Thin Film Solution Processable Perovskite Solar Cell

Abstract

Perovskite has emerged as a promising light-harvesting material for solar cells due to its higher absorption coefficient, bandgap tunability, low-exciton binding energy, and long carrier diffusion length. These lead to high power conversion efficiency >25% for thin film-based perovskite solar cells (PSCs). Additionally, PSCs can be fabricated through simple and cost-effective solution processable techniques, which make this technology more advantageous over the current photovoltaic technologies. Several solution-processable methods have been developed for fabrication of PSCs. In this chapter, the advantages and disadvantages of various solution processable techniques and their scope for large-scale commercialization will be discussed.

1. Introduction

Solar cell technologies have grown in the past few decades across four different generations. The first generation consisted of wafer-based photo active layer which was dominated by silicon wafer and is continuing to conquer the photovoltaic market. However, its high energy and cost of production has allowed the thin film technologies to gain attention among the research community. In the second-generation thin film based inorganic materials are being utilised to develop solar cells, but the efficiency has not reached the first-generation solar technology and the material production cost is also on the higher side. To reduce the cost of production further, organic and hybrid materials are being used in the third generation. The main advantage of this generation is that it allows photoactive layers to be deposited using low-cost solution processable techniques including spin coating, dip coating, etc. [1]. Further, large area fabrication is also facilitated by techniques such as doctor blading, inkjet printing, etc. [2, 3].

This advantage has also been strategically utilised in fabrication of perovskite solar cell which is the most promising and growing solar cell technology [4, 5, 6]. These techniques facilitate quick deposition of perovskite on any substrate at low temperature processing. Further, the solution processing techniques offer added advantage over the well-known thermal evaporation methods which requires complex vacuum systems. It has been also observed that the high-performance perovskite solar cells (PSCs) are generally fabricated using one of the solution processing techniques. This chapter compiles all solution-processing techniques that are being utilised for the fabrication of perovskite solar cell. Each technique has been explained in details covering the advantages and disadvantages as well as its way in controlling the crystallinity of the deposited perovskite film for photovoltaic application.

2. Solution processable techniques

2.1 Anti-solvent dripping: one step deposition technique

Anti-solvent dripping is sub part of spin coating process. Spin coating is a batch process that spreads a liquid film onto a rotating substrate using centrifugal force. This spin coating technique is categorised into two types: (a) one-step process and (b) two-step process. This method is successfully employed to develop a small area and large area PSCs of 0.1 cm² and 1 cm², respectively. From the above-mentioned techniques, the solution processable one step deposition (OSD) technique is widely accepted to develop PSCs from the laboratory to the industrial level. The formation of perovskite in the one-step process involves two stages: (a) evaporation of excess solvent in the active layer and (b) crystallisation of the active layer [7]. Its popularity stems from its ease of use and low cost of equipment. However, like any other method, this technique suffers a significant drawback. This is primarily due to a lower substrate coverage area and the formation of a rough and porous surface at interface layer of the perovskite solar cell. This method relies on uniformity of film thickness and morphology control to achieve a desirable film quality. This method also faces the challenge of reducing pinholes in the perovskite film. This has a drastic impact on the optoelectronic properties of the PSCs resulting in poor performance [8].

2.1.1 Antisolvent in chemistry

In chemistry, antisolvent precipitation is a well-known method of crystallising a substance. In Figure 1a the antisolvent precipitation is illustrated. The unique part of the antisolvent method is its applicability for the manufacture of PSCs. In this method, a non-dissolving liquid, or anti-solvent, is dropped onto a spinning substrate containing a perovskite solution to quickly remove specific solvents like DMF and GBL (Figure 1b) [9]. The treatment causes rapid nucleation in the film and converts it to a homogeneous intermediate film. Annealing the substrate then results in smooth perovskite film.

Figure 1.

(a) Schematic representation of antisolvent assisted precipitation/or crystallisation. (b) Simple representation of anti-solvent treatment for the OSD technique.

2.1.2 Antisolvent in perovskite

To address all of the aforementioned issues in OSD, the Antisolvent dripping (ASD) method is used to control crystal growth kinetics and film quality. A dense layer of larger grain size (100–500 nm) CH₃NH₃PbI₃ crystals is observed after successful ASD treatment. The significant formation of CH₃NH₃PbI₃ crystals by ASD method has catapulted perovskite research into a whole new realm. For the preparation of high-quality Pb-based perovskite crystalline films, anti-solvents such as benzene, toluene, ethanol, methanol, acetonitrile, benzonitrile chloroform, isopropyl alcohol, ethylene glycol, and chlorobenzene have been utilised [10, 11]. This method was first reported by Jeon and his team, who discovered that using an antisolvent in the fabrication of perovskite films resulted in better-quality, dense films with big grain size. The comparison of the various perovskite materials performance with different antisolvents have been shown in the Table 1.

Perovskite material	Anti-solvent	Antisolvent volume (μL)	Post annealing condition	Device structure	PCE, (%)	Ref.
MAPbI3	Toluene	130	80°C, 10 min	FTO/NiOx/perovskite/PCBM/Rhodamine/Au	21.18	[4]
MAPbI3	Di-isopropyl ether	1000	100°C, 10 min	FTO/c-TiO2/mp-TiO2/perovskite/Spiro-MeOTAD/Ag	19.07	[12]
MAPbBr3	Toluene	60	100°C, 10 min	FTO/c-TiO2/mp-TiO2/perovskite/Spiro-MeOTAD/Au	7.54	[13]
FASnI3	Diethyl ether	—	70°C, 20 min	ITO/PEDOT:PSS/perovskite/C60/BCP/Ag	5.41 ± 0.46	[14]
(CH3NH3)3Sb2I9	Chlorobenzene	300	70°C, 20 min	ITO/PEDOT:PSS/perovskite/PCBM/C60/BCP/Al	2.7	[15]
Cs5(MA0.17FA0.83)95; Pb(I0.83Br0.17)	Diethyl ether	800	Post annealing free	FTO/bl-TiO2/mp-TiO2/perovskite/Spiro-MeOTAD/Au	19.5	[16]
(FAPbI3)0.85(MAPbBr3)0.15	Trifluorotoluene	110	100°C, 90 min	FTO/bl-TiO2/mp-TiO2/perovskite/Spiro-MeOTAD/Au	20.3	[17]
FA0.75MA0.25SnI3	Toluene	—	100°C, 10 min	ITO/PEDOT:PSS/perovskite/C60/BCP/Al	6.36 ± 0.64	[18]
FA0.75MA0.25SnI3	Chlorobenzene	350	45°C, 10 min; then 65°C,20 min; and 100°C, 10 min	ITO/PEDOT:PSS/perovskite/C60/BCP/Ag	7	[19]

Table 1.

PCE comparison of various PSCs using different antisolvents.

The ASD method can significantly alter the morphology of the MAPbI₃ film’s surface. During the deposition of the perovskite layer by the OSD method, many voids and pinholes were observed. On the other hand, the ASD method reveals lesser pinholes with large grains, densely packed MAPbI₃ crystals due to its smooth and homogeneous surface morphology [12]. However, the anti-solvent preparation process must be done at the correct time and in the correct quantity as well as with a high level of proficiency. Uncontrolled crystallisation can also result in pinholes and higher defect density, which can reduce device efficiency and stability.

The performance of PSCs and their reproducibility is significantly improved when the MAPbI₃ film is of higher quality and covers the entire surface area by using ASD method. The addition of favourable additives has facilitated perovskite crystal growth to improve the morphology, stability, excitonic, and optoelectronic properties of hybrid inorganic-organic perovskite films. Although it has been discovered that the solution-processable technique is capable of producing the ideal perovskite film, the quality of the film may be compromised due to factors such as temperature, precursor solubility, atmosphere, and annealing time [20].

Zhou et al. used an antisolvent-solvent extraction process to study the crystallisation behaviour of mix halide perovskites at room temperature. A small amount of solvent diffuses within a large amount of antisolvent in this strategy. This antisolvent-solvent extraction method achieves supersaturation state and nucleation for the crystallisation procedure, and nucleation rate can be improved by magnetic stirring of antisolvent. The antisolvent DEE (anhydrous diethyl ether), and magnetic stirrers were used to introduce advection in the antisolvent bath. The antisolvent-solvent extraction process is a straightforward method for producing high-quality perovskite films with improved morphology [21].

Xiao et al. demonstrated a fast, single-step, solution-based deposition crystallisation technique that allows control over the dynamics of nucleation and grain growth of CH₃NH₃PbI₃, resulting in rapid and repeatable fabrication of high-quality perovskite thin films. In this method, a DMF solution of CH₃NH₃PbI₃ perovskite is spin-coated on a substrate, followed by a second solvent, such as chlorobenzene (CBZ), applied on top of the wet film during the spin coating process to induce fast crystallisation. The second solvent is important for lowering the solubility of CH₃NH₃PbI₃ and promoting crystal nucleation and growth within the thin film [22].

For inverted planar perovskite solar cells, Liu et al. reported effective and stable green mixed anti-solvent engineering. This green mixed anti-solvent technique can improve the surface morphology of perovskite films and passivate the grain boundary of perovskite thin films [23].

2.1.3 Modified antisolvent treatment

Modification, particularly during the anti-solvent treatment, is absolutely necessary to manage perovskite crystal growth in a humid environment and obtain a highly efficient device. Currently, the anti-solvent is dropped 15 s after the fast-spinning programme begins, with chlorobenzene, toluene, and diethyl ether being the most common anti-solvents used in PSC fabrication [24]. Determining an appropriate time delay for quenching anti-solvents is the most important part of anti-solvent treatment. The turbid point during the spin deposition stage must be identified in order to achieve this goal. During spin coating, the turbid point is the point at which the precursor film turns from transparent to turbid. The time delay increases as the RH level rises. This result is linked to the previously discussed solvent evaporation dynamics. According to Wang et al., the turbid point appears at approximately 9, 15, 19, and 20 s for RH0, RH50, RH70, and RH90%, respectively. In order to develop PSCs under high-humidity conditions, it appears that determining an accurate dripping time is an inconvenient procedure [25].

Thus, there are three possible strategies to avoid the antisolvent dripping time window: (a) creating an anti-solvent mixture by combining traditional non-polar solvents, such as diethyl ether and chlorobenzene, with other polar solvents (e.g., R-OH); (b) finding another anti-solvent that is suitable for high-humidity processing and completely substituting the commonly used anti-solvent; and (c) applying preheating treatment to the solution [26]. Anti-solvent mixes can be divided into two types: those with (1) a small amount of polar solvent and those with (2) a large amount of polar solvent. Because the anti-solvent contains a polar solvent, the adsorbed H₂O molecules on the perovskite intermediate layer can be dissolved and removed concurrently with DMF and excess DMSO. As a result of the direct contact between the electron transfer layer and the hole transfer layer, a homogenous perovskite intermediate is generated, which gradually converts into a smooth and pinhole-free film, limiting charge recombination [27].

Another viable strategy for forming a high-quality perovskite film in high humidity is to replace the standard anti-solvent, as previously described. For high-humidity PSC production (up to RH75%), Troughton et al. utilised ethyl acetate. This was due to the fact that ethyl acetate, unlike other anti-solvents like toluene, chlorobenzene, and diethyl ether can absorb a significant amount of moisture in the air. Because of these properties, ethyl acetate can absorb moisture from the air and prevent it from interacting with the intermediate phase of perovskite. As a result, regardless of the RH level, uniform and smooth films can be created. Another effective strategy for making the anti-solvent treatment humidity insensitive is to speed up the rate of solvent evaporation by pre-annealing the substrate, causing the turbid point to appear earlier. According to Wang et al., the anti-solvent can be applied to the perovskite film 2 s after the spinning protocol starts, regardless of the humidity level, to facilitate the fabrication of highly efficient PSCs when the substrate is pre-heated to 70°C [28].

2.2 Hot-casting

Hot-casting technology has emerged as an excellent tool for the deposition of high-quality perovskite thin films with notable benefits including rapid crystallisation, a quick film formation process, increased grain size, preferred crystal orientation, and low defect density [29]. Within the framework of nucleation growth theory, a direct formation mechanism is proposed that constitute a driving force for the phase change provided by a high substrate temperature leading to an ultrashort crystallisation process. Meanwhile, a sufficient thermal energy allows atoms to diffuse in a liquid without forming an intermediate phase. Nie et al. were the first to report this method, and they went on to improve the film quality by tweaking deposition parameters like substrate temperature, annealing temperature, and precursor composition. This technology has recently been expanded to include the deposition of organic–inorganic hybrid, all-inorganic, lead-free, and low dimensional perovskite films [30].

2.2.1 Fundamentals of nucleation and crystal growth

The goal of the hot-casting technology is to spin coat a hot precursor solution over a substrate at higher-temperature. The crystal growth is influenced by factors such as substrate temperature, solution concentration, solvent, and supersaturated environment. In Figure 2a, the LaMer diagram represents these deposition parameters correlated with the nucleation and growth processes [31].

Figure 2.

(a) Lar Mer model for nucleation and growth of perovskite thin films, (b) a schematic illustration of nucleation and growth of perovskite films at each stage.

2.2.2 Classical nucleation and classical growth

The Volmer-Weber model, the Frank-van der Merwe model, and the Stranski Krastanov model are three classic thin film nucleation and growth models [32, 33]. The Volmer-Weber model is valid for the nucleation and growth of most polycrystalline thin films if the substrate temperature is sufficiently high and the deposited atoms have a certain diffusion capability.

According to classical nucleation theory, to initiate the crystallisation, the solution must be supersaturated. Nucleation can be either heterogenous or homogenous. Heterogeneous nucleation is defined as nucleation with preferential nucleation sites, which means that new phases form preferentially within certain regions of a liquid phase. The system must overcome an energy barrier known as the maximum free energy of critical nucleation (G) during the thermodynamic nucleation process. Heterogeneous nucleation is defined as nucleation with preferential nucleation sites, which indicates that new phases form preferentially within some parts of a liquid phase. The system must surpass an energy barrier known as the maximal free energy of critical nucleation (G) during the thermodynamic nucleation process.

2.3 Temperature and thermal annealing

2.3.1 Direct formation mechanism

Nucleation growth is influenced by the substrate temperature and thermal annealing. When the substrate temperature is low (less than 100°C), the perovskite film formation process includes three steps- the initial solution stage, the transition-to-solid film stage, and the transformation stage from intermediates to a perovskite film [34]. However, when the substrate temperature is raised to 100–180°C, enough thermal energy is provided to speed reactant diffusion and contact, hence no intermediate phase development occurs during the hot-casting process leading to the direct formation of perovskite film.

2.3.2 Substrate temperature

The supersaturation of the solution is affected by the substrate temperature, which changes the nucleation rate and film shape. The high boiling solvent supports a stable development of the perovskite crystal with big crystal grains when the substrate temperature is higher than the crystallisation temperature of the perovskite phase. A large-grain perovskite layer improves device performance by lowering the defect density and boosting the mobility which suppresses charge trapping [35]. The effect of Substrate temperature on various perovskite material has been provided in the Table 2.

Perovskite material	Device structure	Temperature	Voc, (V)	Jsc, (mA/cm^2)	FF, (%)	PCE, %	Ref.
MAPbI3	FTO/c-TiO2/m-TiO2/Perovskite/spiro-OMeTAD/Ag	175	1.07	21.32	70	16.01	[36]
MAPbI3−xClx	FTO/PEDOT:PSS/perovskite/PCBM/Al	190	0.954	24.21	61	14.11	[37]
FA0.25MA0.75PbI3	FTO/c-TiO2/perovskite/spiro- OMeTAD/Au	240	0.97	22.6	66	14.6	[38]
FAPbI3−xCIx	FTO/c-TiO2/perovskite/spiro-OMeTAD/Ag	175	0.96	18.93	66	12.07	[39]
CsPbI2Br	ITO/SnO2/perovskite/PTAA/MoO3/Al	340	1.19	14.54	74	13.8	[40]
Ag2BiI5	FTO/c-TiO2/m-TiO2/perovskite/PTAA/Au	100	0.69	6.04	62	2.60	[41]
BA2MA3Pb4I13	FTO/PEDOT:PSS/perovskite/PCBM/Al	150	1.01	16.76	74	12.52	[42]
(CPEA)2MA4Pb5I16	FTO/c-TiO2/perovskite/spiro-OMeTAD/Au	120	0.99	19.92	60	11.86	[43]
MA2PbI4	FTO/c-TiO2/perovskite/spiro-OMeTAD/Au	150	1.06	21.00	76	16.92	[44]
(GA)(MA)3Pb3I10	FTO/c-TiO2/perovskite/spiro-OMeTAD/Au	100	1.00	20.70	66	13.87	[45]

Table 2.

Photovoltaic parameters of various hot-casted PSCs.

2.3.3 Thermal annealing

In hot-casting technique, Yang et al. described the Volmer-Weber growth mechanism for the creation of island-like grains and the transition to a compact perovskite film. Thermodynamic energy is important in accelerating perovskite crystallisation and precursor diffusion, which directly affects the film shape (Figure 2b) [46].

Thermal energy can lower the surface tension between the precursor solution and the substrate, increasing the probability of large grain domain with fewer surface defects. As the temperature rises, isolated like grains get larger and begin to form bonds with one another, finally generating a high-quality perovskite film with no pinholes. Thermal annealing not only increases material transport inside the film and facilitates solvent evaporation, it also improves charge extraction in the active layer [47]. Through quick solvent removal, flash infrared annealing (FIRA) can enhance the development and crystallisation of perovskite films. In comparison to the traditional thermal annealing, short heating pulses, compared to traditional thermal annealing, can greatly minimise the deterioration of organic components, even at extremely high temperatures [48]. The grain size distribution along the periphery of the perovskite film made by hot casting is greater than the core size distribution, and the particle size distribution is annular, as proposed by Ren et al. During the evaporation of the solvent, a compensatory flow from the centre of the solution to the edge is thought to occur spontaneously, resulting in a greater concentration and larger particle size near the edges.

2.4 Precursor chemistry

2.4.1 Additives

The wettability of the substrate and consequently the film formation is affected by the viscosity and solvation ability of the solvent. The rate of solution evaporation is determined by the solvent’s boiling point and vapour pressure. Therefore, a selection of extremely polar aprotic organic solvents is required for a weakly soluble inorganic lead salt. Lewis-base character of some highly polar aprotic organic solvents can cause a solvent-solute coordination, which can alter the perovskite crystallisation process. The colloidal skeleton of the perovskite precursor solution is made up of numerous coordination complexes, and it is viewed as a colloidal cluster with a soft colloidal skeleton [49, 50]. Additives (Cl⁻ and Br⁻) influence the size of the colloidal clusters. According to Liao et al., adding chlorine to perovskite films improves their optoelectronic capabilities and environmental resilience. The addition of 10% Cl⁻ to a MAPbI₃ precursor solution increases film uniformity and coverage greatly [51].

2.4.2 Ageing time and solvent

The size and shape of colloidal clusters, as well as the nucleation and growth process, can be affected by the ageing duration of the precursor solution. Mohite et al. found a substantial link between film crystallinity and ageing time. The crystallinity and grain size of perovskite films were greatly improved when the precursor solution was aged for more than 24 h, according to the findings. The precursor solution gradually develops big seeds (or crystals) as it matures. Grain development, phase purity, surface uniformity, and trap state density of the perovskite film have all been shown to be considerably affected by precursor ageing. Meanwhile, the phase development and crystalline characteristics of perovskite films are influenced by the solvents and content of the precursor. To deposit perovskite films, Wang et al. employed gamma-butyrolactone (GBL) and DMF as co-solvents [36]. The grain size and photovoltaic characteristics of the device produced by the GBL solvent were found to be much lower than those produced by the DMF solvent. Iyer et al. also used DMSO as a Lewis base adduct to control perovskite crystallisation and grain development [37].

2.4.3 Composition and other factors

Processing parameters such as substrate temperature, rotation speed, and thermal annealing can be used to fine-tune the perovskite film morphology. Through heat energy and centrifugal force, these processing factors influence solute diffusion and perovskite crystallisation behaviour. The residual organic residue is evaporated during thermal annealing. To create a high-quality perovskite layer, Moon et al. employed MAI and lead acetate (PbAc₂) as precursors. By using by-product gas (3MAI + PbAc₂ MAPbI₃ + 2MAAc) to remove the PbAc₂ residue, the crystal development can be accelerated to generate a fully covered, pinhole-free, and highly crystalline perovskite film [52]. Janssen et al. used a mixture of PbAc₂, PbI, and MAI to make a high-quality perovskite layer. In ambient condition, Huang et al. used methylammonium acetate (MAAc) as a general solvent to produce high-quality perovskite films. To facilitate solvent evaporation, a constant substrate temperature (100°C) was used, resulting in supersaturation and fast nucleation and crystal formation [53].

2.4.4 Atmosphere

The crystallinity and surface morphology of as-cast perovskite films are affected by various deposition circumstances. One benefit of hot-casting method is that the deposition is not affected by the processing environment. Therefore, the perovskite films may be produced in ambient air, and the device revealed great stability under high humidity [54]. Mori et al. combined a gas flow with hot-casting method to make MAPbI₃ films in ambient circumstances (relative humidity = 42–48%) [55]. The flowing gas can greatly expedite mass transfer and eliminate thickness non-uniformity due to the difference in centrifugal force between the centre and edge of the substrate. Yang et al. also used a combination of hot-casting and methylamine (MA) gas treatment to create dense and homogeneous perovskite films at high relative humidity. With MA gas treatment, porous and rough MAPbI₃ perovskite films made by hot casting can be turned into dense and high-quality films. In addition, Hao et al. used non-destructive ethanol/chlorobenzene to treat MAPbI₃ perovskite films, which resulted in coarsening of the perovskite grains and lateral grain expansion of the MAPbI₃ perovskite films. To overcome the challenges of temperature gradient and moisture intrusion during the deposition process, Cheng et al. developed a thermal radiation hot-casting method [56].

2.5 Advantages of hot-casting technique

2.5.1 Grain size, orientation, and film thickness

To create reasonably thick and preferentially oriented large-grain perovskite films, hot-casting process has been frequently used. The grain boundaries of the perovskite films are reduced due to the higher grain size. At the same time, the absorption, charge transport, and crystallinity of perovskite films are all positively affected. Increases in grain size, on the other hand, may increase the density of unwanted pinholes, resulting in direct contact between the HTL and ETL and leakage current [57]. The device’s performance will be severely harmed by the voids in the perovskite coating. A thick perovskite layer helps to gather enough light absorption across the visible light spectrum [58]. To tune the thickness of the perovskite layer, the concentration of the precursor solution and the rotation speed can be varied. The inorganic halide octahedrons are joined at a shared apex and stretched in a 2D direction to form 2D perovskites when a long-chain organic cation layer is placed into the inorganic framework to deviate the tolerance factor from a value of 1. The introduction of an organic chain will make charge extraction and collecting more difficult. Controlling the growth direction of a 2D perovskite film is critical for carrier transport. A hot-casting technique can produce a selectively oriented development of 2D perovskites [59, 60].

2.5.2 Defects and recombination

The performance of the PSCs is harmed by nonradiative recombination in the following ways. Radiative recombination occurs when electrons return to the valence band. Minority recombination occurs at the interface as holes (electrons) are transported back to the perovskite layer. The porous perovskite layer will then come into direct contact with the functional layers, resulting in carrier recombination. A nonradiative recombination process is caused by a number of defect states in the device forming recombination centres to trap the carriers. The major avenue for a carrier loss is through some deep-level traps. Furthermore, the nonradiative recombination of carriers within the device has a direct impact on the PSCs’ V_oc. As a result, forming low-defective perovskite thin films is critical for preventing nonradiative recombination. Large grain sizes help limit the number of grain boundaries in hot-casted perovskite films, which reduces charge trapping [30].

2.6 Efficiency and stability

2.6.1 Device performance

Three key limiting considerations for the actual deployment of PSCs are PCE, stability, and cost. Table 1 shows how PCEs have progressed in previous years while employing a hot-casting approach. The two most common light harvesters in PSCs are MAPbI₃ and FAPbI₃, which have energy bandgaps of 1.55 and 1.47 eV and theoretical maximum PCEs of 31.3 and 32.5%, respectively [61]. Nie et al. were the first to describe the manufacture of perovskite films with a millimetre grain size, resulting a PCE of 17.48% with no hysteresis. Marks et al. used a hot casting technique to manage Cl⁻ incorporation and reported a PCE of 18.2% for a small area (0.09 cm²) and 15.4% for a big area (0.09 cm²) (1 cm²). We believe that in the future, the PCE of PSCs produced using hot-casting technology will catch up to that produced using a solution approach. Meanwhile, the perovskite composition, device structure, and encapsulation all play a role in improving PSC long-term stability [62].

2.6.2 Stability

H₂O and O₂ have a direct impact on PSC device performance and stability throughout film deposition, device testing, characterisation, and storage. In the case of MAPbI₃, water vapour can dissolve the perovskite material, and MAI is dissolved to generate a combination of MA and HI; however, HI will either react with O₂ to form H₂O and I₂, or self-decompose. After being exposed to moisture, MAPbI₃ continues to degrade. The inability hinders the commercialization of PSC devices. Improving long-term stability requires adjusting the ABX₃ perovskite content and boosting crystalline quality [42, 63]. Long-chain organic cations have been routinely used as a site cation substitution in this regard. The organic and inorganic layers alternatively form a layered structure in these 2D perovskites, which have great long-term stability. As long-chain organic cations are hydrophobic, the bigger organic cations in the 2D perovskite crystal structure improve humidity stability. Pure 2D PSCs are stable, but have a low PCE. Combining 3D perovskite and 2D perovskite yields exceptional optoelectronic characteristics and stability. The intrinsic performance of the 2D perovskite deposited using a hot-casting approach can be maintained for a long time, showing greater stability in humid and other environmental conditions [64, 65].

2.7 Two-step coating

Mitzi et al. first introduces the two-step coating method in order to improve the morphology and quality of the active perovskite layer [66]. Coating is done in the first phase by using conventional spin coating process, and the second step uses other coating methods such as immersion and spin coating depending on the materials’ requirements.

2.7.1 Immersion method

The organic and inorganic components of the perovskite material are treated separately in the immersion method. The perovskite material’s inorganic salt (PbI₂) is first spin coated on the substrate at a particular RPM. The spin-coated PbI₂ film substrate is then immersed in an organic salt (MAI) precursor solution for a period of time. After taking the substrates from the precursor solution, rinse them using the same solvent and concentration that was used to prepare the organic precursor solution. This step is used to remove any surplus organic material from the substrate’s surface. Finally, the ultimate perovskite film is obtained by annealing the substrate for a few minutes at a specific temperature. The procedure is depicted schematically in Figure 3. Grätzel et al. used this strategy for the first time in 2013 to obtain MAPbI₃ film in order to optimise the morphology of active perovskite material for photovoltaic devices [67].

Figure 3.

Schematic diagram of immersion coating process of perovskite (MAPbI₃) film.

The concentration of inorganic compound in the spin coated film, the concentration of organic salt in the precursor solution, and the immersion duration all seem to have a significant impact on the growth of the perovskite film, which defines the morphology and quality of the ultimate perovskite film.

After immersion in the precursor solution, two distinct reaction pathways convert the spin coated film to the ultimate perovskite film. The first is the solid-liquid interface conversion reaction, which happens when the precursor solution concentration is low. Due to the low concentration of the MAI precursor solution, MAI tends to diffuse into the PbI₂ structure during immersion of the PbI₂ coated film. Finally, the reaction of MAI with the PbI₂ film at the interface produces the ultimate MAPbI₃ perovskite film. The conversion reaction at the solid-liquid interface is show by Eq. (1).

$$Pb{I}_2\left(s\right)+C{H}_{3}N{H_3}^{+}\left(\mathit{sol}\right)+I^{-}\left(\mathit{sol}\right)\to C{H}_{3}N{H}_{3}Pb{I}_3\left(s\right)$$E1

In the above reaction initially, MAI diffuses into the PbI₂ structure as well as the reaction start occurs at the interface of PbI₂. Once the perovskite crystal is formed at the surface of PbI₂ the diffusion of the organic cation MAI into the PbI₂ structure is at a standstill and the conversion reaction is completed [68].

When the concentration of organic cation is high enough (>10 mg/mL), another reaction called dissolution-recrystallization evolution mechanism occurs. As the concentration of organic cation in this process is high, it causes rapid crystallisation of perovskite on the surface of PbI₂. As a result, organic cation diffusion into the PbI₂ structure is totally stopped after a few times, as stated by Eq. (2). According to Yang et al., a high concentration of the organic cation MAI leads to production of the lead iodide complex PbI₄²⁻, as shown in the Eq. (3). The excess iodine in the system began to dissolve previously created MAPbI₃ crystals and PbI₂ components that had not been covered during the crystallisation process, and the reaction continued until the system could not reach thermal equilibrium [69].

$$C{H}_{3}N{H}_{3}Pb{I}_3\left(s\right)+I^{-}\left(\mathit{sol}\right)\to C{H}_{3}N{H_3}^{+}\left(\mathit{sol}\right)+Pb{I_4}^{2-}\left(\mathit{sol}\right)$$E2

$$Pb{I}_2\left(s\right)+2I^{-}\left(\mathit{sol}\right)\to Pb{I_4}^{2-}\left(\mathit{sol}\right)$$E3

Following these two reactions, when the concentration of excess PbI₄²⁻ approaches the supersaturation state, the gradual re-crystallisation of MAPbI₃ via the reaction mechanism shown in Eq. (4) begins anew.

$$Pb{I_4}^{2-}\left(\mathit{sol}\right)+C{H}_{3}N{H_3}^{+}\left(\mathit{sol}\right)\to C{H}_{3}N{H}_{3}Pb{I}_3\left(s\right)+I^{-}\left(\mathit{sol}\right)$$E4

2.7.2 Sequential spin coating method

The two-step spin coating method, also known as sequential spin coating, is a well-studied laboratory technique for perovskite thin film deposition. Im et al. presented this technology in 2014 to fabricate high efficiency (>16%) perovskite solar cells by growing cuboid perovskite grains of controlled size [70]. The organic precursor is spin coated above the inorganic layer at a certain spin-rpm in this technique, which starts with the inorganic part of the perovskite material being spin coated onto the substrate. The final perovskite thin film is obtained by periodically annealing the substrate, which exhibits the perovskite film’s growth by changing colour during the process. As long as the concentration of organic precursor is high enough, the dissolution crystallisation mechanism produces the resulting perovskite film. Figure 4 illustrates the perovskite film fabrication process using this technique.

Figure 4.

Schematic diagram of Sequential spin coating process of perovskite (MAPbI₃) film.

Huang et al. devised a two-step spin coating process to produce a perovskite film with fewer pinholes at low temperatures. The inorganic precursor is spin coated onto the substrate and dried on the hot plate in the first stage. After that, the inorganic film substrate is spin coated with the organic precursor solution. The solvent used to dissolve the organic component of the perovskite material should have a low solubility of the inorganic part in this approach, such that no inorganic part washed out during the organic precursor coating. Due to the inter-diffusion of spin coated organic and inorganic assembly layers, the substrate ultimate compact, pin-hole free perovskite film is achieved after thermal annealing [71].

Panzer et al. revealed the reaction mechanism for obtaining the ultimate perovskite layer using a two-step spin coating method. According to their findings, the reaction mechanism can be broken down into five steps: perovskite capping layer formation, change in organic layer concentration during solvent evaporation, initial caping layer dissolution, rapid dissolution recrystallization, and complete conversion of residual PbI₂ to perovskite crystal. The perovskite capping layer begins to form immediately in the second step of the spin coating process, i.e., on the surface of PbI₂ after the introduction of MAI, which initiates the formation of the MAPbI₃ crystal. The size of perovskite grains and the time it takes to produce them both decrease as the MAI concentration rises [72]. Furthermore, the reaction temperature has a major impact on the production of perovskite crystals and their grain sizes. Solvent evaporation occurs in the organic precursor during thermal annealing. During this method, a dense perovskite crystal layer is first generated on top of the PbI₂ layer to prevent MAI from penetrating the PbI₂ structure and preventing further perovskite crystal formation in the system [72]. The concentration of MAI solution continues to rise due to the suppression of MAI solution penetration onto PbI₂ and the volatile nature of its precursor solvent. As the concentration of MAI rises, so does the concentration of iodine ions, which react with the perovskite layer on the surface of PbI₂ to generate complex lead iodine PbI₄²⁻ as a by-product. The degradation of the previously produced perovskite layer occurs primarily at grain boundaries and smaller grains in this process [72]. Finally, when the concentration of volatile PbI₄²⁻ approaches the supersaturation point it start recrystallisation of perovskite grain with all the uncovered inorganic component PbI₂ converted into MAPbI₃ crystal.

Two-step coating, also known as sequential deposition, has a number of advantages, including increased processing window flexibility. Solvent engineering, concentration variation, annealing time, annealing temperature adjustment, spin rpm variation etc. can be used to optimise each deposition stage separately, resulting in a superior morphological perovskite film. Furthermore, depending on the solubility of the additive, different additives can be doped directly into the precursor solution to form a defect-free perovskite layer. These approaches also produce a high-quality perovskite film with great reproducibility. When comparing the performance of solar cell devices, sequential deposition delivers a comparable efficiency to the single-step deposition procedure.

Aside from these benefits, the two-step deposition process has a few disadvantages. The formation of a perovskite layer in this approach relies heavily on molecular exchange. The organic component is frequently overreacted in the second step of the deposition technique, making it difficult to precisely control the chemical composition of the film [73]. Another issue with this approach is incomplete inorganic compound conversion to perovskite crystal, which affects overall molecular exchange with the organic component. Table 3 shows some of the best results obtained using these two-step coating methods.

Methods	Device architecture	Jsc, (mA/cm^2)	Voc, (V)	FF, (%)	PCE, (%)	Ref.
Immersion	FTO/c-TiO2/MAPbI3/Spiro-OMeTAD/Ag	22.49	1.08	67	16.21	[67]
	FTO/c-TiO2/m-TiO2/MAPbI3/Spiro-OMeTAD/Au	19.4	1.01	65.8	12.9	[74]
	FTO/TiO2/CsPbBr3/Carbon	5.99	1.33	65.7	5.25	[75]
	TCO/c-TiO2/m-TiO2/MAPbI3/Spiro-OMeTAD/Au	20	0.993	73	15	[76]
	FTO/c-TiO2/m-TiO2/MAPbI3/Spiro-OMeTAD/Au	19.42	1.03	66	13.2	[77]
Sequential spin coating	ITO/SnO2/CsFAMAPbI2Br/PCBM/Ag	22.11	1.13	81.53	20.35	[78]
	ITO/c-TiO2/FAPbI3/PCBM/Au	24.03	1.08	72	18.77	[79]
	ITO/PTAA/MAPbI3/PCBM/Al	23.41	1.05	78	19.5	[80]
	ITO/TiO2/FAPbI3/Spiro-OMeTAD/Au	24.3	1.12	81.1	22.1	[81]
	FTO/c-TiO2/m-TiO2/FAxMA1−xPbI2.55 Br0.45/Spiro-OMeTAD/Au	23.72	1.075	78.8	20.11	[82]

Table 3.

Photovoltaic parameters of different perovskite material using two step deposition method.

The methods described above are only suitable for making solar cells on a laboratory scale because a large amount of precursor solution is wasted during the spinning process, and the deposited film does not provide uniformity across the entire area of the film, resulting in device performance degradation.

3. Techniques for large area device fabrication

The performance of PSCs in large area modules must be maintained for practical implementation in the industrial environment. The methods described above are exclusively used to create small-scale laboratory devices. The performance of a perovskite solar cell declines as the active area grows. With the scaling up of device area, the PCE value reduces by 0.8% in most large-scale devices [83]. The PCE is lost due to an increase in series resistance caused by the huge transparent substrate’s resistance. Furthermore, when the active area of the device grows, it becomes more difficult to maintain homogeneity across all layers, which has a significant impact on its PCE. The scaling up of perovskite solar cells with morphology manipulation for high quality pin hole free perovskite layer is the most significant and challenging of all the deposition layers.

Several attempts have been made to scale up PSCs towards commercialization. For the fabrication of large area devices, numerous solution-based scalable deposition processes have been developed that can retain overall good device performance. Blade coating, slot die coating, bar coating, spray coating, and inkjet printing, screen printing are examples of these processes.

3.1 Blade coating

For the printing of large area devices, the blade coating process is a relatively simple and inexpensive technique. The blade is utilised to spread the precursor solution throughout the substrate. A micrometre screw is used in the system to adjust the distance between the blade and the substrate by rotating it, allowing the thickness, homogeneity, and crystallinity of the deposited layer to be determined. This approach can be utilised to deposit not just the active perovskite layer, but also other charge transport layers [84]. Figure 5(a) illustrates a schematic illustration of the deposition procedure. This approach allows for a slower drying of the wet film, allowing for a broader coverage and higher-quality perovskite film. The creation of a bigger grain perovskite coating as a result of the slow solvent drying process enhances the carrier diffusion length and thus the device’s PCE [85]. In the film creation process of these technologies, two significant regimes exist: the evaporation regime and the Landau-Levich regime, which are depicted in Figure 5(b) and (c), respectively.

Figure 5.

(a) Schematic diagram of blade coating process, (b) schematic diagram of evaporation regime, (c) schematic diagram of Landau-Levich regime.

The thickness of the film tends to decrease in the evaporation regime as the substrate’s movement speed increases. This occurs as a result of the solute’s shorter residence duration, which results in a lower accumulation amount. In the Landau-Levich regime, the thickness of the deposited layer grows as the coating speed increases, owing to the viscous force that pulls the liquid film out and then dries [86]. The Landau-Levich regime is advantageous in terms of practical use. However, if the solvent present in the precursor solution has a high surface-tension, the perovskite material can generate islands with a high roughness surface [87].

3.2 Slot die coating

The slot die coating process is similar to blade coating in that it uses a different coater to apply a thin and homogeneous film. The coater is made up of a head that includes a downstream and upstream die. The precursor solution is initially pushed to the die head using a syringe pump in this procedure. The solution forms a liquid bridge between the head and the substrate during the procedure. As the substrate begins to move, a moist layer of solution forms. The flow rate of the solution, coating speed, spacing between the head and the substrate, viscosity of the solution, and other factors all influence the quality and shape of the deposited film in this process [88]. Figure 6(a) depicts a schematic diagram of the coating process. This method can be used to deposit the inorganic part of the perovskite material in the first step of sequential deposition. Again, annealing can be done by N₂ gas quenching in this process for solvent evaporation to fabricate a pin hole free perovskite layer which is shown by Figure 6(b). It is the only approach that has shown perovskite solar cell role-to-role manufacturing.

Figure 6.

(a) Schematic diagram of slot-die coating process, (b) schematic diagram of gas quenching annealing of slot-die coated film.

Apart from the many advantages of using the slot die coating method to fabricate high-quality large module perovskite films, it also has some failure mechanisms that prevent the development of good morphological films. The failure mechanism includes a low flow limit of the solution, which causes a breakup of the downstream meniscus, resulting in a discontinuity in the wet film [89]. The creation of an air entrainment defect in the wet film is caused by the breaking of upstream meniscus bubbles inside the film. Flooding or dripping occurs when the ink flow to the head is greater than the coating speed, leading in a progressive build-up of ink at the coating head, preventing the desired film thickness from being attained.

3.3 Bar coating

A well-known method for producing high-efficiency PSCs is bar coating, often known as D-bar coating. The procedure is similar to blade coating, but it spreads the solution across the substrate with a cylindrical bar. In this case, the precursor solution is put onto a bar with a small cylindrical tube. As the substrate moves below the cylindrical bar, a wet film is formed due to the solution passing through the cylinder’s wire gap [90]. Figure 7 depicts a schematic diagram of the coating procedure. After annealing, the deposited solution is converted into the ultimate film. The deposited film thickness and quality is directly proportional to the amount of solution passing through the wire gap of the cylinder and the type of solvent used to prepare the solution in this technique, providing for high repeatability and minimal precursor solution loss [91].

Figure 7.

Schematic diagram of Bar coating process.

3.4 Spray coating

Spray coating is a low-temperature coating technology that is advanced. A nozzle sprays the tiny solution droplets over the pre-heated substrate at a high speed in this process. Pneumatic spraying, which provides the solution droplet as a rapid gas flow, is the most commonly used spray coater. The spray coating process consists of four steps: creation of the droplet at the nozzle, transportation of the droplet to the substrate, coalescence of the droplet on the substrate, and drying [92]. The droplets are formed through the nozzle in the first step, which is known as atomization. The substrate is ready for the annealing procedure to obtain the ultimate film after the droplet coalescence on the wet film. Figure 8 depicts a schematic diagram of the overall coating process.

Figure 8.

Schematic diagram of spray coating process.

To achieve thorough wetting of the substrate, the solvent utilised in this approach must have a low surface tension and contact angle. Another way to produce a completely wet substrate is to use a pre-heated substrate, which lowers the surface tension and reduces the contact angle between the solution and the substrate [93]. Furthermore, the type of nozzle utilised, the pressure of the gas jet propeller, the distance between the nozzle and the substrate, and the temperature of the pre-heated substrate all have a significance in obtaining a uniform deposited film. An ultrasonic spray coater was utilised to create a high-quality smooth perovskite coating with homogeneous droplets.

3.5 Ink-jet printing

Ink-jet printing is a common method for fabricating electrical devices, particularly optoelectronics. This approach has a cost advantage over others due to its maskless on demand printing and, more crucially, contactless high-resolution printing. The printing procedure operates in the same way as spray coating. A piezo electronic transducer and a detector are also used to control the droplet size and trajectory, enabling this system to deposit material with precise patterning [94]. This technique is divided into two groups based on how ink droplets are emitted. Continuous ink-jet printing (CIP) and drop-on-demand ink-jet printing is two of them (DOD).

As the name implies, in continuous ink-jet printing, the solution droplet flows continuously towards the substrate under the influence of gravity. When the droplets fall from the nozzle, they acquire up an electric charge. The charged droplets are subsequently sent through a deflection coil, which directs them. A tiny voltage is applied between the nozzle and the ground to achieve this. Figure 9(a) depicts a schematic of the complete printing process. In this method, a piezoelectric transducer (PZT) provides an appropriate frequency for regular separation of the droplets, and the separation force involved is surface tension. When no printing is necessary, this approach also recycles the depositing material by collecting it in a reservoir [95]. Aside from that, because it is a non-contact printing technique, it enables excellent film deposition on both rough and curved substrates.

Figure 9.

(a) Schematic diagram of CIP process, (b) schematic diagram of DOD printing process.

Again, DOD printing is a modern, high-precision printing technology that can print with a single drop precession, allowing it to save a lot of material that would otherwise be wasted throughout the CIP process which is depicted in the Figure 9(b). In this technology, a computer programme controls the movement of either the printing head or the substrate. Due to the contraction of the ink contain volume, the printing material is ejected out the nozzle at a definite pressure pulse. The pressure pulse production method is split into two parts: piezoelectric DOD and thermal DOD.

In piezoelectric DOD, an ink droplet is created by passing an impulse current over the transducer, causing PZT to deform mechanically. The majority of the printing industry employs this printing process because it allows for variable actuation pulses to adjust the velocity and size of ink droplets released from the nozzle. The thermal DOD, on the other hand, uses thermal evaporation to create the ink droplet. This is accomplished by passing electricity through the small resistive heater. When the temperature rises above the boiling point of the printing ink, the vapour entrapment causes bubbles to form. When the heater’s power is turned off, the bubble begins to collapse because of heat transfer to the surrounding tank depending on the temperature difference [96]. Due to the difficulty in generating ink bubbles for high vapour pressure solution ink, this approach is not appropriate for large-area printing.

3.6 Screen printing

The screen-printing process is used to print a pattern that has previously been created on a thread or steel mesh. When the ink has a high viscosity, this approach provides the best printing pattern. The printing ink is spread over the patterned mask with a squeegee, which prints the pattern on the substrate. Due to the high viscosity of the printing fluid, this method produces a somewhat thick film [97]. Figure 10 depicts a schematic diagram of the printing process. Depending on the printing technique, this printing technology is divided into two categories: flatbed and rotary screen printing.

Figure 10.

Schematic diagram of screen-printing process and printed patterns.

The printing is done in a stepwise manner in the flatbed method, with the screen held extremely close to the top of the substrate and the paste transferred over the screen by the squeegee. The screen is then lifted or transferred to continue the printing process over the entire substrate after the printing is completed. For roll to roll and large-area printing, the recurrence of this procedure is not suitable. However, rotary screen printing is a low-cost, high-efficiency large-area printing process for generating rapid, precise patterns. The squeegee and paste are stored in a folded tube in this way. The stationary squeegee constantly spreads the paste across the substrate through the mesh as the tubular screen rotates with the substrate, allowing for complete printing [98].

Table 4 shows some of the best results obtained by all of the large-scale deposition processes.

Methods	Device architecture	Device area (cm^2)	PCE, (%)	Ref.
Blade coating	ITO/PTAA/MAPbI3/C60/BCP/Cu	57.2	14.6	[99]
	ITO/PTAA/MAPbI3/C60/BCP/Cu	63.7	16.40	[100]
	FTO/TiO2/MAPbI3/Spiro-OMeTAD/Au	47	14.7	[101]
Slot-die coating	FTO/NiMgLiO/FA0.83Cs0.17PbI2.83Br0.17/LiF/C60/BCP/Bi/Ag	20.77	16.63	[102]
	FTO/SnO2/ Cs0.15FA0.85Pb(I0.83 Br0.17)3/Spiro-OMeTAD/Au	57.5	16.22	[103]
	ITO/c-TiO2/CH3NH3PbI3−xClx/ Spiro-OMeTAD/Au	168.75	10	[104]
Bar coating	FTO/SnO2/GAxMA1−XPbI3/spiro-OMeTAD/Au	16	13.85	[105]
	FTO/SnO2/(FAPbI3)0.875(CsPbBr3)0.125/spiro-OMeTAD/Au	25	17.01	[106]
	FTO/SnO2/(FAPbI3)0.95(CsPbBr3)0.05/Spiro-OMeTAD/Au	19.69	17.94	[107]
Spray coating	ITO/PEDOT:PSS/CH3NH3PbI3/C60/BCP/Cu	56.25	16.4	[108]
	FTO/SnO2/C60/CsxFA1−xPbIyBr3−y/Spiro-OMeTAD/Au	25	14.7	[109]
	FTO/TiO2/MAPbI3−XClX/PTAA/Au	40	15.5	[110]
Ink-jet printing	PEN/Ag(NWs)/PEDOT:PSS/CH3NH3PbI3/PC71BM/PEI/Ag(NWs)	180	10.68	[111]
	FTO/TiO2/MAPbI3/Spiro-OMeTAD/Au	4	13.27	[112]
Screen printing	FTO/TiO2/ZrO2/CH3NH3PbI3/carbon	70	10.74	[113]
	FTO/TiO2/ZrO2/5PVAXMA1−XPbI3/carbon	49	10.4	[114]

Table 4.

A comparison of the performance of several perovskite large area modules prepared utilising various large scale techniques.

Not only these methods are utilised to fabricate active perovskite layers, but they also provide high-quality deposition of all transport layers, including metal electrodes. However, the type of solvent employed in the precursor solution, the solute concentration, printing speed, surface tension, and viscosity of the solution ink, among other factors, all play key role in achieving high-quality pin hole-free morphological film deposition.

4. Conclusion and prospects

PSCs have been a front runner in competition towards the commercially available photovoltaic technologies. Despite the extensive effort and significant progress in the large area fabrication of perovskite based photovoltaic, the large-area efficiency still lags behind those of small area devices. One of the main reasons for this gap between large and small area PSCs is the difficulty in large area processing of uniform and high-quality perovskite thin-films. This chapter covers all the solution processing techniques used in the fabrication of perovskite solar cells including the large-area coating techniques. The role of each technique is achieving homogeneous, pinhole-free and large grain-sized perovskite thin-films have been explained. The understanding the crystallisation kinetics of perovskite in one or more of these techniques will help in developing commercially viable large-area perovskite based solar modules.

Conflict of interests

There are no conflicts to declare.