References to Related Applications

This application claims priority to Singapore application number 10202011433U filed on 17 November 2020 and Singapore application number 10202109920V filed on 9 September 2021, the disclosures of which are hereby incorporated by reference.

Technical Field

The present invention generally relates to a perovskite precursor. The present invention also relates to a method of making a perovskite precursor. The present invention further relates to a perovskite ink. The present invention further relates to a method of forming a film, the film formed from the method and a cell comprising the film.

Background Art

The meteoric upswing in power conversion efficiencies (PCE) of halide perovskite solar cell (PSC) from 3.85% to 25.6% in over a decade makes this technology the youngest member in the high-efficiency photovoltaic league Apart from a host of photophysical properties that make them ideal for use as light absorbers (e.g., tunable bandgaps, low exciton binding energies and long charge-carrier diffusion lengths), halide perovskites (HPs) offer a significant advantage in terms of process versatility, allowing for more facile and low-cost fabrication. However, conventional high- efficiency PSCs are mostly demonstrated on a small area (<1 cm ² ), and further process engineering is needed to realize industrial-scale production with minimum loss in PCE. Spin-coating is typically used for highly efficient PSCs, which is far from ideal for industrial manufacturing.

Several conventional scalable coating techniques have been developed to address the problems above, such as spray deposition, slot-die coating, inkjet printing, screen printing, and blade coating. Among these scalable techniques, slot-die coating method is especially promising, owing to its compatibility with the roll-to-roll or sheet-to-sheet fabrication process. Other advantages include minimal material wastage and high coating uniformity.

Apart from the deposition method, the perovskite ink precursor plays a crucial role in determining the final film quality, device performance and stability. This is because the perovskite’s constituents tend to exist as haloplumbates (Pb ₍₁ X _m ) ⁺² ' ^1-m complex in the precursor solution (rather than pure Pb ²⁺ and halide X- ions) that can interact with polar aprotic solvents, such as dimethyl sulfoxide (DMSO) and dimethylformamide (DMF). Therefore, any minute change in the ink composition or stoichiometry would result in a large variation in photovoltaic parameters owing to alteration of the crystallization dynamics and final material composition. Consequently, stabilizing the perovskite phase is an important strategy to achieve desired photovoltaic parameters, by employing suitable functional additives. These additives can lower the formation energy, slow down the crystallization rate, control nucleation, enhance hydrophobicity, induce stable intermediate, and defect passivation at grain boundaries and interfaces. Most conventional methods involving additives have been on spin-coated small area devices.

One conventional method adopted the additive engineering and solvent tuning of the perovskite precursor ink which allowed a wide precursor-processing window with fast grain growth rate.

Another conventional method required the addition of ammonium chloride (NH4CI) into perovskite precursor solution to retard the nucleation of the perovskite, which achieved 15.86 % efficiency on modules with an active area of 42.9 cm ² on flexible glass substrate via blade coating.

Alkali metal additive-assisted strategies have been conventionally used to improve crystallization and reduce hysteresis by modulating the interfacial defects or trap densities in spin-coated small area PSCs. For instance, the conventional addition of potassium iodide (KI) to perovskite reduces ion migration and passivates negatively charged trap states. Similarly, Rb ⁺ , Cs ⁺ and Na ⁺ halide salts have also been utilized conventionally to achieve reduced defect density and high carrier lifetime.

A further conventional method demonstrated a high PCE of 21 % with improved stability upon continuous illuminations (10% of loss after 1000 hours) upon addition of sodium fluoride (NaF) salts, which passivated both anion and cation vacancies in the perovskite film.

Another conventional method had CsI (0.1 mol) added into FAPbb, not only to stabilize the photoactive a-FAPbb perovskite phase but also to homogenize the halide distribution within a mixed perovskite phase, resulting in increased charge carrier lifetime and improved PSC performance. While various additives have shown improved optoelectronic properties, remarkable PCEs and exceptional device stability in small area spin-coated devices, their roles and location of those additives in the resultant perovskite film remain unclear. Thus, understanding the impact of the additives on structural and optoelectronic properties of perovskite thin film, for large scale slot-die coating is of significance. Further, the additive engineering strategy has not been demonstrated for large area scalable slot die coating.

Accordingly, there is a need for a perovskite precursor and a method of making the same that address or ameliorate one or more problems above. Summary

In one aspect, there is provided a perovskite precursor having the formula: la-Ib, wherein la is a compound of the formula ABX3 or AB2X5, wherein A is one or more alkaline metals, B is lead and X is one or more halides; wherein lb is a compound of the formula CBY3, wherein C is selected from the group consisting of one or more alkaline metals, one or more monovalent organic cations and combinations thereof, and Y is one or more halides; and wherein the one or more monovalent organic cations is independently of the formula R ^T R ² R ³ R ⁴ N ⁺ , wherein R ¹ , R ² , R ³ and R ⁴ are independently selected from the group consisting of hydrogen, C1-C20 alkyl, C1-C20 haloalkyl, C1-C20 heteroalkyl, C2-C20 alkenyl, C2-C20 haloalkenyl, C2-C20 heteroalkenyl, C2-C20 alkynyl, C2-C20 haloalkynyl, C2-C20 heteroalkynyl, C3-C20 cycloalkyl, C3-C20 halocycloalkyl, C3-C20 heterocycloalkyl, C6-C20 aryl, C6-C20 haloaryl, C6-C20 heteroaryl and combinations thereof; or wherein R ¹ and R ² together form a moiety selected from the group consisting of C1-C20 alkylene, C1-C20 haloalky lene, C1-C20 heteroalkylene, C2-C20 alkenylene, C2-C20 haloalkenylene, C2-C20 heteroalkenylene, C3-C20 cycloalkylene, C3-C20 halocycloalkylene, C3-C20 heterocycloalkylene and combinations thereof; and R ³ and R ⁴ are independently selected from the group consisting of hydrogen, C1-C20 alkyl, C1-C20 haloalkyl, C1-C20 heteroalkyl, C2-C20 alkenyl, C2-C20 haloalkenyl, C2-C20 heteroalkenyl, C2-C20 alkynyl, C2-C20 haloalkynyl, C2-C20 heteroalkynyl, C3-C20 cycloalkyl, C3-C20 halocycloalkyl, C3-C20 heterocycloalkyl, C6-C20 aryl, C6-C20 haloaryl, C6-C20 heteroaryl and combinations thereof.

Advantageously, the compound of formula la may have suitable solubility properties in a perovskite ink composition. This may provide nucleation sites for the precursor, such that when the perovskite ink composition (comprising the perovskite precursor as described herein) is made into a film, the film may have a mean grain size that is improved to at least about 320 nm and at least about 9 % increase compared to a film devoid of the compound of formula la. The film may then have better electric properties. Compared to a film made from a conventional precursor where la is absent, the film of the present disclosure may have a power conversion efficiency that is at least 10% higher.

In another aspect, there is provided a method of making a perovskite precursor, comprising the step of mixing a compound of formula la and a compound of formula lb, wherein la is of the formula ABX3 or AB2X5, wherein A is one or more alkaline metals, B is lead and X is one or more halides; wherein lb is of the formula CBY3, wherein C is selected from the group consisting of one or more alkaline metals, one or more monovalent organic cations and combinations thereof, and Y is one or more halides; and wherein the one or more monovalent organic cations is independently of the formula R ¹ R ² R ³ R ⁴ N ⁺ , wherein R ¹ , R ² , R ³ and R ⁴ are independently selected from the group consisting of hydrogen, C1-C20 alkyl, C1-C20 haloalkyl, C1-C20 heteroalkyl, C2-C20 alkenyl, C2-C20 haloalkenyl, C2-C20 heteroalkenyl, C2-C20 alkynyl, C2-C20 haloalkynyl, C2-C20 heteroalkynyl, C3-C20 cycloalkyl, C3-C20 halocycloalkyl, C3-C20 heterocycloalkyl, C6-C20 aryl, C6-C20 haloaryl, C6-C20 heteroaryl and combinations thereof; or wherein R ¹ and R ² together form a moiety selected from the group consisting of C1-C20 alkylene, C1-C20 haloalky lene, C1-C20 heteroalkylene, C2-C20 alkenylene, C2-C20 haloalkenylene, C2-C20 heteroalkenylene, C3-C20 cycloalkylene, C3-C20 halocycloalkylene, C3-C20 heterocycloalkylene and combinations thereof; and R ³ and R ⁴ are independently selected from the group consisting of hydrogen, C1-C20 alkyl, C1-C20 haloalkyl, C1-C20 heteroalkyl, C2-C20 alkenyl, C2-C20 haloalkenyl, C2-C20 heteroalkenyl, C2-C20 alkynyl, C2-C20 haloalkynyl, C2-C20 heteroalkynyl, C3-C20 cycloalkyl, C3-C20 halocycloalkyl, C3-C20 heterocycloalkyl, C6-C20 aryl, C6-C20 haloaryl, C6-C20 heteroaryl and combinations thereof.

In another aspect, there is provided a perovskite ink composition comprising the perovskite precursor as described herein and a solvent.

In another aspect, there is provided a method of forming a film, comprising the steps of:

(a) heating the perovskite ink composition as described herein to form a heated mixture; and

(b) coating the heated mixture of step (a) onto a substrate to form the film.

Advantageously, heating the perovskite ink composition with constant stirring may aid in better homogeneity of the perovskite precursor, followed by the coating step to form the film. The film thus formed may have a mean grain size that is improved to at least about 320 nm, and by at least about 9 % compared to a film devoid of the compound of formula la, which improves its electric properties. Compared to a film made from a conventional precursor where la is absent, the film of the present disclosure may have a power conversion efficiency that is at least 10% higher.

In another aspect, there is provided a method of improving a mean grain size of a film, comprising the steps of:

(a) heating the perovskite ink composition as described herein to form a heated mixture; and (b) coating the heated mixture of step (a) onto a substrate to form the film, wherein the mean grain size of the film is at least about 320 nm and wherein the mean grain size of the film is improved by at least about 9% compared to a film devoid of the compound of the formula la.

In another aspect, there is provided a film formed by the method as described herein.

Advantageously, the film may have improved electrical properties due to the improved mean grain size with higher crystallinity. Compared to a film made from a conventional precursor where la is absent, the film of the present disclosure may have a power conversion efficiency that is at least 10% higher.

In another aspect, there is provided a cell comprising the film as described herein, a hole transport layer material and a counter electrode.

Definitions

The following words and terms used herein shall have the meaning indicated:

As used herein, the term "alkyl group" includes within its meaning monovalent (“alkyl”) and divalent (“alkylene”) straight chain or branched chain saturated aliphatic groups having from 1 to 20 carbon atoms.

The term "alkenyl group" includes within its meaning monovalent (“alkenyl”) and divalent (“alkenylene”) straight or branched chain unsaturated aliphatic hydrocarbon groups having from 2 to 20 carbon atoms, and having at least one double bond, of either E, Z, cis or trans stereochemistry where applicable, anywhere in the alkyl chain.

The term "alkynyl group" as used herein includes within its meaning monovalent (“alkynyl”) and divalent (“alkynylene”) straight or branched chain unsaturated aliphatic hydrocarbon groups having from 2 to 20 carbon atoms and having at least one triple bond anywhere in the carbon chain.

The term “cycloalkyl” as used herein refers to cyclic saturated aliphatic groups and includes within its meaning monovalent (“cycloalkyl”), and divalent (“cycloalkylene”), saturated, monocyclic, bicyclic, polycyclic or fused polycyclic hydrocarbon radicals having from 3 to 20 carbon atoms.

The term “heterocycloalkyl” as used herein, includes within its meaning monovalent (“heterocycloalkyl”) and divalent (“heterocycloalkylene”), saturated, monocyclic, bicyclic, polycyclic or fused hydrocarbon radicals having from 3 to 20 ring atoms wherein 1 to 5 ring atoms are heteroatoms selected from O, N, NH, or S.

The term “heteroaromatic group” and variants such as “heteroaryl” or “heteroarylene” as used herein, includes within its meaning monovalent (“heteroaryl”) and divalent (“heteroarylene”), single, polynuclear, conjugated and fused aromatic radicals having 6 to 20 atoms wherein 1 to 6 atoms are heteroatoms selected from O, N, NH and S. Examples of such groups include pyridyl, 2,2’- bipyridyl, phenanthrolinyl, quinolinyl, thiophenyl, and the like.

The term “halogen” or variants such as “halide” or “halo” as used herein refers to fluorine, chlorine, bromine and iodine.

The term “heteroatom” or variants such as “hetero-” as used herein refers to O, N, NH and S.

Unless specified otherwise, the terms "comprising" and "comprise", and grammatical variants thereof, are intended to represent "open" or "inclusive" language such that they include recited elements but also permit inclusion of additional, unrecited elements.

The term "about" as used herein typically means +/- 5 % of the stated value, more typically +/- 4 % of the stated value, more typically +/- 3 % of the stated value, more typically, +/- 2 % of the stated value, even more typically +/- 1 % of the stated value, and even more typically +/- 0.5 % of the stated value.

Throughout this disclosure, certain embodiments may be disclosed in a range format. It should be understood that the description in range format is merely for convenience and brevity and should not be construed as an inflexible limitation on the scope of the disclosed ranges. Accordingly, the description of a range should be considered to have specifically disclosed all the possible sub-ranges as well as individual numerical values within that range. For example, description of a range such as from 1 to 6 should be considered to have specifically disclosed sub-ranges such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6 etc., as well as individual numbers within that range, for example, 1, 2, 3, 4, 5, and 6. This applies regardless of the breadth of the range.

Certain embodiments may also be described broadly and generically herein. Each of the narrower species and subgeneric groupings falling within the generic disclosure also form part of the disclosure. This includes the generic description of the embodiments with a proviso or negative limitation removing any subject matter from the genus, regardless of whether or not the excised material is specifically recited herein.

Detailed Disclosure of Embodiments

Exemplary, non-limiting embodiments of a perovskite precursor having the formula la-Ib will now be disclosed.

In the perovskite precursor, la is a compound having the formula ABX3 or AB2X5. Where the same A, B and X are present (whereby A of ABX3 is the same as the A in AB2X5, the B of ABX3 is the same as the B in AB2X5 and the X of ABX3 is the same as the X in AB2X5), compounds of these two formulas may have the same chemical and/or physical properties.

Advantageously, the compound of formula la may have a suitable solubility in a perovskite ink composition. This may provide nucleation sites for the precursor, such that when the perovskite ink composition is made into a film, the film may have a mean grain size that is improved to at least about 320 nm, and by at least about 9 % compared to a film devoid of the compound of formula la. The film may then have better electric properties. Compared to a film made from a conventional precursor where la is absent, the film of the present disclosure may have a power conversion efficiency that is at least 10% higher.

In the compound of formula la, A may be one or more alkaline metals. A may be selected from the group consisting of lithium, sodium, potassium, rubidium, cesium and a combination thereof.

In the compound of formula la, B may be lead.

In the compound of formula la, X may be one or more halides. X may be selected from the group consisting of chlorine, bromine, iodine and a combination thereof.

The compound of formula la may be CsPbBr? or KPb2Br5.

In the perovskite precursor, lb is a compound having the formula CBY3.

In the compound of formula lb, C may be selected from the group consisting of one or more alkaline metals, one or more monovalent organic cations and combinations thereof.

In C, the one or more monovalent organic cations may independently be of the formula R ^T R ² R ³ R ⁴ N ⁺ .

In C, R ¹ , R ² , R ³ and R ⁴ may be independently selected from the group consisting of hydrogen, C1-C20 alkyl, C1-C20 haloalkyl, C1-C20 heteroalkyl, C2-C20 alkenyl, C2-C20 haloalkenyl, C2-C20 heteroalkenyl, C2-C20 alkynyl, C2-C20 haloalkynyl, C2-C20 heteroalkynyl, C3-C20 cycloalkyl, C3-C20 halocycloalkyl, C3-C20 heterocycloalkyl, C6-C20 aryl, C6-C20 haloaryl, C6-C20 heteroaryl and combinations thereof.

In C, R ¹ and R ² may additionally or alternatively together form a moiety selected from the group consisting of C1-C20 alkylene, C1-C20 haloalkylene, C1-C20 heteroalkylene, C2-C20 alkenylene, C2-C20 haloalkenylene, C2-C20 heteroalkenylene, C3-C20 cycloalkylene, C3-C20 halocycloalkylene, C3-C20 heterocycloalkylene and combinations thereof; and R ³ and R ⁴ are independently selected from the group consisting of hydrogen, C1-C20 alkyl, C1-C20 haloalkyl, C1-C20 heteroalkyl, C2-C20 alkenyl, C2-C20 haloalkenyl, C2-C20 heteroalkenyl, C2-C20 alkynyl, C2-C20 haloalkynyl, C2-C20 heteroalkynyl, C3-C20 cycloalkyl, C3-C20 halocycloalkyl, C3-C20 heterocycloalkyl, C6-C20 aryl, C6-C20 haloaryl, C6-C20 heteroaryl and combinations thereof.

In C, the one or more monovalent organic cations may be selected from the group consisting of formamidinium (FA), iodoformamidinium, bromoformamidinium, methylammonium, ethylammonium, propylammonium and combinations thereof.

In the compound of formula lb, where C is a combination of the one or more alkaline metals and the one or more monovalent organic cations, the molar ratio of the one or more alkaline metals to the one or more monovalent organic cations may be in the range of about 0.05:0.95 to about 0.95:0.05, about 0.2:0.8 to about 0.95:0.05, about 0.4:0.6 to about 0.95:0.05, about 0.6:0.4 to about 0.95:0.05, about 0.8:0.2 to about 0.95:0.05, about 0.05:0.95 to about 0.8:0.2, about 0.05:0.95 to about 0.6:0.4, about 0.05:0.95 to about 0.4:0.6 or about 0.05:0.95 to about 0.2:0.8. In an example where C is a combination of cesium and formamidinium, molar ratio of the cesium to the formamidinium may be as defined above and exemplarily about 0.15:0.85.

In the compound of formula lb, Y may be one or more halides. Where Y is a combination of two halides, the molar ratio of one halide to the other halide may be in the range of about 0.05:0.95 to about 0.95:0.05, about 0.2:0.8 to about 0.95:0.05, about 0.4:0.6 to about 0.95:0.05, about 0.6:0.4 to about 0.95:0.05, about 0.8:0.2 to about 0.95:0.05, about 0.05:0.95 to about 0.8:0.2, about 0.05:0.95 to about 0.6:0.4, about 0.05:0.95 to about 0.4:0.6 or about 0.05:0.95 to about 0.2:0.8. In an example where Y is a combination of bromine and iodine, the molar ratio of bromine to iodine may be as defined above and exemplarily about 0.17:0.83.

The molar ratio between the two halides may be suitably selected to change the colour of a cell that is formed from the perovskite precursor.

In the perovskite precursor, the compound of formula la may have a mole percentage in the range of about 1 mole% to about 20 mole% based on the number of moles of the compound of formula lb. The mole percentage of the compound of formula la may be about 3 mole% based on the number of moles of the compound of formula lb.

The perovskite precursor may be of the formula CsPbBr3-Cs _x FAi- _x Pb(I _y Bri- _y )3 (where la is CsPbBr ₃ and lb is Cs _x FAi- _x Pb(I _y Bri- _y )3) or KPb2Br5-Cs _m FAi-mPb(I _n Bri- n)3 (where la is KPb ₂ Br ₅ and lb is Cs _m FAi-mPb(I _n Bri-n)3), wherein x, y, m and n may independently be a number selected from the range of about 0.05 to about 0.95, about 0.2 to about 0.95, about 0.4 to about 0.95, about 0.6 to about 0.95, about 0.8 to about 0.95, about 0.05 to about 0.2, about 0.05 to about 0.4, about 0.05 to about 0.6, about 0.05 to about 0.8, about 0.1 to about 0.2. The perovskite precursor may be of the formula CsPbBr3-Cso.i5FAo.85Pb(Io.83Bro.i7)3 (where la is CsPbBr ₃ and lb is Cso.i5FAo.85Pb(Io.83Bro.i?)3) or KPb ₂ Br ₅ - Cso.i ₅ FAo.85Pb(Io.83Bro.i7)3 (where la is KPb ₂ Br ₅ and lb is Cso.i5FAo.85Pb(Io.83Bro.i7)3).

Exemplary, non-limiting embodiments of a method of making a perovskite precursor will now be disclosed.

The method may comprise the step of mixing a compound of formula la and a compound of formula lb.

In the method, the compound of la may be of the formula ABX3 or AB2X5. Where the same A, B and X are present (whereby A of ABX3 is the same as the A in AB2X5, the B of ABX3 is the same as the B in AB2X5 and the X of ABX3 is the same as the X in AB2X5), compounds of these two formulas may have the same chemical and/or physical properties.

In the compound of formula la, A may be one or more alkaline metals. A may be selected from the group consisting of lithium, sodium, potassium, rubidium, cesium and a combination thereof.

In the compound of formula la, B may be lead.

In the compound of formula la, X may be one or more halides. X may be selected from the group consisting of chlorine, bromine, iodine and a combination thereof.

The compound of formula la may be CsPbBr ₃ or KPb ₂ Br ₅ .

In the perovskite precursor, lb is a compound having the formula CBY3.

In the compound of formula lb, C may be selected from the group consisting of one or more alkaline metals, one or more monovalent organic cations and combinations thereof.

In C, the one or more monovalent organic cations may independently be of the formula R ^T R ² R ³ R ⁴ N ⁺ .

In C, R ¹ , R ² , R ³ and R ⁴ may be independently selected from the group consisting of hydrogen, C1-C20 alkyl, C1-C20 haloalkyl, C1-C20 heteroalkyl, C2-C20 alkenyl, C2-C20 haloalkenyl, C2-C20 heteroalkenyl, C2-C20 alkynyl, C2-C20 haloalkynyl, C2-C20 heteroalkynyl, C3-C20 cycloalkyl, C3-C20 halocycloalkyl, C3-C20 heterocycloalkyl, C6-C20 aryl, C6-C20 haloaryl, C6-C20 heteroaryl and combinations thereof.

In C, R ¹ and R ² may additionally or alternatively together form a moiety selected from the group consisting of C1-C20 alkylene, C1-C20 haloalkylene, C1-C20 heteroalkylene, C2-C20 alkenylene, C2-C20 haloalkenylene, C2-C20 heteroalkenylene, C3-C20 cycloalkylene, C3-C20 halocycloalkylene, C3-C20 heterocycloalkylene and combinations thereof; and R ³ and R ⁴ are independently selected from the group consisting of hydrogen, C1-C20 alkyl, C1-C20 haloalkyl, C1-C20 heteroalkyl, C2-C20 alkenyl, C2-C20 haloalkenyl, C2-C20 heteroalkenyl, C2-C20 alkynyl, C2-C20 haloalkynyl, C2-C20 heteroalkynyl, C3-C20 cycloalkyl, C3-C20 halocycloalkyl, C3-C20 heterocycloalkyl, C6-C20 aryl, C6-C20 haloaryl, C6-C20 heteroaryl and combinations thereof.

In C, the one or more monovalent organic cations may be selected from the group consisting of formamidinium (FA), iodoformamidinium, bromoformamidinium, methylammonium, ethylammonium, propylammonium and combinations thereof.

In the compound of formula lb, where C is a combination of the one or more alkaline metals and the one or more monovalent organic cations, the molar ratio of the one or more alkaline metals to the one or more monovalent organic cations may be may be in the range of about 0.05:0.95 to about 0.95:0.05, about 0.2:0.8 to about 0.95:0.05, about 0.4:0.6 to about 0.95:0.05, about 0.6:0.4 to about 0.95:0.05, about 0.8:0.2 to about 0.95:0.05, about 0.05:0.95 to about 0.8:0.2, about 0.05:0.95 to about 0.6:0.4, about 0.05:0.95 to about 0.4:0.6 or about 0.05:0.95 to about 0.2:0.8. In an example where C is a combination of cesium and formamidinium, the molar ratio of the cesium to the formamidinium may be as defined above and exemplarily about 0.15:0.85.

In the compound of formula lb, Y may be one or more halides. Where Y is a combination of two halides, the molar ratio of one halide to the other halide may be in the range of about 0.05:0.95 to about 0.95:0.05, about 0.2:0.8 to about 0.95:0.05, about 0.4:0.6 to about 0.95:0.05, about 0.6:0.4 to about 0.95:0.05, about 0.8:0.2 to about 0.95:0.05, about 0.05:0.95 to about 0.8:0.2, about 0.05:0.95 to about 0.6:0.4, about 0.05:0.95 to about 0.4:0.6 or about 0.05:0.95 to about 0.2:0.8. In an example where Y is a combination of bromine and iodine, the molar ratio of bromine to iodine may be as defined above and exemplarily about 0.17:0.83.

The molar ratio between the two halides may be suitably selected to change the colour of a cell that is formed from the perovskite precursor.

In the perovskite precursor, the compound of formula la may have a mole percentage in the range of about 1 mole% to about 20 mole% based on the number of moles of the compound of formula lb. The mole percentage of the compound of formula la may be about 3 mole% based on the number of moles of the compound of formula lb.

The perovskite precursor may be of the formula CsPbBr3-Cs _x FAi- _x Pb(I _y Bri-y)3 (where la is CsPbBr ₃ and lb is Cs _x FAi- _x Pb(I _y Bri- _y )3) or KPb2Br5-Cs _m FAi-mPb(I _n Bri- n)3 (where la is KPb ₂ Br ₅ and lb is Cs _m FAi-mPb(I _n Bri-n)3), wherein x, y, m and n may independently be a number selected from the range of about 0.05 to about 0.95, about 0.2 to about 0.95, about 0.4 to about 0.95, about 0.6 to about 0.95, about 0.8 to about 0.95, about 0.05 to about 0.2, about 0.05 to about 0.4, about 0.05 to about 0.6, about 0.05 to about 0.8, about 0.1 to about 0.2. The perovskite precursor may be of the formula CsPbBr3-Cso.i5FAo.85Pb(Io.83Bro.i7)3 (where la is CsPbBr ₃ and lb is Cso.i5FAo.85Pb(Io.83Bro.i?)3) or KPb ₂ Br ₅ - Cso.i ₅ FAo.85Pb(Io.83Bro.i7)3 (where la is KPb ₂ Br ₅ and lb is Cso.i5FAo.85Pb(Io.83Bro.i7)3).

In the method, the mixing step may be undertaken in an inert atmosphere or in a glovebox.

Exemplary, non-limiting embodiments of a perovskite ink composition will now be disclosed.

The perovskite ink composition may comprise the perovskite precursor as described herein and a solvent.

Non- limiting examples of the solvent include DMF, DMSO, gamma-butyrolactone, N-methyl pyrrolidone and combinations thereof. The solvent maybe a combination of DMF and DMSO having a volume ratio of 4: 1.

In the perovskite ink composition, the perovskite precursor may be provided at a concentration in the range of about 0.5 to about 3 M, about 1 M to about 3 M, about 2 M to about 3 M, about 0.5 M to 2 M or about 0.5 M to about 1 M.

The perovskite ink composition may be stored under an inert atmosphere.

Exemplary, non-limiting embodiments of a method of forming a film will now be disclosed.

The method may comprise the steps of:

(a) heating the perovskite ink composition as described herein to form a heated mixture; and

(b) coating the heated mixture of step (a) onto a substrate to form the film.

Advantageously, heating the perovskite ink composition may aid in better homogeneity of the perovskite precursor, followed by the coating step to form the film. The film thus formed may have a mean grain size that is improved to at least about 320 nm, and by at least about 9 % compared to a film devoid of the compound of formula la, which improves its electric properties. Compared to a film made from a conventional precursor where la is absent, the film of the present disclosure may have a power conversion efficiency that is at least 10% higher.

The increased mean grain size may be about at least about 320 nm, at least about 350 nm, at least about 380 nm or at least about 400 nm. This increased mean grain size is with reference to a comparative film that is formed using the same conditions but where the perovskite precursor making up the perovskite ink composition does not have the compound of formula la. Accordingly, the increase in the mean grain size of the perovskite film of the present application can additionally be regarded as a percentage increase from that of the comparative film whereby the percentage increase in the mean grain size of the present perovskite film from the comparative film is an increase of at least about 9%, at least about 10%, at least about 15%, at least about 20%, at least about 25%, at least about 30%, at least about 35%.

The heating step (a) may be undertaken at a temperature in the range of about 25 °C to about 70 °C , about 50 °C to about 70 °C or about 25 °C to about 50 °C. The heating step may be undertaken at a temperature of about 50 °C .

The heating step (a) may be undertaken for a duration of at least about 8 hours or overnight (such as at least about 12 hours, at least about 16 hours, at least about 20 hours, or at least about 24 hours), with constant stirring. This step of heating the perovskite ink composition for at least 8 hours can also be regarded or termed as an ageing step. A perovskite ink composition that did not undergo an ageing step is one which is heated to form a heated mixture and then used straight away to form the film. Here, the perovskite ink composition that did not undergo an ageing step can be one which contains the perovskite precursor of the formula la-Ib or one which contains a perovskite precursor of formula lb only (that is, compound of formula la is absent). In one example, the heating (or ageing step) can be undertaken for a duration of about 8 hours to about 24 hours, or about 24 hours.

The coating step (b) may be undertaken by a coating technique. Non-limiting examples of the coating technique include slot die coating, blade coating, spin coating, inkjet printing, spray coating, and combinations thereof. The coating step (b) may be undertaken by slot die coating.

In the coating step (b), non-limiting examples of the substrate include fluoride-doped tin oxide (FTO) glass, modified FTO glass, indium tin oxide (ITO) glass, polyethylene terephthalate, polyethylene naphthalate and combinations thereof.

The modified FTO glass may be modified by depositing a layer of material. The depositing step may be undertaken by a spin coating, sputtering, thermal evaporation, chemical bath depositing or combinations thereof. The material may be tin oxide (SnO2), TiO2, ZnO, phenyl-C61 -butyric acid methyl ester (PCBM) or combinations thereof.

Exemplary, non-limiting embodiments of a method of improving a mean grain size of a film will now be disclosed.

The method may comprise the steps of:

(a) heating the perovskite ink composition as described herein to form a heated mixture; and

(b) coating the heated mixture of step (a) onto a substrate to form the film, In the method, the mean grain size of the film may be at least about 320 nm. The mean grain size of the film may be improved by at least about 9% compared to a film devoid of the compound of the formula la.

The increased mean grain size may be about at least about 320 nm, at least about 350 nm, at least about 380 nm or at least about 400 nm. This increased mean grain size is with reference to the comparative film (as described above). Accordingly, the increase in the mean grain size of the perovskite film of the present application can additionally be regarded as a percentage increase from that of the comparative film whereby the percentage increase in the mean grain size of the present perovskite film from the comparative film is an increase of at least about 9%, at least about 10%, at least about 15%, at least about 20%, at least about 25%, at least about 30%, at least about 35%.

The heating step (a) may be undertaken at a temperature in the range of about 25 °C to about 70 °C, about 50 °C to about 70 °C or about 25 °C to about 50 °C . The heating step may be undertaken at a temperature of about 50 °C .

The heating step (a) may be undertaken for a duration of at least about 8 hours or overnight (such as at least about 12 hours, at least about 16 hours, at least about 20 hours, or at least about 24 hours), with constant stirring. This step of heating the perovskite ink composition for at least 8 hours can also be regarded or termed as an ageing step. A perovskite ink composition that did not undergo an ageing step is one which is heated to form a heated mixture and then used straight away to form the film. Here, the perovskite ink composition that did not undergo an ageing step can be one which contains the perovskite precursor of the formula la-Ib or one which contains a perovskite precursor of formula lb only (that is, compound of formula la is absent). In one example, the heating (or ageing step) can be undertaken for a duration of about 8 hours to about 24 hours, or about 24 hours.

The coating step (b) may be undertaken by a coating technique. Non-limiting examples of the coating technique include slot die coating, blade coating, spin coating, inkjet printing, spray coating, and combinations thereof. The coating step (b) may be undertaken by slot die coating.

In the coating step (b), non-limiting examples of the substrate include fluoride-doped tin oxide (FTO) glass, modified FTO glass, indium tin oxide (ITO) glass, polyethylene terephthalate, polyethylene naphthalate and combinations thereof.

The modified FTO glass may be modified by depositing a layer of material. The depositing step may be undertaken by a spin coating, sputtering, thermal evaporation, chemical bath depositing or combinations thereof. The material may be tin oxide (SnO2), TiO2, ZnO, phenyl-C61 -butyric acid methyl ester (PCBM) or combinations thereof. Exemplary, non-limiting embodiments of a film form by the method as described herein will now be disclosed.

The film may comprise crystallites having a mean grain size of at least about 320 nm, and that is increased by at least about 9 % compared to a film devoid of the compound of formula la.

The mean grain size may be about at least about 320 nm, at least about 350 nm, at least about 380 nm or at least about 400 nm. This mean grain size is with reference to the comparative film (as described above). Accordingly, the increase in the mean grain size of the perovskite film of the present application can additionally be regarded as a percentage increase from that of the comparative film whereby the percentage increase in the mean grain size of the present perovskite film from the comparative film is an increase of at least about 9%, at least about 10%, at least about 15%, at least about 20%, at least about 25%, at least about 30%, at least about 35%. Advantageously, the film may have improved electrical properties due to the high mean grain size that is at least about 320 nm, and that is increased by at least about 9 % compared to a film devoid of the compound of formula la. Compared to a film made from a conventional precursor where la is absent, the film of the present disclosure may have a power conversion efficiency that is at least 10% higher.

Exemplary, non-limiting embodiments of a cell will now be disclosed.

The cell may comprise the film as described herein, a hole transport layer material and a counter electrode.

In the cell, the hole transport layer may comprise N ² ,N ² ,N ² ,N ² ,N ⁷ ,N ⁷ ,N ⁷ ,N ⁷ - octakis(4-methoxyphenyl)-9, 9 '- spirobi [9H-fluorene] -2,2', 7, 7 '-tetramine (Spiro-

OMeTAD) or a modified Spiro-OMeTAD.

The modified Spiro-OMeTAD may comprise Spiro-OMeTAD and an additive. Nonlimiting examples of the additive include lithium bis(trifluoromethylsulfonyl)imide (Li-TFSI), zinc bis(trifluoromethylsulfonyl)imide (Zn-(TFSI)2) 4-tert-butylpyridine, a solvent or combinations thereof.

In the cell, the counter electrode may be a gold, copper, aluminium, ITO, silver or carbon electrode.

The cell may be a solar cell such as a perovskite solar cell.

Brief Description of Drawings

The accompanying drawings illustrate a disclosed embodiment and serves to explain the principles of the disclosed embodiment. It is to be understood, however, that the drawings are designed for purposes of illustration only, and not as a definition of the limits of the invention. Fig. 1

[Fig. 1] shows chemical reaction schemes for the synthesis of CsPbBr ₃ and KPb ₂ Br ₅ , and theoretical and experimental X-ray diffraction (XRD) patterns of KPb ₂ Br ₅ .

Fig. 2

[Fig. 2] shows characterizations of perovskites. Fig. 2(a) shows X-ray crystal structure of KPb ₂ Br ₅ bromoplumb ate. Fig. 2(b) is a schematic presentation of the precursor solution containing CsPbBr ₃ / KPb ₂ Br ₅ clusters together with [PbI ₆ ] ^4- octahedron and DMF, DMSO solvent molecules. Fig. 2(c) shows dynamic light scattering (DLS) profiles of CsPbBr ₃ , KPb ₂ Br ₅ , CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ - CsFA precursor solutions. Fig. 2(d) shows XRD patterns of CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA perovskite films. Fig. 2(e) shows time-resolved photoluminescence (TRPL) spectra of CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA perovskite films. Fig. 2(f) shows a top morphological field emission scanning electron microscopy (FESEM) image of KPb2Br5-CsFA perovskite film (inset is histogram of KPb2Br5-CsFA perovskite grain size). Fig. 2(g) shows a cross sectional FESEM image of typical KPb2Br5-CsFA perovskite device. Fig. 2(h) shows 2D grazing-incidence wide-angle X-ray scattering (GIWAXS) patterns of CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA perovskite films.

Fig. 3

[Fig. 3] shows transmission electron microscopy (TEM) analysis to support the existence of bromoplumbate cluster in precursor solution. Presented in Fig. 3(a) and Fig. 3(b) are high-resolution TEM (HRTEM) image of drop casted KPb2Br5-CsFA solution showing small nanocrystals (< 10 nm). Lattice spacings of 0.45 and 0.74 correspond to (Oil) perovskite plane and (001) Pbl2 plane, respectively, while 0.27 to (031) KPb ₂ Br ₅ plane. The latter is confirmed with the HRTEM image in Fig. 3(c) where drop casted KPb ₂ Br ₅ solution shows 0.27, 0.31, and 0.36 nm lattice spacings ascribed to (031), (300), and (11-3) plane of KPb ₂ Br ₅ , respectively.

Fig. 4

[Fig. 4] shows a schematic illustration of perovskite nucleation and crystallization in the (a) absence and (b) presence of CsPbBr ₃ or KPb ₂ Br ₅ additives. Fig. 4(c) shows grazing-incidence XRD (GIXRD) patterns of a CsFA perovskite film without both air flow quenching and post-treatment annealing (inset shows photograph of the corresponding CsFA film coated on glass/FTO substrate).

Fig. 5

[Fig. 5] shows (a) an enlarged view of XRD patterns of CsFA, CsPbBr ₃ -CsFA and KPb2Br5-CsFA perovskite films; and (b) UV-Vis absorbance and photoluminescence spectra of CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA perovskite films coated on glass substrates. Fig. 6

[Fig. 6] shows top morphological FESEM images of (a) CsFA and (b) CsPbBr ₃ -CsFA perovskite layers. Fig. 6(c) and Fig. 6(d) are cross sectional FESEM images of CsFA, and CsPbBr ₃ -CsFA perovskite solar cell devices, respectively. Fig. 6(e) shows a cross sectional energy dispersive X-ray spectroscopy (EDX) analysis of KPb ₂ Br ₅ -CsFA perovskite solar cell devices.

Fig. 7

[Fig. 7] shows ID profiles in the out-of-plane direction for CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA perovskite films.

Fig. 8

[Fig. 8] shows Typical J-V curves of perovskite solar cells with varying CsPbBr ₃ and KPb ₂ Br ₅ concentrations (0% to 5%).

Fig. 9

[Fig. 9] shows DLS profiles of CsPbBr ₃ -CsFA with varying ageing durations: 0 day, 1 day, and 2 days.

Fig. 10

[Fig. 10] shows ultraviolet (UV) and photoluminescence (PL) spectra of KPb ₂ Br ₅ - CsFA with varying ageing durations: 0 day, 1 day, 2 days and 7 days.

Fig. 11

[Fig. 11] shows top morphological FESEM images of KPb2Br5-CsFA with varying ageing durations: 0 day, 1 day, 2 days and 7 days.

Fig. 12

[Fig. 12] shows XRD spectra of KPb2Br5-CsFA with varying ageing durations: 0 day, 1 day, 2 days and 7 days.

Fig. 13

[Fig. 13] shows device performance statistics for KPb2Br5-CsFA perovskite solar cell (PSC) (a) short-circuit current density (J _sc ), (b) open-circuit voltage (V _oc ), (c) fill factor (FF), and (d) power conversion efficiency (PCE) with varying ageing durations: 0 day, 1 day, 2 days and 7 days.

Fig. 14

[Fig. 14] shows device performance statistics for CsPbBr ₃ -CsFA PSC (a) J _sc , (b) Voc, (c) FF, and (d) PCE with varying ageing durations: 0 day, 1 day and 2 days. Fig. 15

[Fig. 15] shows characterization results of perovskites obtained from the method of the present disclosure. Fig. 15(a) shows J- V curves of the champion devices of CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA perovskite solar cells (inset: device architecture). Fig. 15(b) shows external quantum efficiency (EQE) spectra with Integrated J _sc curves of CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA devices. Fig. 15(c) shows statistical results of 30 devices of PCE, from one batch for CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA perovskite solar cells. Fig. 15(d) shows forward and reverse scan J- V curves of KPb ₂ Br ₅ -CsFA PSC. Fig. 15(e) shows normalized electroluminescence (EL) spectra at different voltages (inset: KPb2Br5-CsFA devices exhibiting red color light-emitting diode (LED)). Fig. 15(f) shows photostability at maximum power point (MPPT) condition for CsFA, and KPb2Br5-CsFA PSCs measured under continuous illumination in dry air environment. Fig. 15(g) shows a schematic illustration of charge traps at interfaces and hybrid perovskite layer: (gl) CsFA and (g2) KPb ₂ Br ₅ - CsFA PSC.

Fig. 16

[Fig. 16] shows photovoltaic performances of various slot die coated perovskite films. The perovskite films coated on SnO ₂ /FTO substrates (10 cm wide and 10 cm long, after slot die coating) was further annealed at 100 °C for 15 minutes. Then the substrates were cut into equal dimension. Device mappings of PSC across the 100 cm ² are shown for (a) CsFA, (e) CsPbBr ₃ -CsFA and (i) KPb ₂ Br ₅ -CsFA, respectively. Jsc, Voc, and FF across the 100 cm ² are shown for (b) to (d) CsFA, (f) to (h) CsPbBr ₃ - CsFA and (j) to (1) KPb ₂ Br ₅ -CsFA, respectively. Arrows indicate coating direction. Photovoltaic parameters displayed here were based on the reverse scanned J-V data.

Fig. 17

[Fig. 17] shows forward and reverse J-V curves of CsFA and CsPbBr ₃ -CsFA PSC.

Fig. 18

[Fig. 18] shows (a) normalized EL spectra of CsFA; and (b) to (c) EQE data for KPb ₂ Br ₅ -CsFA.

Fig. 19

[Fig. 19] shows high-resolution X-ray photoelectron spectroscopy (XPS) spectra of the core levels used to construct the depth profile: (a) Pb 4f, (b) I 3d, (c) Cs 3d, (d) Br 3d, (e) K 2p, (f) N Is, (g) C Is, (h) O Is and (i) Sn 3d.

Fig. 20

[Fig. 20] shows characterization of perovskite films. Fig. 20(a) shows XPS depth profile acquired on a KPb ₂ Br ₅ -CsFA perovskite film on FTO glass with Ar gas cluster ions (10 keV Arl000+). The inset shows the same depth profile focusing on the hybrid perovskite/substrate interface. Fig. 20(b) shows an evolution of Pb 4f, I 3d, Cs 3d and K 2p core levels with etching cycles (60 s/cycle) for KPb ₂ Br ₅ -CsFA perovskite films. Fig. 20(c) and Fig. 20(d) show normalized K 2p and Br 3d spectra for the different bromide compounds (KBr, KPb ₂ Br ₅ and the hybrid lead perovskite), respectively. Fig. 20(e) and Fig. 20(f) show ²⁰⁷ Pb MAS NMR spectra of the control sample, the doped sample and KPb ₂ Br ₅ covering the (e) higher and (f) lower frequency ranges, with an additional spectrum taken with 10 times the number of scans marked in grey.

Fig. 21

[Fig. 21] shows corresponding survey spectra for the depth profile.

Fig. 22

[Fig. 22] shows core level XPS spectra of (a) Pb 4f (before etching), (b) Pb 4f (after etching), (c)Br 3d, (d) I 3d, (e) Cs 3d, (f) Nls, (g) C Is and (h) K 2p.

Fig. 23

[Fig. 23] shows halide-to-lead ratio variation with etch cycles (60 s/cycle). The substrate signals were detected from cycle 15 onwards.

Fig. 24

[Fig. 24] shows core level XPS spectra of KBr K2p, Br and KPb ₂ Br ₅ K2p, Br and Pb.

Fig. 25

[Fig. 25] shows ²⁰⁷ Pb magic-angle spinning (MAS) NMR spectra of KPb ₂ Br ₅ with a simulated chemical shift anisotropy (CSA) spinning sideband manifold (MAS frequency: 30 KHz; δ _iso = -279.89 ppm; J _CSA = -1112.18 ppm; η _CSA = 0.99).

Fig. 26

[Fig. 26] shows NMR analysis of KPb ₂ Br ₅ . Fig. 26(a) shows ¹³³ Cs MAS NMR spectra of control and doped samples, with an additional ¹³³ Cs[ ¹ H] cross polarisation/MAS (CPMAS) spectra of the doped sample marked in grey. Fig. 26(b) shows ⁸¹ Br MAS NMR spectra of KBr and the control sample. MAS sideband manifolds (simulated date) are marked by asterixes throughout.

Fig. 27

[Fig. 27] shows coordination tailored perovskite inks for slot die coating of perovskite films. Fig. 27(a) is a schematic presentation for N2-knife-assisted slot-die coating of perovskite films at 11 mm/s at 55 °C using coordination tailored ink. Insets show the pictures of as-coated perovskite ink, perovskite/intermediate film, perovskite film, and PL mapping of KPb ₂ Br ₅ -CsFA perovskite film. Fig. 27(b) is a schematic representation of drying of perovskite ink followed by perovskite/intermediate film and completely crystallized perovskite film. Fig. 27(c) shows UV-Vis absorption at 9 different areas on a 10 xlO cm ² glass/perovskite substrate. Fig. 27(d) shows a J-V curve of the best performing KPb ₂ Br ₅ -CsFA perovskite solar module with series connection of 13 sub-cells with an active area of 57.5 cm ² , the module being measured under 1 sun illumination without an aperture mask (inset: Perovskite solar module with 13 sub-cells). Fig. 27(e) shows device performance statistics for CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA PSC modules.

Fig. 28

[Fig. 28] shows camera images of perovskite films of the present disclosure. Fig. 28(a) shows a perovskite wet film without air knife quenching. Fig. 28(b) to Fig. 28(d) show a perovskite film with varying gaps between the air knife and the substrate, being 200 pM, 300 pM and 500 pM, respectively. Fig. 28(e) shows a reflective nature of optimized perovskite film. Fig. 28(f) shows a 10cm x 30 cm FTO substrate to support the homogeneity of the perovskite film

Fig. 29

[Fig. 29] shows UV-Vis absorption at 15 different spots on 10 xlO cm ² glass/perovskite substrate.

Fig. 30

[Fig. 30] shows a schematic representation of a module with Pl, P2 and P3 etching.

Fig. 31

[Fig. 31] shows forward and reverse J-V curves of CsFA and CsPbBr ₃ -CsFA perovskite solar modules.

Fig. 32

[Fig. 32] shows device performance statistics for CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA PSC modules by the following parameters: (a) 7 _S c, (b) V _O c, (c) FF, and (d) PCE.

Fig. 33

[Fig. 33] shows stability analyses of PSCs of the present disclosure. Fig. 33(a) shows a stability analysis for CsFA and KPb ₂ Br ₅ -CsFA perovskite devices, stored in dark at room temperature with relative humidity (RH) of 30 % over 10,000 hours (without encapsulation). Fig. 33(b) shows a thermal stability analysis of CsFA and KPb ₂ Br ₅ - CsFA perovskite devices at 65 °C over 1,000 hours (without encapsulation). Fig. 33(c) shows a thermal stability analysis of CsFA and KPb ₂ Br ₅ -CsFA perovskite devices at 85 °C over 900 hours (PIB based blanket encapsulation). Fig. 33(d) shows a stability analysis for CsFA and KPb ₂ Br ₅ -CsFA perovskite mini-modules, stored in dark at room temperature with RH of 30 % over more than 4,500 hours (without encapsulation). Fig. 33(e) shows a photostability analysis at maximum power point (MPPT) condition for CsFA and KPb ₂ Br ₅ -CsFA perovskite mini-module measured under continuous illumination in the dry air environment.

Fig. 34

[Fig. 34] shows device performance statistics for CsFA perovskite solar cells at 65 °C with the following parameters: (a) J _sc , (b) V _oc , (c) FF, and (d) PCE. Device parameters were obtained from reverse J-V scans.

Fig. 35

[Fig. 35] shows device performance statistics for KPb ₂ Br ₅ -CsFA perovskite solar cells at 65 °C with the following parameters: (a) J _sc , (b) V _oc , (c) FF, and (d) PCE. Device parameters were obtained from reverse J-V scans.

Fig. 36

[Fig. 36] shows device performance statistics for CsFA perovskite solar cells at 85 °C with the following parameters: (a) J _sc , (b) V _oc , (c) FF, and (d) PCE. Device parameters were obtained from reverse J-V scans.

Fig. 37

[Fig. 37] shows device performance statistics for KPb ₂ Br ₅ -CsFA perovskite solar cells at 85 °C with the following parameters: (a) J _sc , (b) V _oc , (c) FF, and (d) PCE. device parameters were obtained from reverse J-V scans.

Fig. 38

[Fig. 38] shows device stability analyses of CsFA and KPb ₂ Br ₅ -CsFA modules.

Examples

Non- limiting examples of the invention and a comparative example will be further described in greater detail by reference to specific Examples, which should not be construed as in any way limiting the scope of the invention.

Example 1 - Synthesis of CsPbBr ₃

14.7 g of PbBr2 (99.999% trace metal basis, purchased from Sigma Aldrich of St. Louis, Missouri of the United States of America) was dissolved in 32 mL of HBr (48 weight% in H2O, purchased from Sigma Aldrich). 8.52 g of CsBr (99.999% trace metals basis, purchased from Sigma Aldrich) was dissolved in 20 mL of deionized water (purified from Milli-Q® Integral 3 (Model number: ZRXQ003T0 F7CA88729D)) and was added to the previous solution dropwise. Orange-coloured precipitate was observed after the complete addition of CsBr solution. Further, the precipitate was filtered and washed two times with ethanol (purchased from Alfa Aesar of Haverhill, Massachusetts of the United States of America) to remove all unreacted material and later dried in a vacuum oven at 60 °C for 24 hours.

Example 2 - Synthesis of KPb ₂ Br ₅

14.7 g of PbBr2 was dissolved in 32 mL of HBr. 4.76 g of KBr (>99% trace metals basis, purchased from Sigma Aldrich) was dissolved in 20 mL of deionized water and was added to the previous solution drop by drop. White precipitate was observed after complete addition of KBr solution. Further, the precipitate was filtered and washed two times with ethanol to remove all unreacted material and later dried in a vacuum oven at 60 °C for 24 hours. A person skilled in the art would appreciate that KPb ₂ Br ₅ is the product of the synthesis steps as described above and as characterized below. Thus, references to KPbBr? in priority application SG10202011433U as the product of the synthesis steps above was an error and the correct formula meant to be used in priority application SG10202011433U is KPb ₂ Br ₅ .

Example 3 - Growth of Single Crystals of KPb ₂ Br ₅

A vial was charged with a concentrated solution (about 0.25 to 0.33 M) of presynthesized KPb ₂ Br ₅ powder (as described in Example 2) in dimethyl sulfoxide (DMSO, anhydrous, >99.9%, purchased from Sigma Aldrich), after which an antisolvent acetone (purchased from Sigma Aldrich) was introduced by the vapour diffusion method under ambient conditions. The single crystals, which formed within 2 to 7 days (ca. 40% yield based on Pb content), were then isolated and used for X- ray crystallographic studies.

Example 4 - Preparation of Rigid FTO Substrates

10 x 10 cm ² FTO substrates (purchased from Yingkou Shangneng Optoelectronic material Co., Ltd. from China) were first etched (Pl etching) using a femtosecond laser machine. Further, they were cleaned through subsequent ultrasonic cleaning by Decon 90 Liquid Detergent (purchased from East Sussex of the United Kingdom), deionized water, and ethanol for 30 minutes each, respectively. Later, nitrogen gas flow was used to dry up the films. UV-ozone treatment was done with UV light at a wavelength of 365nm for 15 minutes before the deposition of SnO2 to remove the organic residues.

Example 5 - Preparation of SnO ₂ by Spin Coating (SC)

0.04 M of SnCl ₂ .2H ₂ O (>99.99% trace metals basis, purchased from Sigma Aldrich) solution was prepared in ethanol and kept stirring for 1 hour at room temperature. Later, the solution was deposited on cleaned FTO substrate with three-step of rotation: 1) 1,000 rpm, 500 acceleration, 4 seconds (puddling); 2) 1,000 rpm, 1,000 acceleration, 10 seconds; and 3) 5,000 rpm, 1,000 acceleration, 30 seconds, followed by pre-drying at 80 °C for 10 minutes and then annealed at 180 °C for 1 hour. Example 6 - Preparation of SnO ₂ by Chemical bath (CBD)

A CBD-S11O2 film was grown by chemical bath deposition method on spin-coated layers of SnO ₂ substrate. 10 g urea (purchased from Sigma Aldrich) was first dissolved in 4.2 L of deionized water, followed by the addition of 300 pL mercaptoacetic acid (purchased from Sigma Aldrich) and 10 mL HC1 (37 weight%, purchased from Sigma Aldrich). Finally, SnO ₂ .2H2O was dissolved in the solution at 0.003 M concentration followed by stirring for 3 minutes. The substrates were kept vertically in a glass container filled with the abovementioned solution for 4 hours at 80 °C in a lab oven. Later, the treated CBD samples were washed with deionized water in a sonication bath for 2 minutes to eliminate the loosely bound material, followed by drying by gas gun blowing and pre-drying at 80 °C for 10 minutes and then annealed at 180 °C for 1 hour. UV-ozone treatment was done for 15 minutes for all SnO ₂ layers before the deposition of perovskite film.

Example 7 - Preparation of Mixed Perovskite Precursor

Reference Solution'. A mixed cation lead mixed halide perovskite solution was prepared by dissolving PbI2 (99.99%, trace metal basis, purchased from Tokyo Chemical Industry of Tokyo, Japan), formamidinium iodide (FAI, purchased from Greatcell Solar of Queanbeyan, Australia), PbBr2 and CsI (99.999% trace metal basis, purchased from Sigma Aldrich) in a mixed solvent of DMF : DMSO by 4:1 volume ratio to achieve 1.2 M solution of Cso.i5FAo.85Pb(I0 ₈₃ Br _0.17 )3 perovskite precursor solution with 5 % excess lead halide in a nitrogen -filled glovebox. The above solution was kept at 50°C overnight for stirring and transferred to slot die for coating the next day.

CsPbBr -CsFA: 0 to 5 mol% of CsPbBr ₃ was added with respect to lead iodide in the above reference solution and followed the same condition for heating overnight.

KPb2Br5-CsFA'. 0 to 5 mol % of KPb ₂ Br ₅ was added with respect to lead iodide in the above reference solution and followed the same condition for heating overnight.

Example 8 - Perovskite Film Deposition via Slot Die

Perovskite layers with and without different additives were slot-die coated using nRad slot-die coater (nTact company) on the 10 x 10 cm ² UV-ozone FTO/CBD- SnO ₂ substrate. The complete instrument was present in an enclosed environment with a relative humidity of about 30% to 40%. The condition needed to obtain a uniform and highly crystalline perovskite film was optimized with the perovskite ink. At substrate stage, the reaction temperature was kept at 52 °C to coat the perovskite layer with a coating speed of 10.5 mm/s and dispense rate of 4.5 pL/sec. An air knife was attached to the slot-die head and N2 gas was used as a carrier gas with a flow rate of 140 L/minute for a fast crystallization of perovskite. The distance between the slotdie head and the substrate was 100 pm and the distance between the air knife and substrate was 300 m. After completing the perovskite layer, the substrate was heated at 100 °C for 15 minutes in a dark closed chamber with a relative humidity of 10%.

Example 9 - Hole Transporting Layer and Counter Electrode

A spiro-OMeTAD solution with additive was prepared with 72.3 mg of N ² ,N ² ,N ² ,N ² ,N ⁷ ,N ⁷ ,N ⁷ ,N ⁷ -octakis(4-methoxyphenyl)-9,9'-spirobi-[9H-fluorene]-2 ,2',7,7'- tetramine (Spiro-OMeTAD, purchased from Lumtec Technology Corp., Taiwan) in 1 mL of chlorobenzene (purchased from Sigma Aldrich). 520 mg of lithium bis(trifluoromethylsulfonyl)imide (Li-TFSI, 99.95% trace metals basis, purchased from Sigma Aldrich) was dissolved in 1 mL of acetonitrile (ACN, purchased from Sigma Aldrich) and 29 μL of 4-tert-butylpyridine (purchased from Sigma Aldrich), and 17.23 pL of this solution was added to the spiro-OMeTAD solution above. The combined solution (HTM solution) was spin-coated dynamically over perovskite at 4,000 rpm, 1,000 acc, for 40 seconds in a glovebox. Finally, a 100 nm gold top electrode (purchased from ACI Alloys of San Jose, California of the United States of America) was thermally evaporated under a high vacuum.

Example 10 - Characterisation of Products

UV-vis spectroscopy and. photoluminescence'. UV-vis absorption spectra of perovskite thin films coated from slot-die were recorded in the range 300 to 820 nm wavelength using a SHIMADZU UV-3600 spectrophotometer. The steady-state photoluminescence spectra were recorded by using Spectro-fluorophotometer (Shimadzu, RF-5301PC), under the excitation of a 650 nm light source.

Slot die instrument'. Perovskite film was coated by N-Rad slot die coater (N-Takt company) under controlled environment condition of 30% to 40% relative humidity.

Thickness measurements'. NanoMap 500 LS (AEP Technologies) instrument was used to measure the thickness of perovskite film.

Surface analysis'. The top morphology and cross-section of the device were measured by Jeol JSM-7600F Field Emission Scanning Electron Microscope (FE-SEM) at 5 to 10 kV with a working distance of about 8 mm.

XRD analysis'. The XRD spectra of different perovskite film were collected by a Bruker D8 Advance X-ray diffractometer with a Copper source and a detector range from 10 to 50°.

GIWAXS XRD'. The Grazing incidence wide-angle X-ray scattering (GIWAXS) data of perovskite deposited on Glass substrate were obtained at beamline BL14B1 of the Shanghai Synchrotron Radiation Facility (SSRF). Example 11 - Solar Cell Devices and Incident Photon -To- Current Efficiencies

The J-V curves of these perovskite solar cells were measured under AM 1.5G (100 mW/cm ² ) spectral irradiation from the solar simulator (Newport), which was calibrated with a Si reference cell (Oriel PN91150) incorporating a 450 W xenon lamp (model 81172, Oriel). The active area of the device was determined by a black mask metal aperture of 0.09 cm ² . PVE300 (Bentham) was used to record the Incident Photon-to-Current Conversion Efficiency (IPCE) measurements, with a dual xenon/quartz halogen light source, measured in DC mode, (Newport Oriel Sol3A solar simulator with a 450-W Xenon lamp) in the wavelength range 300 - 850 nm. (Dark scan 1.2 to -0.1 V; Forward scan -0.1 to 1.2 V; Reverse scan 1.2 to -0.1 V).

Example 12 - Solid-State NMR Measurements

All solid-state NMR experiments in this study were completed on a 14.1 T Bruker Advance III HD 600 MHz spectrometer with a 1.9 mm Bruker HXY probe. All spectra were processed using the Topspin software package and referenced to the unified scale using IUPAC recommended frequency ratios.

The ²⁰⁷ Pb NMR (v ₀ ( ²⁰⁷ Pb) = 125.416 - 125.758 MHz) experiments were acquired at MAS frequencies of 24 or 30 KHz. They employed Hahn-echo pulse sequences utilising ^π / 2 and it pulses of 4.125 and 8.25 μ s (determined on Pb(NO ₃ )2(s)), sample dependent recycle delays of 1 to 4 seconds, and a rotor synchronised echo delay. The ²⁰⁷ Pb experiments utilized variable offset cumulative spectroscopy (VOCS) frequency stepping where appropriate to allow for full excitation of the large frequency range present.

The ¹³³ Cs NMR (v ₀ )( ¹³³ Cs) = 78.724 MHz) one -pulse experiments were acquired at an MAS frequency of 24 KHz, utilising a selective π/ 2 pulse of 6.5 μ s (determined on CsCl(s)) and a recycle delay of 60 seconds. The ¹³³ Cs[ ¹ H] cross polarization NMR experiment was acquired at an MAS frequency of 12 KHz, utilising a 5000 μ s contact pulse length, a ¹ H π/ 2 pulse 2.75 μ s, high power proton decoupling and a recycle delay of 77.5 seconds.

The ⁸¹ Br NMR (v ₀ )( ⁸¹ Br) = 162.10 MHz) one-pulse experiments were acquired at an MAS frequency of 24 KHz, utilising a selective π/ 2 pulse of 3.3 μ s (determined on KBps)) and a recycle delay of 0.5 second.

Example 13 - LED Measurement

The electrical poling was performed using the same Keithley 2612B by applying a fixed constant bias (1, 2 and 3 V) for a set duration of 1 minute. All PeLED devices were tested at ambient conditions. The characteristic current density -voltage- luminance (J-V-L) were recorded with a Keithley 2612B source meter and an OceanOptics QE Pro spectrometer connected to an integrating sphere and operated using Ciemo Lab VIEW software.

Example 14 - XPS Measurement

X-ray photoelectron spectroscopy (XPS) depth profiling was performed using an AXIS Supra spectrometer (Kratos Analytical Inc., UK) equipped with a hemispherical analyser and a monochromatic Al K-alpha source (1487 eV) operated at 15 mA and 15 kV. The etching was done with an Ar Gas Cluster Ion Source (GCIS, Kratos Analytical Inc., Minibeam 6) operated at 10 keV, Ariooo ⁺ with a raster size of 2 x 2 mm ² . The high-resolution XPS spectra were acquired from an area of 700 x 300 pm ² after 60-second etch for each cycle. The sample was electrically grounded to the sample holder to prevent charge build-up on the sample surface.

To generate the depth profile, in principle the core-level spectra were fitted using mixed Gaussian-Lorentzian line shapes after background subtraction. The composition was further determined based on the transfer-function corrected integrated peak area for each core level and the corresponding relative sensitivity factor (RSF).

Example 15 - TRPL and PL Measurements

TRPL dynamics were collected using the micro-PL setup, employing a Nikon microscope, and using a Picoquant PicoHarp 300 time-correlated single-photon counting (TCSPC) system. A picosecond pulsed laser diode, Picoquant P-C-405B, 1 = 405 nm with 2.5 MHz repetition (40 Hz frequency divided by a 16 factor) rate was used as the excitation source. The excitation fluence was signal was coupled to an avalanche diode synchronized with an excitation laser via TCSPC electronics. Overall, the full width at half the maximum of the system instrument response function was around 50 ps.

PL measurements were performed using Allalin 4027 Chronos from Attolight AG.

Example 16 - X-Ray Crystallography Measurement

Single crystals were mounted on a Bruker X8 Quest CPAD area detector diffractometer, and data was collected at room temperature using an IpS 3.0 Microfocus Mo Ka radiation source ( = 0.71073 A). Data reduction and absorption corrections were performed using the SAINT and SADABS software packages, respectively. All structures were solved by direct methods and refined by full-matrix least-squares procedures on F2, using the Bruker SHELXTL-2014 software package. Non-hydrogen atoms were anisotropically refined, after which hydrogen atoms were introduced at calculated positions and subsequent further refinement of the data performed. Graphical depictions of the crystal structures were created using the programs Mercury and VESTA. Example 17 - Characterisation Results

The solution chemistry of perovskite ink has a huge impact on film properties and performance of PSCs. Here, to consistently improve the crystallization of high- quality perovskite films, pre-synthesized CsPbBr? and KPb ₂ Br ₅ perovskites were introduced as processing additive into the perovskite precursor inks (Cso.i ₅ FAo.85Pb(Io.83Bro.i7)3) (see Table 1 for detailed synthesis of CsPbBr ₃ perovskite or KPb ₂ Br ₅ bromoplumbate together with the crystallographic data of KPb ₂ Br ₅ , and Fig. 1 for the corresponding powder XRD pattern). The crystallographic and structure refinement data for KPb ₂ Br ₅ are shown in Table 1. The X-ray crystal structure of KPb ₂ Br ₅ bromoplumbate is shown in Fig. 2a. Since these inorganic alkali-based perovskites had lower solubility in conventional organic solvents in comparison to their organic counterparts such as MA and formamidinium (FA) cations, they were expected to induce the formation of clusters in the precursor solution, playing a crucial role in perovskite crystallization as schematically proposed in Fig. 2b. Dynamic light scattering (DES) measurements were performed on pristine CsPbBr ₃ and KPb ₂ Br ₅ solutions and peaks were observed at 0.68 nm, 443 nm, and 5500 nm in the DES profile of CsPbBr ₃ sample and at 1.2 nm and 5400 nm in that of KPb ₂ Br ₅ (Fig. 2c, dashed lines). This indicated that bromoplumbates, such as CsPbBr ₃ and KPb ₂ Br ₅ , were not completely dissolved but existed as various clusters in the solution. High-Resolution Transmission Electron Microscopy (HRTEM) analysis also supported the existence of such cluster, as shown and discussed in detail in Fig. 3.

Table 1. Crystallographic and structure refinement data for KPb2Br5. ^a

Compound

KPb ₂ Br ₅

Empirical formula Br ₅ KPb ₂ Formula weight 853.03 g/mol Temperature 296(2) K Wavelength 0.71073 A Crystal size 0.060 x 0.100 x 0.120 mm

Crystal habit colorless block

Crystal system monoclinic

Space group P 1 21/c 1

Unit cell dimensions a = 9.2675(10) A a = 90° b = 8.3809(7) A p = 90.013(8)° c = 13.0581(13) A y = 90°

Volume 1014.22(17) A ³

Z 4 Density 5.587 g/cm ³

Absorption coefficient 53.184 mm ¹

F(000) 1432

Theta range for data >

„ . ⁵ 2.20 to 28.81° collection

Reflections collected 7370

Coverage of independent ... . 99.0% reflections

Absorption correction Multi-Scan Max. and min.

0.1430 and 0.0610 transmission Function minimized Data / restraints /

2622 / 0 / 74 parameters Goodness-of-fit on F ² 1.003 A/omax 0.001 Final R indices 1809 data [I > 2G(I)] R1 = 0.0625, wR2 = 0.1343

R indices [all data] R1 = 0.1024, wR2 = 0.1559

Largest diff. peak and

4.438 and -4.322 eÅ ^-3 hole

R.M.S. deviation from ,

0.804 eA ^-3 mean

Following that, further DLS measurements were carried out on control ink solution

(CsFA), control ink with CsPbBr3 additive (CsPbBr ₃ -CsFA) and control ink with KPb ₂ Br ₅ additive (KPb ₂ Br ₅ -CsFA) to investigate how the ink chemistry of the perovskite precursors changed with the addition of the alkali-based additives. As shown in Fig 2c (solid lines), the pristine CsFA sample was found to feature only one single peak at around 1.4 nm that corresponded to the cluster of small particles that were typically present in perovskite precursors. Interestingly, however, both CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA samples exhibited large size particle peaks that appear at similar positions as those observed in pristine CsPbBr ₃ and KPb ₂ Br ₅ . Those peaks were ascribed to the presence of the soft coordination complex of Pb-I frameworks or solvated perovskite colloids in the ink. The presence of such large particle peaks was conventionally deemed disadvantageous for perovskite film fabrication and various ways had been developed to supress the presence of big colloidal aggregates by adding either acids or ionic liquids into precursor solutions. However, conventional studies also found that peaks at diameters of larger than 100 nm were preferrable to induce enlarged grain sizes. The presence of the large particle peaks in the present disclosure suggested that alkalibased additives could induce a higher supersaturation state in the precursor ink relative to that of pristine CsFA (due presumably to their low solubility) and it was found that the large colloidal particles were useful in terms of improving the morphology and crystallinity of slot-die coated perovskite films (vide infra). The observed effect could stem from the different nucleation and crystal growth processes that the perovskite solutions undergo upon additives inclusion (see Fig. 4 for the schematic representation of perovskite film formation in the absence and presence of additives). Generally, the crystallization of perovskites (e.g., CsFA) would not occur until the formation of nuclei, in which a critical free energy barrier is overcome at the point where the nuclei grow beyond the critical radius. Such nucleation of particles (which was believed to be related to polymeric Pb(I/Br)2:DMSO/DMF complexes represented in Fig. 4c.) is triggered by supersaturation in the precursor solution and in an uncontrolled condition, this would result in high number of nucleation sites during the slot die coating process. In the presence of the additives, the perovskite nucleation and growth could be promoted as it is easier to attain solute critical concentration required for the nucleation of [Pbl6] ^4- complexes to initiate under higher supersaturation condition. As a result, the CsPbBr3-CsFA and KPb ₂ Br ₅ -CsFA films exhibit bigger mean grain sizes, better crystallinity, and homogeneity in comparison to the control CsFA film (Fig. 4b).

Fig. 2d showed the XRD patterns of CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA films on FTO/SnO2 substrates. The diffraction pattern was in good agreement with the cubic phase of CsFA perovskites, showing 20 peaks of 14.04°, 19.96°, 24.44°, 28.13°, 31.75°, 37.83° and 43.13° for (001), (Oi l), (111), (002), (012), (112) and (003) planes, respectively. On the other hand, the peak at 12.67° corresponded to excess PbI2 in the precursor solution. The inclusion of CsPbBr3 and KPb ₂ Br ₅ into CsFA perovskite solution did not cause any peak shift. However, it changed the perovskite (001)-to-Pbl2 peak intensity ratio from 1.10 (CsFA) to 1.35 and 1.75 in CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA, respectively (Fig. 5a). This indicated that the presence of CsPbBr3 and KPb ₂ Br ₅ clusters promoted the crystallization of CsPbBr3- CsFA and KPb ₂ Br ₅ -CsFA slot-die coated perovskite films. In addition, the full widths at half maximum (FWHM) for CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA films based on the (012) peak were 0.30 ±0.02°, 0.28 ±0.02° and 0.26 ±0.02°, respectively. This was in agreement with the increasing mean grain size upon additive inclusion.

The optical absorption and photoluminescence (PL) spectra of the films were presented in Fig. 5b to 5d for CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA films. CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA perovskites showed strong absorption in the visible region with absorption onset at 782 nm, 770 nm and 775 nm, respectively. The slight blue shift of absorption onset could be due to the additional bromide ions from CsPbBr ₃ and KPb ₂ Br ₅ clusters, which is also reflected in the luminescence peaks. Time-resolved photoluminescence (TRPL) measurements of CsFA, CsPbBr?- CsFA and KPb2Br5-CsFA films were conducted at room temperature (Fig. 2e) and decay curves were fitted with a biexponential function. The two characteristic lifetimes ( τ1 and τ2) resulted from the fitting procedure were reported in Table 2. The films incorporating bromoplumbate clusters showed increased lifetime: indeed of the nonradiative ( τ1) and the radiative lifetime ( τ2) increased in CsPbBr3-CsFA and KPb ₂ Br ₅ -CsFA films, compared to the pristine CsFA film. Consequently, the average lifetime (x _a v) increased from 118 ns in CsFA, to 149 ns and 367 ns in CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA perovskites. The enhanced carrier lifetime indicated the additive- included films have lower defects which reduce the non-radiative pathways.

Table 2. Time-resolved photoluminescence fitting data for different perovskite films. α1 τ1 α2 τ2 Tavg (ns)

CsFA 0.85 2.9 0.15 132 118 CsPbBr ₃ -CsFA 0.60 8.9 0.40 159 148

KPb ₂ Br ₅ -CsFA 0.42 3.6 0.58 370 368

Fig. 2f showed field emission scanning electron microscopy (FESEM) images of KPb2Br5-CsFA films (see Fig. 6a and 6b for CsFA and CsPbBr ₃ -CsFA films). The average mean grain sizes of CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA were 320 ± 10 nm, 350 ± 11 nm and 460 ± 12 nm, respectively (the substrate of this experiment was FTO layer). The increased mean grain size was attributed to the perovskite clusters in the ink solution, which induced the controlled nucleation and result in higher crystallinity. As shown in the cross-sectional FESEM image of the complete perovskite solar cell (Fig. 2g), the average thickness of the slot-die coated perovskite layer was approximately 550 nm, which was in good agreement with the result obtained from the surface profilometer (in the following section). In the CsFA-based PSCs (Fig. 6a), smaller grains were observed along with a visible void/gap between the electron transport layer (ETL) and perovskite films. Owing to the cluster-assisted nucleation during the formation of CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA films, significantly larger mean grains were obtained together with better crystallization at the ETL/perovskite interface. As part of the preliminary investigation, energy dispersive X-ray spectroscopy (EDX) was performed on the cross-section of the KPb ₂ Br ₅ -CsFA perovskite PSCs (Fig. 6e) to elucidate the existence of K ⁺ . The result clearly indicated that the K ⁺ was mainly distributed at the SnO2/perovskite interface. The effect of cluster inclusion on perovskite crystallinity was subsequently monitored using grazing-incidence wide-angle X-ray scattering (GIWAXS). The 2D GIWAXS patterns of CsFA, CsPbBr ₃ -CsFA, and KPb ₂ Br ₅ -CsFA perovskite films were depicted in Fig. 2h while the corresponding ID profiles in the out-of-plane direction were plotted in Fig. 7. Although there was a variation in grain size, the perovskite films displayed similar diffraction rings, with the (110), (220) and (310) diffraction peaks located at qz values of 1.48, 2.11 and 3.00 A respectively, and similar XRD results represented in Fig. 7. Example 18 - Electrical Properties of Perovskite Solar Cells

PSCs using CsFA, CsPbBR ₃ -CsFA and KPb ₂ Br ₅ -CsFA as light-harvesting layers were fabricated in n-i-p configuration (FTO/SnO2/perovskite/Spiro-OMeTAD/Au). Typical J-V curves of perovskite solar cells with varying CsPbBR ₃ and KPb ₂ Br ₅ concentration (0 to 5 %) were presented in Fig. 8, where 3 % of additives showed optimized device performance. Their corresponding photovoltaic parameters were summarized in Table 3.

Table 3. Photovoltaic parameters of perovskite solar cells with varying CsPbBR ₃ and KPb ₂ Br ₅ concentration (0% to 5%).

Ageing of CsPbBR ₃ -CsFA and KPb ₂ Br ₅ -CsFA perovskite inks was carried out to observe its impact on the final film and device performance (see Fig. 9 to Fig. 14). It was found that ageing the precursor solution for 1 day gave the best efficiency with narrow efficiency distribution for CsPbBR ₃ -CsFA and KPb ₂ Br ₅ -CsFA perovskite solar devices as depicted in Fig. 13 and Fig. 14. The J-V curves of the champion devices of CsFA, CsPbBR ₃ -CsFA and KPb ₂ Br ₅ -CsFA perovskites were shown in Fig. 15a and the photovoltaic parameters were summarized in Table 4.

Table 4. Photovoltaic parameters of slot-die coated PSCs. aChampion device efficiency; ^b Average reverse scan efficiency of 30 devices.

The CsFA device showed power conversion efficiency (PCE) of 18.06 % (with short- circuit current density (J _sc ) = 22.33 mA cm ^-2 , open-circuit voltage (V _oc ) = 1.01 V, and fill factor (FF) = 0.79) while CsPbBr ₃ -CsFA and KPb ₂ Br5-CsFA devices showed PCE of 18.44 % (with J _sc = 22.53 mA cm" ² , V _oc = 1.08 V, and FF = 0.77) and 18.94 % PCE (with Jsc = 22.56 mA cm" ² , V _oc = 1.09 V, and FF = 0.77), respectively (for small device with active area 0.09 cm ² ).

Perovskite solar cells with functional additives showed significant improvement in PCE from 18.06 % (reference cell CsFA without any additive) to 18.94 % for KPb ₂ Br ₅ -CsFA devices.

There was a comparable J _sc increase with the addition of CsPbBr? and KPb ₂ Br5 in CsFA from 22.33 mA cm" ² (reference CsFA) to 22.53 mA cm" ² and 22.56 mA cm" ² .

A drastic enhancement was observed in V _oc from 1.02 V (reference CsFA) to 1.08 V and 1.09 V for CsPbBr ₃ -CsFA and KPb ₂ Br5-CsFA, respectively.

Incident photon-to-electron conversion efficiency (IPCE) spectra and the integrated Jsc curves for CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br5-CsFA PSCs were shown in Fig. 15b.

The integrated short circuit current density of CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br5- CsFA perovskites was 21.24 mA cm" ² , 21.41 mA cm" ² and 21.59 mA cm" ² respectively, which was within ± 5 % from the short-circuit current density measured from the J~V curves.

Interestingly, as shown in Fig. 15c, CsPbBr ₃ -CsFA and KPb ₂ Br5-CsFA PSCs showed narrower PCE distribution in PCEs with higher reproducibility using slot-die coating technique compared to the reference CsFA perovskite devices.

The average PCE of CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br5-CsFA was 16.56 %, 17.90 % and 18.25 %, respectively, obtained from 30 devices for each composition. Furthermore, Fig. 16 depicted the mapping of PCE obtained from slot-die coated KPb2Br5-CsFA perovskite devices on 100 cm ² substrate together with the mapping of other photovoltaic parameters (PCE, J _sc , V _oc , and FF) of CsFA, CsPbBr?-CsFA and KPb2Br5-CsFA perovskite solar cells. 18.94 % was the highest efficiency for CsFA-based small area perovskite solar cells using the slot-die coating technique.

The hysteresis of the perovskite devices was monitored by scanning J-V curves in forward and reverse direction (Fig. 15d and Fig. 17). The hysteresis index of CsFA was 7.27 %, while for CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA devices it was less than 5 %. The hysteresis behaviour was typically attributed to charge accumulation at the interface. A negligible hysteresis in CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA devices indicated a significant reduction of the defect and/or traps at interface and grain boundaries. This was further validated by examining the electroluminescence (EL) of KPb ₂ Br ₅ -CsFA solar cell under different biases (Fig. 15e, 15g and 18). The EL spectra for CsFA were positioned at 764 nm and observed the peak shift of 4 nm with increasing bias from 1.8 to 2.7 V depicted in Fig. 18a, while the EL spectra for KPb2Br5-CsFA positioned at 765 nm do not experience any peak shift with increasing bias from 1 to 3 V at high electric field generation (higher than the V _oc of about 1.1 V). This was in good agreement with the suppressed charge accumulation and reduced ion migration at ETL/perovskite interface upon the addition of KPb ₂ Br ₅ additive. The PSCs fabricated from the cluster-assisted approach also exhibited stabilized efficiency under continuous illumination in a dry air environment over 5,000 seconds as represented in Fig. 15f.

Example 19 - Influence of Potassium Additive

After the demonstration of high efficiency from the additive approach, a deeper analysis was performed to understand the impact of potassium additive in obtaining superior optoelectronic property.

For the spin-coated small-area devices, the presence of potassium had mostly been reported at the grain boundaries. In the present disclosure, the perovskite crystallization was completely different from spin-coating and it was pertinent to further correlate the impact of the potassium additives on the perovskite formed by slot-die coating. As previously shown by EDX of the cross-section sample, the alkali cation K ⁺ is primarily distributed at the SnCb/pcrovskitc interface. The distribution of the alkali-based cluster additives in the resultant perovskite film could be further verified through an XPS depth profiling study on KPb ₂ Br ₅ -CsFA film. To generate the depth profile, high-resolution XPS spectra of Pb 4f, I 3d, Cs 3d, Br 3d, K 2p, N Is, C Is, O Is and Sn 3d were collected intermittently, followed by the etching process with Ar gas cluster ions (10 keV Ariooo ⁺ ) over an area of 2 mm x 2 mm. The evolution of the individual spectra as a function of etching cycles was presented in Fig. 19, while the compositional profile showing atomic % of the elements of interest as a function of etching time was shown in Fig. 20 (inset of the enlarged view from 500 s to 1,500 s). The corresponding survey spectra could be found in Fig. 21. The surface scan showed doublet peaks for Pb 4f core level, i.e., 138.6 eV and 143.4 eV corresponding to the spin-orbit components of Pb ²⁺ 4f?/2 and Pb ²⁺ 4fs/2, respectively, while the I 3ds/2 and I 3d?/2 core levels could be observed at 619.4 eV and 630.8 eV, respectively as shown in Fig. 22a and 22d. On the other hand, Cs 3d core level showed peaks at 725 eV and 738.9 eV, which were attributed to the 3ds/2 and 3d?/2 components of Cs ⁺ ; for the Br 3d core level, the spin-orbit components 3ds/2 and 3d?/2 at 68.6 eV and 69.7 eV also agreed well with other reports on mixed halide perovskites represented in Fig. 22e. The surface composition for Cs : Pb : I : Br was calculated to be 0.07 : 1 : 2.52 : 0.28, suggesting that the surface was slightly poor in Cs and Br. After the first etch, the ratio changed to 0.2 : 1 : 2.13 : 0.18, indicating some loss in the halide content after etching. A second set of asymmetrical peaks in the Pb 4f spectra were also observed after the etching. These peaks were observed at about 136.8 eV and 141.6 eV, which were attributed to the 4f?/2 and 413/2 components of metallic Pb (Pb°), and the intensity gradually increasing with respect to the Pb ²⁺ doublet peaks shown in Fig. 22b. The formation of the metallic species was likely due to preferential sputtering of the halide groups, as different moieties in the sample have different sputtering yield. The etching process could also result in the breakdown of the hybrid perovskite into the precursor materials on the surface. The contribution of Pb° in the composition calculation was factored out based on the deconvoluted Pb 4f spectra.

The chemical state or identity of the K ⁺ ions after perovskite formation was investigated afterwards. As can be seen in the depth profile, most elements could already be detected on the surface, while K, O and Sn were only observed after 600 to 800 seconds of etching. The decline in the perovskite signals and the rise in the substrate signals were less abrupt at the interface, likely caused by the large analysis area and the substrate roughness. The composition in the bulk of the thin film remained consistent albeit a gradual reduction in both C and N concentrations with depth, which could be associated with the loss of the organic components from the sputtering process, as shown in Fig. 22g and 22h. In addition, there was also a slight increase in the bromide content around the interface. In fact, the rate of loss for bromide and iodide was different - the hybrid perovskite film became more enriched in Br at the film/ substrate interface. The small increase in Br could be associated with the emergence of the K signal, which overall suggested that the presence of the KPb ₂ Br ₅ clusters at the bottom of the thin film. This could be clearly seen in Fig. 20b, in which K 2p signals could still be persistently detected despite the declining signals from the other major core levels (Pb 4f, I 3d and Cs 3d). The loss of halide with depth could be immediately seen from the decrease in the halide-to-lead ratio (Fig. 23). The stoichiometry of the synthesized KPb ₂ Br ₅ power and KBr was analyzed by XPS and related data were shown in Tables 5 and 6.

Table 5. Table of KPb ₂ Br ₅ XPS analysis for confirmation of stoichiometry.

TF-Corrected Peak Area RSF Peak/Area Ratio

K 2p 4120.78 1.466 2810.90 1.00 Pb 4f 50987.47 8.329 6121.68 2.18

Br 3d 13595.02 1.055 12886.27 4.58

Table 6. Table of KBr XPS analysis for confirmation of stoichiometry.

TF-Corrected Peak Area RSF Peak/Area Ratio

K 2p 19128.46 1.466 13048.06 1.00

Br 3d 13567.46 1.055 12860.15 0.99

To determine the chemical states of the K-based additive, the K 2p and Br 3d core levels detected during the depth profiling of the KPb2Br5-CsFA sample were compared with those from pristine KBr and KPb ₂ Br ₅ , as presented in Fig. 20c and 20d. In the hybrid perovskite film, the K 2p3/2 and K 2pi/2 peaks were observed at 293.6 eV and 296.3 eV, respectively, while in the other alkali bromides, similar peaks appeared at slightly lower energy, i.e., 292.9 eV and 293.0 eV for the K 2p3/2 components of KBr and KPb ₂ Br ₅ , respectively. The deconvoluted fitted spectra of KBr and KPb ₂ Br ₅ for K 2p, Pb 4f and Br 3d were represented in Fig. 24. The corelevel shift could be associated with the different chemical environments experienced by the K atoms in the different compounds. Likewise, for Br 3d core level there was a slight variation among the compounds - Br 3ds/2 increased from 68.3, 68.48 to 68.6 eV for KPb ₂ Br ₅ , KBr to KPb ₂ Br ₅ -CsFA, respectively. The difference in the peak positions suggested that the K-based additive was unlikely to be present as either KBr or KPb ₂ Br ₅ . It was hypothesized that the added KPb ₂ Br ₅ could have been integrated into the structure of CsFA. It was also worth mentioning that throughout the course of the depth profiling, the Br 3d core levels remained unaltered in terms of the peak positions, which suggested that there was little change in terms of the chemical environment of the Br atoms on the surface, in the bulk of the film, as well as at the film-substrate interface. In summary, clear evidence for the distribution of alkali cation K ⁺ at the SnC>2/perovskite interface was obtained.

NMR spectroscopy was then carried out to probe the existence of potassium-based additive in the perovskite powder. The ²⁰⁷ Pb NMR spectra of the KPb ₂ Br ₅ -CsFA sample were compared to the CsFA in Fig. 20e. Both appeared with a broad resonance centered around 1,450 ppm, which was between the reported positions of FAPbL (1,515 ppm) and FAPbBn (515 ppm). This broadened resonance confirmed that the samples had formed a mixed cation/halogen solid solution perovskite, rather than separate phases, with the resonance appearing closer to FAPbL due to the higher concentration of FA and I. The similarity between the additive-treated and control spectra confirmed that the incorporation of KPb ₂ Br ₅ did not disturb the crystal structure of CsFA perovskite. Fig. 20f showed ²⁰⁷ Pb NMR spectra (at a lower frequency) of the KPb ₂ Br ₅ -CsFA perovskite in comparison to pure KPb ₂ Br ₅ . The KPb ₂ Br ₅ spectra presented with a singular sharp resonance at -280 ppm, accompanied by a spinning sideband manifold resulting from the large Chemical Shift Anisotropy (CSA) broadening present in the sample. A simulation of the CSA sideband manifold is provided in Fig. 25. By having a much longer ²⁰⁷ Pb scan at the frequency observed for KPb ₂ Br ₅ , an additional resonance was detected in the doped sample around -190 ppm. This resonance was hypothesized to be the small percentage of KPb ₂ Br ₅ additive in the KPb ₂ Br ₅ -CsFA perovskite sample. The resonance was more amorphous, and shifted to a higher frequency, than the pure KPb ₂ Br ₅ resonance observed, due to the nano-sized KPb ₂ Br ₅ particles. The inherent distortion to the crystalline lattice in small nanoparticles results in these changes to the NMR line shape, as has been shown in previous NMR studies of metallic nanoparticles. The ¹³³ Cs NMR spectra of the KPb ₂ Br ₅ -CsFA sample were compared to the control sample in Fig. S23a and the ⁸¹ Br NMR spectra of the KPb ₂ Br ₅ -CsFA sample in comparison to pure KBr were shown in Fig. 26b. In agreement with XPS data discussed previously, NMR analysis also probed the existence of potassium additivebased cluster in the KPb ₂ Br ₅ -CsFA perovskite sample.

Example 20 - Preparing Films by Slot Die Coating

Homogenous coating of large-area perovskite layer was critical as it was necessary for reproducibility in high throughput manufacturing. Fig. 27a illustrated the slot die coating of perovskite thin film in this work. During the perovskite coating, a nitrogen gas knife was introduced to accelerate the quenching of perovskite at low temperature. The optimization of perovskite film by slot-die coating was explained in detail in Fig. 28. The intermediate temperature between 50 °C to 60 °C would result in compact and homogenous surface film coverage due to fast crystal growth at the liquid-solid interface (Fig. 27b). Thickness mapping of the slot-die coated perovskite film on 100 cm ² substrate is illustrated in Fig. 29a. The thickness of the slot-die coated perovskite film was in the range of 540-580 nm and its average thickness on 100 cm ² substrate was 558 nm. The uniformity of the slot-die coated perovskite film was further confirmed by measuring optical absorption at 9 different areas on 100 cm ² substrate as shown in Fig. 27c. The absorption spectra overlapped with each other which indicated the consistency and homogeneity of the slot-die coated perovskite film over 100 cm ² . This also agreed with the hyperspectral mapping collected from 15 different spots (Fig. 29b). Perovskite modules on 100 cm ² conductive glass substrates were then fabricated with this optimized slot-die coated film in the same device stack as the small cell discussed earlier.

For the module fabrication, Pl and P2 etching was carried out by laser process while P3 etching was done manually by utilizing a 500-pm tape. The overall schematic representation of Pl, P2 and P3 etching was illustrated in Fig. 30. The perovskite module contained 13 cells that were connected in series. The I-V curve for the best performing KPb ₂ Br ₅ -CsFA module (active area of 57.5 cm ² ) was given in Fig. 27e showing module PCE of 16.22 % (V _O c 14.23 V, I _sc 0.100 A and FF of 0.64). The I-V curves of CsFA and CsPbBr ₃ -CsFA perovskite module were depicted in Fig. 31, showing module PCE of 14.34 % and 15.85 %, respectively. In addition, perovskite modules with CsPbBr? and KPb ₂ Br ₅ additives displayed a narrower distribution of conversion efficiency, as compared to the pristine CsFA perovskite module (Fig. 32).

Average PCE of CsFA, CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA modules were 14.21 %, 15.48 %, and 15.90 % respectively, obtained from 5 modules for each perovskite. Detailed photovoltaic parameters of perovskite modules were summarized in Table 7. Table 7. Photovoltaic parameters of slot-die coated perovskite solar modules. a Champion device efficiency; ^b Average reverse scan efficiency of 5 modules.

The incorporation of KPb ₂ Br ₅ additive not only improved the photovoltaic performance but also resulted in large-area devices with higher ambient and thermal stability under different testing conditions. The high stability of module could be directly related to improved grain size and reduced interfacial defects in additive based perovskite. Comparisons of the long-term stability between CsFA and KPb ₂ Br ₅ -CsFA perovskite solar cells under 30 % RH, 65 °C and 85 °C were given in

Fig. 33a to 33c and the detailed photovoltaic parameters are shown in Fig. 34 to 37.

For monitoring the thermal stability at 85 °C, perovskite solar cells were encapsulated using PIB based (PVS101) polymer encapsulant (Fig. 33c inset), while for 65 °C the devices were kept unencapsulated. Fig. 33a displayed the ambient stability for CsFA, and CsPbBr ₃ -CsFA and KPb ₂ Br ₅ -CsFA by periodically measuring the photovoltaic parameter of devices which were stored in dark with 30 % RH at room temperature, following the ISOS-D-1 protocol. KPb ₂ Br ₅ -CsFA additive based perovskite solar cell showed outstanding stability, retaining about 72 % of its initial PCE after 426 days (10,224 hours) in 30 % RH. Meanwhile, the CsFA devices showed 32 % of retention in PCE under the same testing condition. Under 65 °C and 85 °C thermal stability testing, KPb2Br5-CsFA devices still outperformed the CsFA devices, retaining about 81 % of its initial PCE after 1,152 hours at 65 °C and about 53 % of its initial PCE after 960 hours at 85 °C. On the other hand, CsFA devices could only sustain 64 % and 44 % of their initial efficiency at 65 °C and 85 °C, respectively. KPb2Br5-CsFA additive based perovskite solar module stability was tested under 30% RH and showed outstanding stability, retaining about 82 % of its initial PCE after 200 days (4,800 hours). The CsFA and KPb2Br5-CsFA perovskite solar module retained about 50 % of its initial PCE after 206 hours and 560 hours, respectively, without encapsulation under continuous illumination in the dry air environment represented in Fig. 33e. The related photovoltaic normalized data were shown in Fig. 38.

Example 21 - Ageing of Perovskite Ink

The aging study of the precursor solution was carried out by monitoring the dynamic light scattering (DES) profiles of the inks as they aged over a duration of two days. Relatively more dynamic ink transformation was observed in the case of CsPbBr?- CsFA sample where different peaks present were identified approximately at 1.5 nm and 280 nm for the freshly prepared ink; 1.4 nm, 700 nm, and 5500 nm for the 24 hour-aged ink; and only 1.5 nm only for a 2 day-aged precursor shown in Fig. 9. Meanwhile, for KPb ₂ Br ₅ -CsFA samples, the presence of both small and large size particles could be observed throughout the ageing period with the intensity of large particle size intensity being the highest upon 1 day ageing time.

It was postulated that the formation of large aggregates of nanoclusters was useful to act as nucleation sites during crystallization process. The postulation was corroborated by the fact that the photoluminescence intensity of KPb ₂ Br ₅ -CsFA film was the highest after aging the corresponding solution for 1 day (Fig. 10a and 10b), despite the absorption spectra not showing any shift in terms of position or intensity with solution ageing. This suggested the formation of perovskite films with an improved defect profile or characteristic.

A deeper analysis of film was carried out by taking the FESEM images of the KPb2Br5-CsFA films (see Fig. 11). The perovskite mean grain size initially increased to 1 day and later decreased with time. The average mean grain sizes at different interval ageing times were 252 ± 14 nm, 327 ± 25 nm, 260 ± 22 nm and 252 ± 20 nm, for 0 day, 1 day, 2 days and 7 days, respectively (the substrate of this experiment was glass, thus the overall mean grain sizes were smaller than those in Example 17, where the substrate was FTO layer). As such, while grain size was observed to be large for KPb ₂ Br ₅ -CsFA films after ageing the solution for 1-day, perovskite film fabricated from 7-day aged solution showed the high number of small grains with few large ones. In addition, the film for 1 -day-aged solution showed a more compact, large grain and homogenous film in comparison to others. XRD analysis was then performed to compare the crystallinity of different perovskite films. The ensuing XRD patterns were shown in Fig. 12. After ageing the solution for 1 day, the crystallinity of KPb2Br5-CsFA perovskite (for (Oil) plane) was found to improve 2.73 times compared to that of 0 hour, but after ageing the solution for 2 days and 7 days the crystallinity of the corresponding perovskite decreased.

Consistent with the DLS, PL, FESEM, and XRD data discussed previously, we found that ageing the precursor solution for 1 day gave us solar cell devices with the best power conversion efficiency and narrow distribution, as depicted in Fig. 13 and 14. In particular, the average of CsPbBr ₃ -CsFA perovskite solar devices were 16.97 %, 18.15 % and 17.83% for 0-day, 1 day, 2 days of ageing. Meanwhile, the average of KPb ₂ Br ₅ -CsFA perovskite solar devices at different ageing interval of 0-day, 1 day, 2 days and 7 days were 16.93 %, 18.46 %, 17.83 %, and 16.77 %, respectively. The decrease in efficiency was mainly due to decrease in voltage and fill factor as the time ageing time increases. The above observation was attributed to the nonoptimal perovskite nucleation and crystallization processes as the colloidal particles seed undergo changes upon ageing.

Example 22 - Analysis of NMR Results

The ¹³³ Cs NMR spectra of the doped sample were compared to the control sample in Fig. 18a. Like the ²⁰⁷ Pb data, the ¹³³ Cs spectra appeared identical with a single broad resonance around 148 ppm. This appeared between the resonances for CsPbL/CsPbBrS (167/120 ppm) corroborating that a mixed cation/halogen phase has formed. The effect of doping Cs into organic-cation Pb halides had been observed previously, and a shift to lower frequency was also observed in the ¹³³ Cs NMR in those case. For further confirmation that the Cs and FA cations were in close proximity in a mixed phase, a ¹³³ Cs[ ¹ H] cross polarisation experiment was performed on the doped sample. In cross polarization experiments the NMR polarisation was transferred from ^T H nuclei to nearby ¹³³ Cs nuclei, thus a signal should only be observed if the nuclei were in close proximity. The 148-ppm resonance was clearly still present in the CP spectrum demonstrating the mixed phase of Cs and FA. Additionally, the ⁸¹ Br NMR spectra of the doped sample in comparison to pure KBr was shown in Fig. 18b. The lack of any ⁸¹ Br signal at the KBr position of 61 ppm in the doped sample spectrum confirmed that KPb ₂ Br ₅ nanoclusters were forming, rather than breaking down into KBr.

Example 23 - Formation of Perovskite Films

Without providing the nitrogen gas quenching process, perovskite phase could not be crystallized properly in which yellow colored film was obtained with poor surface coverage (see Fig. 19a). Nonetheless, if the temperature of substrate was in the range of 25°C to 50 °C, the rate of solvent evaporation was insufficient for crystallizing perovskite film, leaving it as a wet film. Coating process at a temperature higher than 60°C would induce fast solvent evaporation which would trigger the heterogeneous nucleation and non-uniform quenching that led to poor film coverage. The intermediate temperature between 50 °C to 60 °C would result in compact and homogenous surface film coverage due to fast crystal growth at liquid-solid interface (Fig. 27b). Providing intense nitrogen gas flow (140 L/min) through air knife to the as-coated wet film generated a temperature gradient within the wet film and allowed homogenous crystallization process which led to the flattened surface. The gap between the coater and substrate was crucial in adjusting the thickness of perovskite layer. The coating gap was hence adjusted not less than 100 pm to deposit a perovskite layer with at least 500 nm thick film for ensuring sufficient light harvesting. The optimized coating gap for our perovskite layer was 300 pm, which resulted in a highly crystalline and uniform perovskite film with reflective surface. Eventually, the quenched perovskite film was subsequently annealed at 100 °C for 15 minutes under a slightly dry condition (30 % RH) for complete conversion to perovskite film.

For perovskite solar cell, more than 500 nm of perovskite film was sufficient for efficient charge extraction. Therefore, the coating gap between the substrate and blade was optimized to be 100 pm, and the solution concentration was optimized to be 1.2 M to obtain more than 500 nm perovskite observer layer. Fig. 19a showed the perovskite wet film without air knife quenching. The gap between the air knife and the substrate was further adjusted with different distance, namely 200 pm, 300 pm and 500 pm and the film obtained was shown in Fig. 19b to 19d respectively. When the distance was 200 pm, the film was too thick in the centre and quite rough in nature. When the distance was 300 pm, the quenching of perovskite did not take place sufficiently. 300 pm was the optimized distance for the present disclosure and showed highly uniform and crystalline perovskite film with high reflection in nature as represented in Fig. 19c and 19e. As anticipated, the nitrogen flow of 140 E/min and 300 pm distance resulted in smooth perovskite with full coverage evident by SEM image. Eater, a 10 cm x 30 cm FTO substrate was slot die coated to show the homogeneity of perovskite film as shown in Fig. 19f.

Summary of Examples

The present application discloses the role of alkali-based additives (that is, the compound of formula la) in the crystallization of large-area perovskite films and have examined their effects on optoelectronic properties and photovoltaic performances. The present application also discloses that the lower solubility of alkali cations in organic solvents, compared to the control, induces the formation of clusters in the precursor ink solution which could serve as nucleation centres for forming uniform perovskite films over a large area. The presence of cluster seeds was scrutinized by using dynamic light scattering (DES), transmission electron microscope (TEM), solid-state nuclear magnetic resonance (ss-NMR) spectroscopy and X-ray photoelectron spectroscopy (XPS). This additive approach not only led to the homogeneous and compact coating of perovskite film but also improved the quality of the resultant film. The additives served as seeds for perovskite growth and resulted in larger mean grain perovskite with better photovoltaic properties (as compared to films containing perovskite precursors that did not contain the compound of formula la or compared to a nonaged sample). As a result, perovskite solar cells prepared from this additive approach delivered a champion efficiency of 18.94 % (0.09 cm ² active area) with negligible hysteresis and average PCE of 18.06 %. The mapping of PCE over 100 cm ² substrates indicated a high homogeneity in perovskite film coating, which was a good factor for highly reproducible device fabrication. The perovskite solar module fabricated via this approach showed a champion efficiency of 16.22 % with an active area of 57.5 cm .

Moreover, the resulting small-area devices showed a significantly enhanced operational stability compared to additive-free controls, retaining about 72 % of its initial PCE after 426 days (10,224 hours) in 30 % RH and retaining 82 % and 54 % of initial PCE after 1,150 hours and 958 hours at 65 °C and 85 °C.

Furthermore, KPb2Br5-CsFA additive based perovskite minimodule retained about 82 % of its initial PCE after 200 days (4,800 hours) under 30% RH and the perovskite solar module retains about 50 % of its initial PCE after 560 hours without encapsulation under continuous illumination in the dry air environment.

The present application demonstrates that such a controlled “bottom-up crystallization” approach for perovskite film formation would offer great insight for scalable and reproducible fabrication process, which is the key for industrializing perovskite photovoltaic technology.

Industrial Applicability

The perovskite ink may be made into solar cells that can be used in a variety of applications such as biosensors, portable electronic devices, smart home appliances, aerospace energy supply, etc.

It will be apparent that various other modifications and adaptations of the invention will be apparent to the person skilled in the art after reading the foregoing disclosure without departing from the spirit and scope of the invention and it is intended that all such modifications and adaptations come within the scope of the appended claims.