskite precursors with an organic additive 310 to form a perovskite mixture, a step 1304 of applying the perovskite mixture to the semiconductor substrate 102, 902, and a step 1306 of annealing the perovskite mixture to form an additive-treated perovskite layer 124. An amount of the organic additive is equal to or less than 1% of a total mass of the additive-treated perovskite layer, and the organic additive includes carbon, hydrogen and nitrogen, which are distributed at borders between grains (Gi) of the perovskite layer.

[0058] The disclosed embodiments provide a semiconductor device that includes an additive-treated perovskite layer. It should be understood that this description is not intended to limit the invention. On the contrary, the embodiments are intended to cover alternatives, modifications and equivalents, which are included in the spirit and scope of the invention as defined by the appended claims. Further, in the detailed description of the embodiments, numerous specific details are set forth in order to provide a comprehensive understanding of the claimed invention. However, one skilled in the art would understand that various embodiments may be practiced without such specific details.

[0059] Although the features and elements of the present embodiments are described in the embodiments in particular combinations, each feature or element can be used alone without the other features and elements of the embodiments or in various combinations with or without other features and elements disclosed herein.

[0060] This written description uses examples of the subject matter disclosed to enable any person skilled in the art to practice the same, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the subject matter is defined by the claims, and may include other examples that occur to those skilled in the art. Such other examples are intended to be within the scope of the claims.

References

The entire content of all the publications listed herein is incorporated by reference in this patent application.

[1 ] Wang, R., Xue, J., Wang, K. L., Wang, Z. K., Luo, Y., Penning, D., Xu, G., Nuryyeva, S., Huang, T., Zhao, Y. et al. (2019). Constructive molecular configurations for surface-defect passivation of perovskite photovoltaics. Science 366, 1509-1513.

[2] Bai, S., Da, P., Li, C., Wang, Z., Yuan, Z., Fu, F., Kawecki, M., Liu, X., Sakai, N., Wang, J. T. et al. (2019). Planar perovskite solar cells with long-term stability using ionic liquid additives. Nature 571, 245-250. [3] Lin, Y. H., Sakai, N., Da, P., Wu, J., Sansom, H. C., Ramadan, A. J., Mahesh, S., Liu, J., Oliver, R. D. J., Lim, J. et al. (2020). A piperidinium salt stabilizes efficient metal-halide perovskite solar cells. Science 369, 96-102.

[4] Zhou, Y., Wang, F., Cao, Y., Wang, J.-P., Fang, H.-H., Loi, M. A., Zhao, N. & Wong, C.-P. (2017). Benzylamine-Treated Wide-Bandgap Perovskite with High Thermal- Photostability and Photovoltaic Performance. Adv. Energy Mater. 7, 1701048.

[5] Isikgor, F. H., Furlan, F., Liu, J., Ugur, E., Eswaran, M. K., Subbiah, A. S., Yengel, E., De Bastiani, M., Harrison, G. T., Zhumagali, S. et al. (2021 ). Concurrent cationic and anionic perovskite defect passivation enables 27.4% perovskite/silicon tandems with suppression of halide segregation. Joule 5, 1566-1586

Claims

29 WHAT IS CLAIMED IS:

1 . A tandem solar cell (100) comprising: a first-type base (102) having first and second surfaces (102A, 102B), opposite to each other; a second-type layer (108) formed on the second surface (102B) of the first- type base (102), wherein the first-type is one of n- or p-type and the second-type is another of the n- or p-type, and the second-type layer (108) and the first-type base (102) form a first pn junction; and an additive-treated perovskite layer (124) formed on the first surface (102A) of the first-type base (102), wherein the additive-treated perovskite layer (124) and the first-type base (102) form a second pn junction, wherein the additive-treated perovskite layer (124) has an organic additive (310) that includes carbon, hydrogen and nitrogen, and is distributed at borders between grains (Gi) of the perovskite layer (124), and wherein the additive-treated perovskite layer (124) includes a perovskite material that has a band gap equal to or larger than 1 .65 eV.

2. The tandem solar cell of Claim 1 , wherein the additive is carbazole.

3. The tandem solar cell of Claim 1 , wherein the additive is imidazole or piperidine. 30

4. The tandem solar cell of Claim 1 , wherein the perovskite material includes

CSO.O5FAo.8MAo.15Pb(lo.75Bro.25)3.

5. The tandem solar cell of Claim 1 , wherein a mass of the additive is equal to or less than 1 % of a total mass of the additive-treated perovskite layer.

6. The tandem solar cell of Claim 1 , wherein the first and second surfaces of the first-type base are structured to have corresponding random pyramids.

7. The tandem solar cell of Claim 1 , further comprising: a layer of [2-(9H-carbazol-9-yl)ethyl]phosphonic acid, 2PACz, formed between the first-type base and the additive-treated perovskite layer.

8. The tandem solar cell of Claim 7, wherein the additive-treated perovskite layer is formed directly onto the 2PACz layer.

9. The tandem solar cell of Claim 1 , further comprising: a layer of Poly[bis(4-phenyl)(2,4,6-trimethylphenyl)amine, PTAA, formed between the first-type base and the additive-treated perovskite layer.

10. The tandem solar cell of Claim 9, wherein the additive-treated perovskite layer is formed directly onto the PTAA layer.

1 1 . A single-junction device (900) comprising: a semiconductor base (902); and an additive-treated perovskite layer (124) formed on a surface of the semiconductor base (102), wherein the additive-treated perovskite layer (124) and the semiconductor base (102) form a pn junction, wherein the additive-treated perovskite layer (124) has an organic additive (310) that includes carbon, hydrogen and nitrogen, and is distributed at borders between grains (Gi) of the perovskite layer (124), and wherein the additive-treated perovskite layer (124) includes a perovskite material that has a band gap equal to or larger than 1 .65 eV.

12. The single-junction device of Claim 11 , wherein the additive is carbazole.

13. The single-junction device of Claim 11 , wherein the additive is imidazole or piperidine.

14. The single-junction device of Claim 11 , wherein the perovskite material includes Cso.o5FAo.8MAo.i5Pb(lo.75Bro.25)3.

15. The single-junction device of Claim 11 , wherein a mass of the additive is equal to or less than 1 % of a total mass of the additive-treated perovskite layer.

16. The single-junction device of Claim 11 , further comprising: a layer of [2-(9H-carbazol-9-yl)ethyl]phosphonic acid, 2PACz, formed between the semiconductor base and the additive-treated perovskite layer.

17. The single-junction device of Claim 16, wherein the additive-treated perovskite layer is formed directly onto the 2PACz layer.

18. The single-junction device of Claim 11 , further comprising: a layer of Poly[bis(4-phenyl)(2,4,6-trimethylphenyl)amine, PTAA, formed between the semiconductor base and the additive-treated perovskite layer.

19. The single-junction device of Claim 18, wherein the additive-treated perovskite layer is formed directly onto the PTAA layer.

20. A method for making a perovskite-based device (100, 900), the method comprising: providing (1300) a semiconductor substrate (102, 902); mixing (1302) perovskite precursors with an organic additive (310) to form a perovskite mixture; applying (1304) the perovskite mixture to the semiconductor substrate (102,

902); and annealing (1306) the perovskite mixture to form an additive-treated perovskite layer (124), 33 wherein a mass of the organic additive (310) is equal to or less than 1% of a total mass of the additive-treated perovskite layer (124), wherein the organic additive (310) includes carbon, hydrogen and nitrogen, which are distributed at borders between grains (Gi) of the perovskite layer (124), and wherein the additive-treated perovskite layer (124) includes a perovskite material that has a band gap equal to or larger than 1 .65 eV.