Two⁃Dimensional Perovskite Photovoltaic Devices

Chao Ji; Tuo Li; Xiaofeng Zou; Lu Zhang; Chunjun Liang

doi:10.7536/PC211022

Progress in Chemistry ›› 2022, Vol. 34 ›› Issue (9) : 2063-2080 DOI: 10.7536/PC211022

research-article

Scan to read & Share

Two⁃Dimensional Perovskite Photovoltaic Devices

Chao Ji ¹^,²^,³
, Tuo Li ¹^,²
, Xiaofeng Zou ¹^,²
, Lu Zhang ¹^,²
, Chunjun Liang ³^,^*

Author information +

History +

PDF

Abstract

Organic⁃inorganic hybrid halide perovskite solar cells (PSCs) have attracted more attention because of their low cost, simple preparation process and high power conversion efficiency(PCE). It is widely considered as ideal candidates for the next generation semiconductor photovoltaic technology. However, the instability caused by moisture, light and heat is still the main factor restricting the commercialization of PSCs devices. Layered two⁃dimensional (2D) perovskite has attracted extensive attention because of its good environmental stability. By introducing different kinds of hydrophobic large volume organic ammonium cations, a stable 2D structure can be formed. However, due to the existence of insulating organic spacer cations, its charge transport capacity is blocked and its power conversion efficiency is poor. Therefore, improving the PCE of 2D PSCs on the premise of maintaining excellent stability is the key problem. According to the development process of different kinds of 2D perovskite photovoltaic devices, this paper summarizes the key problems affecting the structure and performance of 2D perovskite, such as vertical orientation, mixed dimensional engineering, quantum well regulation, organic spacer cation replacement engineering and so on. Finally, the future development of 2D PSCs is prospected.

Graphical abstract

Keywords

perovskite photovoltaic devices(PSCs) / two⁃dimensional structure / high efficiency / stability

Cite this article

Download citation ▾

Chao Ji, Tuo Li, Xiaofeng Zou, Lu Zhang, Chunjun Liang. Two⁃Dimensional Perovskite Photovoltaic Devices. Progress in Chemistry, 2022, 34(9): 2063-2080 DOI:10.7536/PC211022

登录浏览全文

4963

注册一个新账户忘记密码

1 Introduction

2 Crystal structure and physical properties of two⁃dimensional perovskite

3 RP phase perovskite solar cell

3.1 Low n⁃value 2D⁃RP perovskite

3.2 High n⁃value 2D⁃RP perovskite

4 DJ phase perovskite solar cell

4.1 Asymmetric diammonium structure

4.2 Symmetrical diammonium structure

5 ACI phase perovskite solar cell

6 Conclusion and Prospect

1 Introduction

Since Miyasaka first reported the use of organic and inorganic metal halide perovskite semiconductor materials for dye-sensitized solar cells in 2009, perovskites have made unprecedented progress as potential semiconductor photovoltaic (PV) materials, and have rapidly become one of the hottest topics in semiconductor materials science in the past few years^{[1⇓⇓⇓⇓~6]}. The unprecedented progress of photovoltaic devices is attributed to the discovery of elemental ions. For example, 2D perovskite has better thermal stability, larger organic cations can effectively hinder the movement of internal ions, and 2D perovskite has good hydrophobic characteristics to improve the humidity stability of thin films and devices^[7⇓⇓~10]. However, as a photovoltaic device, pure 2D perovskite has an optical band gap of at least 2.5 eV or wider, which limits photon absorption to less than 500 nm, and has a larger exciton binding energy (≈ 300 meV) than three-dimensional (3D) perovskite (10-50 meV), which affects the output of photovoltage as a semiconductor photovoltaic device, thereby reducing the power conversion efficiency (PCE)^[11⇓~13]. However, the excellent stability of 2D perovskite makes people have very high expectations for this material, which is one of the reliable ways to achieve stable PSCs^[14]. The disadvantages of 2D perovskites can also be avoided by preparing hybrid 2D/3D^[15]. In this paper, 2D perovskite structure devices for photovoltaic power generation are discussed in detail, in an attempt to classify 2D halide perovskites, and the development prospects of 2D/3D hybrid perovskites are briefly analyzed, and the crystal structure, optoelectronic properties, carrier dynamics, and their effects on photovoltaic performance and device stability are discussed. Finally, this paper highlights some of the key challenges that remain to be addressed, while outlining the application prospects of 2D perovskite structures in efficient and stable solar cells.

2 Crystal structure and physical properties of two-dimensional perovskite

Tolerance factor (TF) is usually used as a general calculation formula to explain whether the perovskite crystal structure can be stably formed, and the change of TF can also be used to predict the transition from three-dimensional to two-dimensional perovskite structure^[16]. As shown in Figure 1A, a typical 3D halide perovskite can be represented as ABX₃, where A is a monovalent cation such as methylammonium (MA⁺), formamidine (FA⁺), cesium (Cs⁺), or rubidium (Rb⁺),B is a divalent metal cation (Pb²⁺, Sn²⁺, etc.), and X is a halide anion (I^-, Br^-, or Cl^-). This structure consists of corner-sharing [BX₆]^4- octahedra in which the A⁺ cations occupy the interstitial sites. TF only depends on the ionic radius of A, B and X sites in the perovskite structure.

(1)

T F = r A + r X 2 (r B + r X)

In formula (1), r_A, r_B and r_x are the atomic radii of A, B and X elements, respectively. For most typical 3D perovskites, such as MAPbI₃ or FAPbI₃,0.8<TF<1 (Figure 1B). The deviation of the tolerance factor from the above range leads to a crystal structure distortion, with octahedra in edge-sharing and face-sharing, rather than the corner-sharing structure of classical ABX₃ perovskites. When the size of the A-site cation is too large to be well accommodated in the octahedral cage of the original 3D perovskite, the initial tetragonal phase structure will undergo a phase transition to a two-dimensional structure. Although 2D perovskites have large exciton binding energy and large band gap structure, which limit their photoelectric conversion efficiency, 2D perovskites still attract people's interest because of their better stability.

2D perovskites can be cut along crystallographic planes < 100 >, < 110 >, or < 111 > of the 3D structure to form three different types of layered structures, namely < 100 >, < 110 >, or < 111 > oriented layered perovskites^[17]. Among the perovskites of these three phases, < 100 > -oriented layered perovskites are the most common (fig. 2a). This oriented perovskite can be divided into layered RP, DJ, and ACI-type perovskite structures, and the class of 2D perovskites can be defined by shifting the central perovskite layer of the reference structure by half of the unit cell vector length in the phase space plane direction (a-axis or b-axis) (Figure 2B)^[18]. For example, fig. 2C shows an RP layered perovskite structure with n = 1. From a crystallographic point of view, the layered perovskite is derived from an oxide perovskite, and in the RP layered perovskite, the inorganic semiconductor layer shows a (1/2,1/2) in-plane displacement along the ab plane in each unit cell, so that the inorganic layers are stacked in a staggered arrangement of half a crystal unit in the in-plane direction^[19]. The other extreme is that the layers in the DJ phase do not show any displacement and stack perfectly with each other (Figure 2D), with a displacement of (0,0). The ACI-type structure is in between and combines the layer-stacking characteristics of the RP and DJ structure types, showing a (1/2,0) shift.

The name 2D perovskite originates from the compound K₂Ni

F 4

discovered by Ruddlesden and Popper in 1957^[20]. The 2D perovskite material grows along a single orientation during the growth process, and finally forms an alternating organic and inorganic lamellar structure^[21]. Depending on the type of organic ammonium cation, the organic layer may consist of a monolayer or bilayer organic sheet. The common 2D materials are (NH₃⁃R⁃NH₃)MX₄ and , as shown in Figures 2C and 2D, where R

NH 3 +

and

NH 3 +

NH 3 +

can be aliphatic or aromatic organic ammonium cations, which are called monoammonium and bisammonium structures, respectively, and the resulting 2D perovskites are also called Ruddlesden Popper (RP) perovskite and Dion Jacobson (DJ) perovskite, respectively^[22]. In (R⁃NH₃)₂MX₄, the inorganic layer interacts with the organic layer through ionic and hydrogen bonds. The tails of the organic moieties interact via van der Waals forces to constitute a stable overall structure. While in the (NH₃⁃R⁃NH₃)MX₄, the hydrogen atoms on both sides of the head and tail of the organic part are respectively combined with the inorganic layers at both adjacent ends, thereby eliminating the van der Waals gap between the tail layers of the organic part.

The alternating layer structure properties of organic and inorganic parts attract researchers to explore in depth. To sum up, 2D perovskites have the following characteristics: 1) dielectric confinement effect. The reason is that the low dielectric constant of the organic layer and the high dielectric constant of the inorganic layer change periodically, and the large difference in dielectric constant between the inorganic layer and the organic layer in the two-dimensional material greatly increases the exciton binding, which is not conducive to the charge dissociation efficiency of free carriers^[23]. 2) The band gap of organic materials is usually higher than that of inorganic materials. When the thickness of the 2D grid is reduced to the de Broglie wavelength of the carrier, the Quantum confinement effect on the charge behavior will occur, and a typical Quantum well (QW) structure will be formed, which will halve the Bohr radius and increase the exciton binding energy by four times^[24]. This is shown in Figure 2E, where the inorganic layer acts as a potential well and the organic layer acts as a potential barrier. This special quantum confinement effect makes 2D perovskite have a large binding energy and suppresses the generation of free charges. But at the same time, this unique property makes these two-dimensional perovskites one of the potential candidate star materials for luminescence applications^[25⇓~27]. In addition, the organic moiety is usually an aliphatic or aromatic organic ammonium cation, and this component imparts hydrophobicity to the perovskite structure, which in turn improves its chemical stability compared to the 3D structure. 3) The wider band gap of pure 2D perovskites (> 2.3 eV) makes them less suitable for photovoltaic applications. However, the band gap can be changed by changing the composition of the inorganic conductor layer of 2D perovskite, or by replacing Pb with Sn, or by controlling different organic cations^[28⇓~30]. In this way, the device efficiency can be effectively improved on the premise of maintaining stability. In conclusion, compared with 3D perovskites, 2D perovskites with periodic quantum well structures can significantly enhance the environmental stability of active layer materials, which has attracted considerable research interest.

3 RP-phase perovskite

3.1 Low n value 2D RP perovskite

Layered RP-phase perovskites can be represented by A'₂A_n_-1B_n

X 3 n + 1

, where A ′ is an aromatic or aliphatic alkylammonium spacer cation in the + 1 valence state, while A is an organic cation, B is a divalent metal, and X is a halide anion. The 2D perovskite structure can be thought of as bulky alkylammonium cation organic layers separating octahedral [MX₆]^4- inorganic layers of a specific thickness, and Coulomb forces and hydrogen bonds between the organic and inorganic layers maintain the integrity of the 2D structure. Therefore, the photoelectric properties of 2D RP perovskites are affected by the number of layers and the nature of the spacer cation. The 2D layered structure is regarded as a spontaneously formed multiple quantum well structure, in which the organic layer with a relatively wide band gap acts as an "insulating layer" and the semiconductor inorganic layer acts as a "well". Moreover, because of the larger molecular weight and the better hydrophobicity of the organic component with longer alkyl chain, the addition of long-chain alkyl ammonium can effectively isolate the erosion of water and oxygen on perovskite and improve the stability of the device in air. The value of n can be defined as the number of inorganic layers in the perovskite structure (specifically, the number of octahedral layers composed of halogen cations sandwiched between two organic layers), which can be achieved experimentally by reasonably adjusting the stoichiometric ratio of the initial precursor. In this paper, two-dimensional perovskites with n < 10 are called low-n 2D/RP perovskites. In 2014, Smith et al. First prepared a high-quality mixed perovskite (PEA)₂(MA)₂(Pb₃I₁₀) active layer film with a 2D RP structure, and the device had an open circuit voltage (V_oc) of 1.18 V and a power conversion efficiency of 4.73%^[13]. And the 2D perovskite thin film has better hydrophobicity and can be prepared even in a humid environment. XRD also showed that the 2D perovskite film was basically unchanged after 46 days of storage under 52% RH humidity aging test conditions. Larger bandgap 2D perovskites may also be suitable for use as high bandgap absorber layers in tandem stack devices. Therefore, the excellent stability and wide band gap of 2D perovskites have gradually attracted people's interest. Cao et al. Subsequently used BA(CH₃(CH₂)₃NH₃) to prepare 2D perovskites with different n values (n = 1, 2, 3, 4), and found that the band gap of this series of 2D perovskite films decreased with the increase of n value, from 2.32 eV(CH₃(CH₂)₃NH₃)₂PbI₄(n=1) to 1.60 eV CH₃NH₃PbI₃(n=∞) (Figure 3A)^[31]. And the compound exhibits strong light absorption in the visible region while exhibiting intense photoluminescence at room temperature, making it a promising light-absorbing material for photovoltaic applications (Figure 3B). The growth patterns of 2D perovskites with different n values were studied, and it was found that when the layer becomes thicker (n > 1), there is a competition between BA ions (trying to achieve growth within the planar layer) and MA ions (trying to expand growth outside the layer) (Figure 3C). The final XRD shows that the stability of the n = 3 perovskite film is further improved, and the efficiency of the optimal device is 4.02%. The main reason for the low photoelectric conversion efficiency of 2D | RP perovskite is that the electron transmission is mainly concentrated in the inorganic layer of Pb | I | Pb, and the inorganic groups are separated by the organic ammonium cations with long alkyl chains, resulting in the electron transmission being blocked by the organic layer, just like the insulating layer cutting off the electron transmission path. Large organic cations hinder out-of-plane carrier transport. Highly packed, vertically oriented crystallites thus help to optimize the charge transport channel.

Based on the above research, Tsai et al. Proposed a further improvement scheme to successfully prepare vertically oriented quasi-two-dimensional perovskite by hot cast deposition, which can effectively make the perovskite film preferentially oriented in the crystallization process, and proved that the inorganic part (Pb I Pb) is in the preferred growth direction of the out-of-plane direction (Fig. 4A)^[32]. The crystalline phase of the inorganic component in the nearly single-crystal quality layered perovskite film prepared by the method shows a strong out-of-plane ordered arrangement to promote efficient charge transport, and the PCE of the optimal device reaches 12.52% without hysteresis. The efficiency of the unencapsulated two-dimensional perovskite device is still more than 60% of the initial value after 2250 H of continuous standard sunlight irradiation. Based on the importance of the vertical orientation of 2D perovskite, Sargent et al. Systematically explained the growth mode of low-dimensional perovskite (RDP), and studied that the layered intermediate complex containing solvent provided a scaffold during the formation of perovskite by in situ X-ray scattering, which contributed to the nucleation and growth of RDP during annealing^[33]. Transient absorption spectroscopy can be used to analyze the distribution of quantum well thickness. Choosing a larger cation leads to a change in the QW distribution. The use of BTA instead of PEA in RDPs leads to a rise in the composition of RDPs with low n values, and at the same time to a quantum well redshift with large n values. Selective reactions between various ions in the precursor solution and their regulation by changing large organic cations also affect the kinetic rate of RDP formation. Finally, from the RDP film prepared by the hot casting method, it was found that the perovskite with low n value tended to form parallel to the bottom, while the wells with high n value clustered near the surface of the film and grew vertically (Fig. 4 B). Meanwhile, Shang et al. Made a deep understanding of carrier dynamics in (C₆H₅CH₂CH₂NH₃)₂(CH₃NH₃)_n_-1Pb_n

I 3 n + 1

(n = 3, 5) -based perovskite thin films^[34]. It is found that electrons are efficiently transferred from the small n component to the large n perovskite component, and holes are transferred in the opposite direction (Fig. 4C). The hybrid perovskites are aligned perpendicular to the substrate from small to large n values and coexist randomly in the horizontal plane. Therefore, the crystal structure of perovskite thin films can be designed to adjust the carrier transport dynamics, so as to achieve controllable carrier dissociation transport characteristics. These results are important for the design of optoelectronic devices such as solar cells, light-emitting diodes, and lasers. Vertically ordered (n from small to large) quasi-2D perovskite thin films can effectively promote the directional transfer of charge, in addition to the hot-cast deposition method, Zhang et al. Reported a different strategy for ordered vertical crystallization, that is, by introducing vacuum polarization treatment during the crystallization process to enhance nucleation, resulting in the preparation of uniformly arranged nanolayers with different n values (PEA₂MA_n_-1Pb_n

I 3 n + 1

)^[35]. By contrast, without vacuum poling, nanolayers with different n values form a vertically dispersed arrangement from small to large between the bottom and top surfaces of the quasi-2D perovskite film. In addition to the observation from absorption, photoluminescence (PL) and energy dispersive X-ray spectroscopy (EDS) that the vacuum polarization step would promote the uniform distribution of nanolayers with different n values, the distribution conjecture of uniform n values was also successfully verified by the precise mechanical tape stripping method. In the case of uniform distribution, the transient absorption shows an efficient carrier transfer within 10 ps and finally reaches a record high fill factor (FF) of 82.4%. The distribution of vertical crystalline phases plays an important role in quasi-two-dimensional perovskite solar cells, and the modification of transport layer also has a great impact on the crystallization of perovskite. Liu et al. discussed how the crystallization driving force affects the vertical distribution of the number of two-dimensional layers, and attempted to modify the surface of the hole transport layer PEDOT: PSS to BrB PEDOT: PSS to change the acidic surface properties of the hole transport layer, thereby producing the vertical distribution required for charge transport^[36]. Solvent engineering is also a viable strategy to improve phase purity and device efficiency. Caiazzo et al. Studied the crystallization mechanism of quasi-2D perovskite thin films under the coexistence of mixed solvents, and analyzed the molecular interactions between different perovskite precursors (MAI, BAI, PbI₂) and solvents (DMF, DMAc, NMP and DMSO)^[37]. The enhancement of the crystallinity and crystal orientation of the titanite film by adjusting the ratio of the mixed solvent leads to an increase in the efficiency of photovoltaic devices based on BA < n = 4 > perovskite from 3.5% without the mixed solvent to 11.1% and 10.2% with DMF/DMSO and DMF/NMP mixtures.

Although the crystal orientation in BA-based films can be well controlled by casting solution on hot substrates, resulting in the formation of efficient charge transport channels along the out-of-plane vertical direction, the defects of the materials themselves still hinder the further improvement of their optoelectronic properties and device stability. In 2017, Zhang et al. Effectively improved the grain size and surface quality of the doped (BA)₂(MA)₃Pb₄I₁₃ film by partially replacing MA⁺ with Cs⁺ (Figure 5A), and obtained good crystal orientation. Meanwhile, Cs⁺ doping helped to reduce the trap density, improve the carrier mobility and optimize the charge transfer path. The excellent film quality significantly increased the PCE of the device from 12.3% to 13.7% (Figure 5B)^[38]. Moreover, the 5% content Cs⁺ doped device retained 89% of its initial cell efficiency after exposure to ambient conditions of 30% relative humidity for 1400 H, and exhibited improved stability against high humidity (under test conditions of 65% and 85% relative humidity) and continuous thermal aging (80 ° C). These results are expected to help perovskite devices toward commercial applications. As mentioned above, when the thickness of the inorganic layer grid structure of 2D perovskite is gradually reduced to the order of the carrier de Broglie wavelength, the periodic QW structure will be formed, which will produce quantum confinement of charge transport behavior, and the increased exciton binding energy will block the dissociation of free charges. In addition, the large dielectric constant difference between the inorganic layer and the organic layer in 2D perovskite materials also increases the exciton binding energy, which is not conducive to the charge dissociation process, resulting in the low photoelectric conversion efficiency of 2D perovskite. Therefore, controlling the dielectric constant between layers to reduce the exciton binding energy is one of the means to improve the efficiency of 2D perovskite photovoltaic devices. Subsequently, in 2018, Zhu et al. Prepared a two-dimensional perovskite thin film MA₂PbI₄ based on methylammonium (MA) by vapor phase fumigation technology^[39]. Because MA (~ 11) has a larger dielectric constant than BA (~ 4), the exciton binding energy of the MA₂PbI₄ film is reduced from 510 meV to 172 meV compared with the traditional 2D perovskite film composed of BA cations (Figure 5C). According to the Wentzel Kramers Brillouin approximation, the tunneling probability of carriers through the QW was found to increase by four orders of magnitude due to the reduced interlayer thickness, which was also confirmed by XRD (the interlayer spacing decreased from 13.3939 Å for BA₂PbI₄ to 9.0808 Å for MA₂PbI₄). In addition, the band gap of MA₂PbI₄ perovskite is narrower and the conductivity is more excellent. The PCE of the optimal 2D PSCs reached 16.92%, and the device showed excellent long-term stability under both illumination and exposure to ambient conditions. The efficiency was maintained at 97.2% of the initial value under constant temperature of 60 ℃ for 500 H in argon environment. The PCE of the unpackaged device was only slightly degraded by 3. 8% after 500 H of continuous illumination in air.

Passarelli et al. Tried to dope the electron acceptor TCBQ into the electron-donating naphthyl (Nap) organic layer lattice, and the interaction between the electron donors and acceptors of Nap and TCBQ increased the electrostatic screening of excitons, thus reducing the binding energy by nearly 50% relative to undoped perovskite^[40]. And because the strong interaction of Nap and TCBQ overcomes the unfavorable factor that the increase of thermodynamic upper layer spacing destroys the lattice, TCBQ can be inserted between naphthyl organic layers without destroying the geometric structure of Pb and I networks. The dielectric environment of the insulating organic layer is eventually changed without affecting the inorganic layer lattice, reducing the exciton binding energy of the perovskite. In 2D perovskite, most of them use MA⁺ as the organic spacer cation, and the spacer cation based on FA is rarely studied. In 2020, Dong et al. First successfully synthesized the FA derivative ThFA (2 thiophene formamidine hydroiodate), and used it in the 2D perovskite (ThFA)₂MA₂Pb₃I₁₀^[41]. The quality of the film is adjusted by inducing the orientation of crystal growth with the precursor organic salts assisted crystal growth (PACG) method. The results show that the grain size of two-dimensional perovskite thin films treated by PACG increases, the grain boundary decreases, the trap density decreases, the preferred growth direction is perpendicular, the carrier lifetime increases, and the non-radiative recombination loss decreases. The PCE of the optimized device reaches 16.72%. In the same year, Xi et al. Designed a new chemical derivative from the molecular point of view, a spacer cation with a conjugated backbone, 3phenyl2 propylammonium (PPA), to prepare more efficient 2D perovskite photovoltaic devices^[42]. It is found that the multiple hydrogen bonds between PPA and inorganic octahedra strengthen the 2D layered structure while forming the RP phase. Fig. 5d shows that as the thickness of the organic layer increases (n = 2, 3, 4), the crystal orientation in the horizontal direction is reduced, so that the free charge carrier transport performance in the direction perpendicular to the substrate is improved. Finally, the planar p/I/n perovskite photovoltaic device based on the PPA/RP structure achieved an efficiency of 14. 76%, and the PCE of the packaged device remained at 93. 8% ± 0.25% of the initial efficiency under the condition of continuous aging at 85 ℃ and 85% humidity for 600 H.

The structural stability of 2D RP layered perovskites comes from the protection of organic ammonium molecules and the weak van der Waals force and hydrogen bond-dominated interaction between layers. This interaction controls the self-assembled structure and charge transport channels of 2D RP layered perovskites. In-depth study of this interaction mechanism is expected to improve the stability and photoelectric conversion efficiency of 2D RP perovskite thin films. Ren et al. innovatively introduced an organic ammonium containing S atom (MTEA) through a combination of theoretical calculations and experimental studies. In addition to the weak van der Waals interaction, the interaction between the S and S atoms in the two MTEA molecules makes the (MTEA)₂(MA)₄Pb₅I₁₆2D⁃RP perovskite structure more stable (Fig. 6a)^[43]. The S/S interaction induced 2D/RP perovskite thin film has a strong out-of-plane preferred growth, which promotes the dissociation and transport efficiency of charges and reduces the trap density. Results The efficiency of the optimized device has been significantly improved. The optimal device PCE is as high as 18.06% (certified 17.8%). The enhanced interlayer molecular interaction also significantly improves the environmental working stability and thermal stability of 2D | RP perovskite thin films. MTEA-based 2D RP PSCs maintained more than 85% of their initial efficiency under continuous illumination for more than 1000 H at the maximum power point (MPP). Based on the comparison of a series of alkyl ammonium spacers with different chain lengths (ethylammonium to hexylammonium), Wu et al. Proposed a strategy to control the crystallization, crystal phase distribution and quantum well constraints of 2D RP perovskite (n = 4) by changing the alkyl chain length to adjust the intermolecular van der Waals interaction^[44]. The fabrication of high-quality 2D perovskite thin films with good phase distribution was achieved by using the optimal AA cation spacer. The lower exciton binding energy also enables highly efficient charge transfer. The optimal PCE based on the (AA)₂MA₃Pb₄I₁₃ 2D perovskite device is 18.42% (Fig. 6 B). Wei et al. studied the effects of formamidine (FA) and 4-fluorophenethylammonium (FPEA) cations on the performance of thin films and devices by controlling the organic components of 2D | RP perovskite materials, and found that FA cations are essential to promote the vertical orientation growth of crystals and improve the quality of thin films^[45]. FA-based perovskites have a narrow band gap so that their absorption sidebands extend into the infrared region, which greatly increases the photocurrent. At the same time, the addition of hydrophobic FPEA cations with large dipole moment can improve the humidity stability, reduce the defect density, and improve the charge mobility. The device efficiency based on the (FPEA)₂(FA)₈Pb₉I₂₈structure can reach 16.15% (Fig. 7 a), and the unencapsulated sample maintains 95% of the initial efficiency after 2112 H in ambient air with relative humidity ranging from 30% to 70% (Fig. 7 B). In 2021, in order to overcome the problem of random phase distribution in perovskite thin films, Huang et al. Proposed an interface modification strategy based on 2D perovskite GA₂MA₄Pb₅I₁₆, using guanidinium bromide (GABr) treatment on the top interface of perovskite thin films to induce the secondary crystallization process (Fig. 7D)^[46]. This strategy can control the formation of crystal nuclei on the surface of the film and control the crystal orientation, resulting in the reordering of the two-dimensional perovskite phase distribution. In addition, the incorporation of GA⁺ can improve the environmental and light stability of the device and inhibit the degradation of perovskite materials. Stable 2D PSCs with the highest PCE of 19.3% were obtained through the GABr modification strategy (Fig. 7 C), and the unencapsulated optimized device could retain 94% of the initial efficiency when stored under ambient conditions for more than 3000 H. The interface modification strategy can be used as a conventional method to fabricate stable and low-defect 2D perovskite thin films. N is a key factor in determining the optoelectronic properties of layered perovskites, affecting the band gap and thus their absorption as well as conductivity/mobility and stability. In most of the reported PSCs with efficiency > 10%, the n value was ≥ 3. However, the higher the n value, the more difficult it is to obtain pure perovskite, so some strategies of mixed 2D/3D perovskite or 2D perovskite as 3D perovskite passivation layer are derived.

3.2 High-n 2D RP perovskite

Two-dimensional perovskites with low n value have excellent stability, but their efficiency is very low due to their weak carrier transport ability, so two-dimensional perovskites with appropriate n value are the key to both efficiency and stability. Usually, we call the modification of three-dimensional structure by n > 10 or a small amount of two-dimensional materials as high-n 2D perovskite. Quan et al. systematically analyzed the efficiency and stability of 3D perovskite and 2D perovskite with different n values, and found that the initial PCE of 3D perovskite device was 16.6%, but after 8 weeks of storage in nitrogen, its PCE decreased to less than 3%^[47]. Moderate n-value perovskites (n = 10) exhibit improved properties and good long-term stability; However, due to the poor carrier transport performance, the absolute performance of 2D samples is still low compared with the current most efficient 3D perovskite structure devices, which limits their further application. Therefore, they also appropriately increased the n value of the perovskite, and prepared a two-dimensional perovskite device with n = 60 and a certified PCE of 15.23%. It was found that after 60 days in a humid environment, the PCE of the device with n = 60 decreased to 11.3%, while the initial PCE of the device with N = 40 was similar to that of device with N = 60, but only decreased to 13.1% after 60 days (Fig. 8A). By comparing the thermal stability of different devices, it is also found that the stability of n = 60 devices is better, so it is speculated that when a physical moisture-proof layer (such as carbon electrode) is deployed into the device structure, combined with the use of quasi-2D perovskite (n = 60, 40), long-term stable samples can be expected. 2D perovskites have higher humidity stability, which can provide more stable perovskite photovoltaic devices. In addition, 2D/3D mixed-dimensional perovskites are able to tune their structural and optoelectronic properties through compositional regulatory engineering. By carefully adjusting the BA content, Wang et al. Obtained plate-like "layered" perovskite crystallites oriented perpendicular to the surface and embedded between 3D perovskite grains^[48]. With this unique 2D/3D heterostructure, a greatly enhanced crystallographic orientation was observed (Figure 8 B), resulting in a reduction in defects due to non-radiative recombination, thereby reducing the current-voltage hysteresis, increasing the efficiency to 17.3%, and significantly improving the operational stability of the device at 80 ° C (Figure 8 C).

2D and 3D heterostructure perovskites are very promising as a conventional preparation method for efficient and environmentally stable PSCs. Ye et al. Demonstrated an in-situ method to prepare 2D@3D((AVA)₂PbI₄@MAPbI₃) perovskite active layer thin film^[49]. The 3D perovskite film was treated by rapid thermal annealing to volatilize the organic components, and then the 2D perovskite structure was obtained by coating AVAI on the surface of the treated film. The ammonium and carboxyl functional groups of AVAI can interact with 3D perovskite through hydrogen and halogen bonds to grow an excellent 2D/3D interface in situ, and 2D perovskite also plays a role in passivating the surface defects of 3D perovskite, thereby reducing non-radiative recombination. Combining the advantages of two-dimensional and three-dimensional perovskite structures, the method achieves a device efficiency of up to 18.0%, and it still shows significantly enhanced stability under package-free environmental conditions. The 2D @ 3D perovskite structure with both efficiency and stability is a further trend for future device development. Lee et al. Added 1.67 mol% 2D phenylethylammonium lead iodide (PEA₂PbI₄) to the precursor solution to form phase-pure formamidine perovskite, and the photoluminescence lifetime was improved by one order of magnitude^[50]. The 2D perovskite is spontaneously formed at the grain boundary of FAPbI₃ (Figure 9 a) to protect FAPbI₃ from moisture, contributing to charge separation/collection and suppressing ion migration (Figure 9 B and C), finally achieving a PCE of 20.64% (19.77% for third-party certified PCE), thereby significantly improving the operational stability of the device under ambient conditions. Hybrid dimensional 2D/3D hybrid perovskites are currently emerging as one of the most reliable ways to improve the stability of PSCs while maintaining the high efficiency of 3D perovskites. One of the most critical issues of this technology is the balance between hydrophobicity and charge transport characteristics. Therefore, it is very important to further explore 2D perovskite materials with more excellent properties.

4 DJ phase perovskite

4.1 Asymmetric diammonium structure

DJ perovskite is also known as bis-ammonium-based perovskite. Compared with RP perovskite, the inorganic layers in DJ perovskite do not show any slight displacement in the (100) plane and are perfectly stacked with each other, thus forming a neat stacking arrangement, which is slightly different from the staggered stacking of RP perovskite. The layered DJ phase perovskite can be represented as A″A_n_-1B_n

X 3 n + 1

, where A ″ is a divalent (+ 2) aromatic or aliphatic alkylammonium spacer cation, while A is an organic cation, X is a halide anion, and B is a divalent metal. Recently, it has been reported that DJ perovskite can be combined with the inorganic layer of perovskite on both sides through chemical bonds due to the special structure of bisammonium group, while the monoammonium cation only interacts with the inorganic layer of perovskite on one side, so there is a van der Waals gap between adjacent cations. Although the stability of 2D RP perovskite PSCs is higher than that of traditional 3D PSCs due to the hydrophobic long-chain alkyl ammonium organic spacer cations (PEA⁺, BA⁺, etc.) and the van der Waals interaction between layers, the very weak van der Waals interaction between layers can not guarantee the structural stability of 2D perovskite framework, resulting in its easy degradation when exposed to external harsh high temperature and other environments. DJ perovskite may have better stability than RP perovskite based on the starting point of overcoming the weak van der Waals interaction between layers and improving the structural stability. In 2018, Mao et al. First used 3AMP and 4AMP as special spacer cations to prepare a new type of DJ perovskites^[51]. It is found that the slight difference of cations can affect the angular distortion process of Pb I Pb, resulting in a huge change in optical properties. DJ perovskites composed of 3AMP (aminomethyl piperidinium) and 4AMP (amino methyl piperidinium) have narrower band gaps compared to the most common 2D RP perovskites based on the BA structure, and the 3AMP series has larger Pb I Pb angles and smaller band gaps than the 4AMP series (fig. 10a). Finally, the PCE of the DJ-PSCs device based on the (3AMP)MA₃Pb₄I₁₃ as the active layer in the planar p-i-n structure is as high as 7.32% (fig. 10B). Subsequently, Li et al. Expanded the composition of 2D DJ perovskite again, prepared 3AMPY and 4AMPY (n = 1 – 4) -based thin film devices, and systematically studied the influence of the properties of organic spacer cations, such as size, shape, and electronegativity, on the optical properties of inorganic halide [Pb_n

I 3 n + 1

] layers (Figure 10C)^[52]. Through the comparison of XRD (single crystal X-ray diffraction), it is found that the organic cation not only affects the tilt of the octahedron, but also affects the stacking of the inorganic layer. 2D DJ halide perovskites can be templated by aromatic 4AMPY cations. For the 3AMPY cation, the octahedra are slightly tilted, and the smaller tilt of the octahedra results in a smaller band gap for the 3AMPY series than in the 4AMPY series. The aromatic cation exhibits a larger dielectric constant than the previously reported aliphatic series, and thus the corresponding compound exhibits a smaller exciton binding energy,

This may be beneficial for charge transport. The photovoltaic device using (3AMPY)(MA)₃Pb₄I₁₃ as the active layer showed excellent photoelectric conversion performance with a PCE of up to 9.2% (Figure 10D). In 2019, Ke et al. Used a mixed cation of methylammonium (MA⁺) and formamidine (FA⁺) to assist film formation in the 2D DJ perovskite structure with (3AMP²⁺) as the spacer cation host, and found that the new 2D DJ 3A MP-based perovskite containing both mixed cations MA and FA in the perovskite structure had a narrower band gap than the perovskite containing only MA, and this change was attributed to the larger Pb/I/Pb angle in the structure (Figure 10e)^[53]. DJ perovskite with mixed A-site cations shows a preferred (011) orientation, which can promote vertical growth on the substrate (Figure 10 f). This means minimizing charge depletion and improving charge transport. Therefore, the solar cell based on 3AMP perovskite with mixed MA/FA cations has higher short-circuit current density (J_sc) and PCE (Figure 10G). When the FA content is 25%, the recombination rate of 3AMP perovskite is lower, and the FF and V_oc of the device are higher. Thus, the (3AMP)(MA_0.75FA_0.25)₃Pb₄I₁₃ solar cell achieves the best PCE of 12.04% with FF as high as 81.04% and average PCE of 9.83% ± 0.94%. In 2021, Yin et al. Studied the hot carrier cooling process of two different 2D DJ perovskites, (3AMP)PbI₄ and (4AMP)PbI₄, and demonstrated that the Rashba band splitting has a significant effect on the hot carrier cooling dynamics of 2D DJ perovskites^[54]. The existence of Rashba band splitting in (4AMP)PbI₄ has been verified by electronic band structure calculations and temperature-dependent photoluminescence spectra and PL decay line measurements, which can slow down the cooling process by controlling the spin state of hot carriers. The Rashba band splitting in 2D perovskites is a powerful tool to control the hot carrier cooling dynamics, and this analysis can be used to design and synthesize 2D materials with excellent optoelectronic properties.

4.2 Symmetric bis-ammonium structure

The bisammonium group reported above is an asymmetric structure for the long alkyl chain. In order to further expand the DJ material, Ahmad et al. Introduced an organic cation (PDA²⁺) containing a symmetrical ammonium group structure at both ends of the alkyl chain. The symmetrical bisammonium group structure can remove the van der Waals interaction between the layers of the RP-type 2D perovskite (Figure 11A)^[55]. The results show that the van der Waals gap is eliminated by the introduction of diammonium cations, and the strong hydrogen bonds between the organic layer and the perovskite layers on both sides maintain the stability of the structure. Compared with the RP structure,The DJ structure does not have the dislocation and distortion characteristics of the crystal structure, which makes the perovskite layers closely packed, resulting in the increase of crystallinity, phase purity and charge mobility of DJ phase 2D perovskite. The maximum PCE of the device based on (PDA)(MA)₃Pb₄I₁₃ is 13. 3%, and the unpackaged device can still retain about 95% of the initial PCE after exposure to ambient air (40% ~ 70% relative humidity) for 4000 H, damp heat test (85 ℃ and 85% relative humidity) for 168 H, and illumination for 3000 H, respectively. The PSCs based on the DJ phase 2D perovskite have higher efficiency and stability than those based on the RP phase. Subsequently, a series of efficient and stable symmetric DJ phase 2D perovskites followed, and Cohen et al. Proposed Dion Jacobson two-dimensional perovskites using phenyldimethylammonium as a barrier molecule^[56]. The results show that the bis-ammonium molecules determine the orientation of the perovskite layer, and efficient 2D perovskite solar cells can be fabricated without any additives or other additional treatments. The efficiency of the 2D DJ perovskite cell device can reach 15.6%, which is one of the highest efficiencies reported for low-dimensional perovskite photovoltaic devices so far. Compared with the 3D structure, the stability of 2D perovskite is obviously improved. Because the crystal growth direction is parallel to the substrate, the efficiency of DJ phase PSCs is often limited by the poor charge transport between perovskite layers, and Li et al. Promoted perovskite crystal growth direction regulation by using NH₄SCN additives^[57]. It was found that different amounts of additives could effectively adjust the crystal orientation of the DJ phase (BDA)(MA)₄Pb₅I₁₆(n=5) perovskite film on the surface of the PTAA substrate from random to perpendicular (fig. 11B). The results of X-ray diffraction (XRD) and grazing incidence wide-angle X-ray scattering (GIWAXS) measurements show that the crystal structure of the whole layered perovskite is vertical and uniform from top to bottom. The reason is that the electrostatic interaction between the substrate and the precursor colloid affects the vertical orientation of the quasi-two-dimensional perovskite. The transient absorption (TA) results also show that the additive NH₄SCN can react with the precursor by ion exchange, which leads to the change of the size of colloidal particles and the membrane structure, resulting in the reduction of the number of layers concentrated in the range of n = 3 and n = 4. The highest PCE of the vertically grown 2D DJ layered perovskite device is 14.53%, which is one of the highest values reported for 2D PSCs fabricated at room temperature. Moreover, the inverted PSCs also showed very good environmental stability, maintaining 85% of their initial PCE for 900 H at 50% ± 5% relative humidity. It has been mentioned earlier that 2D ~ PR phase perovskite photovoltaic devices based on thienyl-based spacer cations (ThFA⁺) have shown excellent photoelectric conversion performance. However, the use of thiophene derivatives as organic spacer cation materials has not been reported in 2D DJ phase perovskite PSCs. In 2020, Lu et al. Developed a thiophene-based spacer cation material organic ammonium salt ThDMAI (2,5 thiophene dimethyl ammonium hydroiodate), which was prepared by using DMF/DMSO (V ∶ V, 9 ∶ 1) mixed solvent, and could obtain high-quality 2D DJ perovskite (ThDMA)(MA)_n_-1Pb_n

I 3 n + 1

(n = 5) with high crystallinity, preferential vertical orientation, and prolonged carrier lifetime^[58]. It is found that the strong coordination molecule DMSO mainly acts as a Lewis base, which delays the crystallization rate, increases the grain size, induces the preferred vertical orientation of perovskite crystals, reduces the density of defect States and increases the carrier fluorescence lifetime (Fig. 11c). The optimized 2D perovskite has better film and device stability. The unpackaged 2D device can maintain more than 95% of its initial PCE after 1655 H of storage in N₂. Subsequently, Zhang et al. Elaborated the effect of casting process on the formation of DJ perovskite (PDMA)(MA)_n_-1Pb_n

I 3 n + 1

(< n > = 4, PDMA refers to 1,4 benzenedimethylammonium) thin film, and the solvent could be removed from the solution more quickly by hot casting or anti-solvent drop flushing, resulting in a more uniform thickness distribution of quantum wells^[59]. This ultimately greatly enhances the carrier transport efficiency in the vertical direction, which in turn increases the photoelectric conversion efficiency. The PCE of the hot-cast device is as high as 15.81% (Figure 11 d), which is also the highest efficiency of the aromatic spacer-based DJ perovskite photovoltaic device. Recently, Gao et al. Reported a new DJ perovskite (mPDA)MA_n_-1Pb_n

I 3 n + 1

, using single crystal X-ray observation to show that the I · I interlayer distance is 4.00 – 4.0404 Å in the n = 2 and 3 structures, which is the shortest among DJ perovskites^[60]. At the same time, the Pb I Pb angle is also as small as 158 ~ 160 °. These changes reflect the large deformation of the inorganic framework, resulting in a larger band gap of the material than other DJ analogues, enriching the family of large band gap DJ perovskite materials.

In perovskite materials, methylammonium (MA) cations were initially used to prepare. However, the photothermal stability of MA-based devices is poor, and ions are easy to migrate, so the removal of MA is one of the trends in the development of perovskite technology. In 2020, Cheng et al. Developed a new type of all-FA-based 2D DJ-phase perovskite in the form of (PDA)(FA)_n_-1Pb_n

I 3 n + 1

(n = 2, 3, 4), with FA as the central cation and propane diammonium (PDA) as the interlayer cation^[61]. The ammonium moieties in the PDA molecule are connected by long carbon chains and hydrogen bonds to the adjacent Pb and I octahedral layers to form a stable framework structure. By introducing FACl as an additive during the film preparation, the ion-exchange behavior performed between I^- and Cl^- effectively eliminated the residues of non-perovskite δ-phase FAPbI₃ and PbI₂, resulting in a well-oriented two-dimensional perovskite film (Fig. 12 a). The effect of Cl was studied by grazing incidence X-ray diffraction (GI-XRD), and it was found that the Cl-containing intermediate could not be preserved for a long time, but existed in a transient form and disappeared after annealing. The final perovskite film has well-oriented large grains, a dense pinhole-free surface, and excellent optoelectronic properties. The device based on the optimized (PDA)FA₃Pb₄I₁₃(n=4) shows a photoelectric conversion efficiency of 13.8%, which is the highest efficiency of FA-based 2D PSCs. In 2021, Su et al. Also followed the same strategy, using FA⁺ with better stability to replace MA⁺, effectively broadening the spectral absorption range and improving the stability of perovskite materials^[62]. At the same time, combined with the precursor additive strategy, the high-quality MA-free quasi-2D DJ perovskite thin film (BDA)FA₄Pb₅I₁₆ was prepared. The incorporation of CDTA results in a more favorable gradient phase distribution and gradient band alignment, which is beneficial to the carrier transport and collection process. The increased grain size allows both crystallinity and crystal orientation to be optimized, helping to reduce the defect density.

The synergistic effect of these advantages makes the PCE of the CDTA-modified device as high as 16.07% (Figure 12b), which is one of the highest PCEs reported for MA-free benchmark 2D DJ PSCs. In the same year, Su et al., based on the (BDA)FA₄Pb₅I₁₆ of two-dimensional DJ phase perovskite without methylammonium, completed the in-situ exchange reaction of I^- and Br^- ions to realize gradient Br doping (GBD) by FABr post-treatment of perovskite thin film, and constructed a gradient band alignment strategy^[63]. First, the gradient band alignment due to GBD can facilitate the transport, extraction, and transfer of carriers (Figure 12C). Secondly, after FABr treatment, the crystallinity is improved and the defect density is reduced due to the recrystallization process. Finally, the addition of Br mainly contributes to the stability of the device.

As a result, PCE and stability increased simultaneously. Compared with the pristine device (BDA)FA₄Pb₅I₁₆(13.78%), the device with GBD achieves a higher PCE of 16.75% (BDA)FA₄Pb₅I_16-_xBr_x (Fig. The significant increase of 12d),V_oc from 0.970 V to 1.107 V,V_oc is attributed to the suppression of non-radiative recombination at the bulk and interface. Finally, the DJ-type perovskite structure has attracted more attention because of its better structural stability and better carrier transport ability due to the elimination of van der Waals interaction.

5 ACI phase perovskite

Compared with 3D perovskites, 2D RP-phase perovskites show great advantages in stability, because the long alkyl chain organic cations in RP perovskites are more hydrophobic than the short chain MA cations commonly used in 3D perovskites. However, the charge is mainly transferred in the Pb-I inorganic layer, and the long organic cation also hinders the charge transfer, increasing the band gap and exciton binding energy, which are unfavorable for the charge collection efficiency. Moreover, the introduction of bisammonium-based cations into 3D perovskites resulted in the formation of a series of DJ phase perovskites, which exhibited better stability compared with RP perovskites due to the elimination of weak van der Waals interactions and the strengthening of two-dimensional framework structure by hydrogen bonds. However, the PCE of perovskite photovoltaic devices is still poor due to the low dissociation degree of photogenerated carriers in the DJ structure. In order to solve these problems, partial replacement of long-chain organic cations with smaller volume of cations can shorten the distance between inorganic layers, which can improve the dissociation degree of photogenerated carriers in perovskite and theoretically achieve excellent device performance with good stability. Therefore, in addition to the RP and DJ two-dimensional perovskite structures described above, the structural characteristics of RP and DJ perovskites can also be combined to form ACI (Alternating Cations in the Interlayer Space) type 2D layered perovskites with Alternating cation structures by inserting two different Cations between the inorganic layers. In 2017, Kanatzidis et al. First created a new 2D halide perovskite with a molecular formula of (C(NH₂)₃)(CH₃NH₃)_nPb_nI₃_n₊₁ by using guanidinium cation as a spacer, adding a new member to the 2D perovskite family^[64]. It is characterized by the inclusion of alternating cations in the inorganic interlayer (Figure 13 a), and the structure is also known as ACI two-dimensional perovskite. When n = 1,2,3, the crystals of GA(MA)_nPb_n

I 3 n + 1

are red, deep red and black, and their optical band gaps are 2. 27 eV, 1.99 eV and 1. 73 eV, respectively. Compared with typical RP perovskites, these materials have narrower band gaps, and their optical gaps show obvious red shifts of 80, 90 and 230 meV, respectively, in the corresponding n = 1 ~ 3 range. The band gap reduction of ACI perovskite is attributed to its distorted crystal structure by analysis. This trend is successfully confirmed by theoretical calculation. The change of dielectric constant in quantum wells with different structures is analyzed, and the exciton binding energy is calculated to be very small, which is more conducive to charge transport. The maximum PCE = 7.26% and the optimal FF ~ 80% were obtained by using GAMA₃Pb₃I₁₀ perovskite with the smallest band gap to make planar structure devices. Perovskites with two different alternating cations between layers continue to be the focus of attention due to their unprecedented special structure, and the crystallization process of materials deserves further study and analysis. In 2019, aiming at the film formation kinetics of ACI-type 2D perovskite, Zhang et al. Systematically studied the dynamic transformation of the film during the formation process and how to control the morphology and various n-phase distributions of 2D perovskite^[65]. Antisolvent engineering and hot casting technology were used to control the transition from disordered colloid to ordered crystal phase in the precursor solution based on ACI perovskite (GA)(MA)_nPb_nI₃_n₊₁(n=3) materials, and how the kinetic transition process affected the film quality and photovoltaic performance was studied. The complex film formation process was revealed by GIWAX in situ, from the disordered colloid to the intermediate phase, and finally to the 2D ACI perovskite. During the thermal annealing process, part of the GA₂PbI₄ intermediate phase disintegrated to form a mixture of one-dimensional GAPbI₃ and ACI perovskite (Figure 13b). The formation of the intermediate phase was affected by the complex interaction between lead halide, organic cations and solvent. The formation of various n-phases is controlled and the photoelectric properties are changed by adjusting the intermediate phase. The distribution and internal charge transfer of different n phases were studied by transient absorption spectroscopy (TA). The effects of film formation kinetics on the morphology and optoelectronic properties of the films were further studied. A high PCE of 14.7% was finally achieved for photovoltaic devices based on high-quality ACI perovskite thin films, emphasizing the importance of the regulation of intermediate phases for high-quality thin films and high-performance photovoltaic devices. Subsequently, Gu et al. Demonstrated a nanoscale hybrid multidimensional ACI perovskite structure with preferred crystal orientation grown along the [Pb_n

I 3 n + 1

] inorganic layer and perpendicular to the substrate based on multidimensional (GA)(MA)₃Pb₃I₁₀ perovskite material analysis^[66]. High-resolution scanning transmission electron microscopy (STEM) shows that in each microcrystalline region, the low-dimensional ACI structure crystallizes in a 3D perovskite nanoscale network. This nanoscale heterojunction facilitates an ultrafast (~ 0.3 ps) carrier fast energy transfer process from the ACI to the 3D perovskite and subsequent efficient carrier extraction from the 3D mesh to the transport layer (Figure 13C), optimizing the charge transport channel. Planar PSCs with high PCE (16.65%) and high stability (80% initial efficiency maintained after 2400 H in the glove box without encapsulation) were achieved based on the optimized ACI film. These results provide new insights into the crystal growth and carrier dynamics of 2D ACI. In the same year, Li et al., in order to improve the absorption capacity and carrier dissociation efficiency of the photoactive layer, reported an interlayer space alternating cation structure (B ∥ ACI) perovskite containing both double ammonium (BEA²⁺) and single ammonium (MA⁺) structures, with the expression of (BEA)_0.5MA_nPb_n

I 3 n + 1

^[67]. Unlike traditional ACI perovskites, B ACI perovskites have hydrogen bonds (between the BEA and the inorganic layer), resulting in a more stable layered perovskite structure. In addition, B ACI perovskite uses fewer long-chain organic ligands, which can significantly improve the carrier transport properties compared with DJ perovskite. Therefore, in theory, B ACI perovskite has the advantages of both ACI and DJ perovskite. Compared with the typical RP perovskite, the B/ACI perovskite has a wide absorption range, which is characterized by ultraviolet photoelectron spectroscopy (UPS) and ultraviolet-visible absorption spectroscopy. Density functional theory (DFT) calculations show that (BEA)_0.5MA₃Pb₃I₁₀ is a direct gap semiconductor with a lower exciton binding energy compared to (BA)₂MA₂Pb₃I₁₀. At the same time, the ultrafast transient absorption (TA) also proves that the B ACI perovskite structure forms a narrow quantum well width distribution, resulting in an efficient carrier transport path. As a result, the barrier for charge transport is effectively lowered, enhancing the diffusion length of carriers (electrons and holes) over 350 nm and lifting the carrier mobility. Finally, the efficiency of the device based on the (BEA)_0.5Cs_0.15(FA_0.83MA_0.17)_2.85Pb₃(I_0.83Br_0.17)₁₀structure reached 17.39% (Fig. 13d). In addition, the tri-cation B ACI device can maintain more than 90% of its initial efficiency when stored for 2400 H under ambient atmospheric conditions, and does not show significant degradation under continuous illumination for more than 500 H. The above results indicate that ACI-type perovskite is a promising potential star semiconductor material structure for improving the performance and stability of low-dimensional perovskite devices.

6 Conclusion and prospect

Layered 2D perovskites have received much attention from researchers due to their good environmental stability. In this paper, the properties and development of 2D perovskites with different structures, such as RP, DJ and ACI, are introduced in detail, and the key issues affecting the performance of 2D PSCs in recent years are summarized, such as the vertical orientation distribution of crystals, the distribution of quantum wells, the exploration of organic spacer cations, and the improvement of device engineering technology. Specifically, it can be summarized into the following four aspects: (1) Generally, the orientation of 2D perovskite depends on the QW thickness. Quantum wells with small values of n are more likely to grow parallel to the substrate direction, while thicker wells (n > 4) tend to grow in a perpendicular orientation. However, the charge transport is mainly distributed in the direction of the inorganic layer, so one of the key factors for high-performance devices is to orient the 2D perovskite layer perpendicular to the electrode to achieve efficient charge transport. At the same time, finding a way to horizontally orient two-dimensional perovskite is also important for optimizing devices such as field effect transistors. (2) In addition to the phase distribution in the vertical direction, the QW width distribution also has an important effect on charge transfer and recombination. Unlike 3D perovskite thin films, 2D perovskite thin films are a mixture of multiple perovskite phases with different n values rather than a homogeneous perovskite phase with the same n value. These multiple phases exhibit a spontaneous arrangement, with the perovskite phase with small n located at the bottom of the film and the perovskite phase with large n located at the top of the film. Light-generated excitons in the quantum well with small n value will transfer energy to the quantum well with large n value, thus realizing efficient radiative recombination, which can be used to prepare light-emitting diodes. On the other hand, the continuous electron and hole transfer from the small n to the large n perovskite phase is beneficial to their application in the inversion p _ I _ n structure PS/Cs. (3) Organic spacer cations have an important effect on the properties, performance, and stability of 2D perovskites. Aromatic organoammonium with π-electron delocalization in the conjugated backbone is expected to enhance charge transport, visible light absorption, and have smaller exciton binding energy, thereby improving the current and efficiency of photovoltaic devices. In-depth study of the structure and crystallization kinetics, crystal growth orientation, phase distribution and photoelectric properties of organic spacer cations will help to synthesize two-dimensional perovskites with more excellent properties. (4) Component engineering is also an effective strategy to improve the performance and stability of two-dimensional perovskite solar cells by using multi-cation mixtures or halogen replacement. Interface engineering and additive engineering can effectively regulate the matching of interlayer energy level structure, and play a key role in controlling film morphology, phase distribution, charge collection and recombination. Finally, the two-dimensional layered structure can significantly improve the device stability due to its natural multiple quantum well structure and the larger molecular weight of the alkyl ammonium component, which makes it more hydrophobic. It is important to find a balance between hydrophobicity and charge transport efficiency.

Appendix

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	Kojima A, Teshima K, Shirai Y, Miyasaka T. J. Am. Chem. Soc., 2009, 131(17): 6050.

[2]	Kim H S, Lee C R, Im J H, Lee K B, Moehl T, Marchioro A, Moon S J, Humphry-Baker R, Yum J H, Moser J E, Grätzel M, Park N G. Sci. Rep., 2012, 2: 591.

[3]	Lee Michael M, Teuscher J, Miyasaka T, Murakami Takurou N, Snaith Henry J. Science, 2012, 338: 643.

[4]	Zhou H P, Chen Q, Li G, Luo S, Song T B, Duan H S, Hong Z R, You J B, Liu Y S, Yang Y. Science, 2014, 345(6196): 542.

[5]	Leijtens T, Bush K A, Prasanna R, McGehee M D. Nat. Energy, 2018, 3(10): 828.

[6]	Yang W S, Noh J H, Jeon N J, Kim Y C, Ryu S, Seo J, Seok S I. Science, 2015, 348(6240): 1234.

[7]	Saliba M, Matsui T, Domanski K, Seo J Y, Ummadisingu A, Zakeeruddin S M, Correa-Baena J P, Tress W R, Abate A, Hagfeldt A, Grätzel M. Science, 2016, 354(6309): 206.

[8]	Bi D Q, Li X, Milić J V, Kubicki D J, Pellet N, Luo J S, LaGrange T, Mettraux P, Emsley L, Zakeeruddin S M, Grätzel M. Nat. Commun., 2018, 9: 4482.

[9]	Hu Y H, Hutter E M, Rieder P, Grill I, Hanisch J, Aygüler M F, Hufnagel A G, Handloser M, Bein T, Hartschuh A, Tvingstedt K, Dyakonov V, Baumann A, Savenije T J, Petrus M L, Docampo P. Adv. Energy Mater., 2018, 8(16): 1870073.

[10]	Min H, Kim M, Lee S U, Kim H, Kim G, Choi K, Lee J H, Seok S I. Science, 2019, 366(6466): 749.

[11]	Slavney A H, Smaha R W, Smith I C, Jaffe A, Umeyama D, Karunadasa H I. Inorg. Chem., 2017, 56(1): 46.

[12]	Misra R K, Cohen B E, Iagher L, Etgar L. ChemSusChem, 2017, 10(19): 3712.

[13]	Smith I C, Hoke E T, Solis-Ibarra D, McGehee M D, Karunadasa H I. Angew. Chem. Int. Ed., 2014, 53(42): 11232.

[14]	Li X T, Hoffman J M, Kanatzidis M G. Chem. Rev., 2021, 121(4): 2230.

[15]	Krishna A, Gottis S, Nazeeruddin M K, Sauvage F. Adv. Funct. Mater., 2019, 29(8): 1806482.

[16]	Goldschmidt V M. Naturwissenschaften, 1926, 14(21): 477.

[17]	Saparov B, Mitzi D B. Chem. Rev., 2016, 116(7): 4558.

[18]	Blancon J C, Even J, Stoumpos C C, Kanatzidis M G, Mohite A D. Nat. Nanotechnol., 2020, 15(12): 969.

[19]	Mao L L, Stoumpos C C, Kanatzidis M G. J. Am. Chem. Soc., 2019, 141(3): 1171.

[20]	Beznosikov B V, Aleksandrov K S. Crystallogr. Rep., 2000, 45(5): 792.

[21]	Ruddlesden S N, Popper P. Acta Crystallogr., 1958, 11(1): 54.

[22]	Cheng Z Y, Lin J. CrystEngComm, 2010, 12(10): 2646.

[23]	Ishihara T, Hong X, Ding J, Nurmikko A V. Surf. Sci., 1992, 267(1/3): 323.

[24]	Hong X, Ishihara T, Nurmikko A V. Phys. Rev. B, 1992, 45(12): 6961.

[25]	Era M, Morimoto S, Tsutsui T, Saito S. Appl. Phys. Lett., 1994, 65(6): 676.

[26]	Hattori T, Taira T, Era M, Tsutsui T, Saito S. Chem. Phys. Lett., 1996, 254(1/2): 103.

[27]	Zou Y, Yuan Z, Bai S, Gao F, Sun B. Mater. Today Nano, 2019, 5: 100028.

[28]	Zhu T, Yang Y R, Gong X. ACS Appl. Mater. Interfaces, 2020, 12(24): 26776.

[29]	Yang W F, Igbari F, Lou Y H, Wang Z K, Liao L S. Adv. Energy Mater., 2020, 10(13): 1902584.

[30]	Zhu H L, Choy W C H. Sol. RRL, 2018, 2(10): 1800146.

[31]	Cao D H, Stoumpos C C, Farha O K, Hupp J T, Kanatzidis M G. J. Am. Chem. Soc., 2015, 137(24): 7843.

[32]	Tsai H, Nie W Y, Blancon J C, Stoumpos C C, Asadpour R, Harutyunyan B, Neukirch A J, Verduzco R, Crochet J J, Tretiak S, Pedesseau L, Even J, Alam M A, Gupta G, Lou J, Ajayan P M, Bedzyk M J, Kanatzidis M G, Mohite A D. Nature, 2016, 536(7616): 312.

[33]	Quintero-Bermudez R, Gold-Parker A, Proppe A H, Munir R, Yang Z Y, Kelley S O, Amassian A, Toney M F, Sargent E H. Nat. Mater., 2018, 17(10): 900.

[34]	Shang Q Y, Wang Y N, Zhong Y G, Mi Y, Qin L, Zhao Y F, Qiu X H, Liu X F, Zhang Q. J. Phys. Chem. Lett., 2017, 8(18): 4431.

[35]	Zhang J, Qin J J, Wang M S, Bai Y J, Zou H, Keum J K, Tao R M, Xu H X, Yu H M, Haacke S, Hu B. Joule, 2019, 3(12): 3061.

[36]	Liu T F, Jiang Y Y, Qin M C, Liu J X, Sun L L, Qin F, Hu L, Xiong S X, Jiang X S, Jiang F Y, Peng P, Jin S Y, Lu X H, Zhou Y H. Nat. Commun., 2019, 10: 878.

[37]	Caiazzo A, Datta K, Jiang J K, GÉlvez-Rueda M C, Li J Y, Ollearo R, Vicent-Luna J M, Tao S X, Grozema F C, Wienk M M, Janssen R A J. Adv. Energy Mater., 2021, 11(42): 2102144.

[38]	Zhang X, Ren X D, Liu B, Munir R, Zhu X J, Yang D, Li J B, Liu Y C, Smilgies D M, Li R P, Yang Z, Niu T Q, Wang X L, Amassian A, Zhao K, Liu S F. Energy Environ. Sci., 2017, 10(10): 2095.

[39]	Zhu X J, Xu Z, Zuo S N, Feng J S, Wang Z Y, Zhang X R, Zhao K, Zhang J, Liu H R, Priya S, Liu S F, Yang D. Energy Environ. Sci., 2018, 11(12): 3349.

[40]	Passarelli J V, Mauck C M, Winslow S W, Perkinson C F, Bard J C, Sai H, Williams K W, Narayanan A, Fairfield D J, Hendricks M P, Tisdale W A, Stupp S I. Nat. Chem., 2020, 12(8): 672.

[41]	Dong Y X, Lu D, Xu Z Y, Lai H T, Liu Y S. Adv. Energy Mater., 2020, 10(28): 2000694.

[42]	Xi J, Spanopoulos I, Bang K, Xu J, Dong H, Yang Y G, Malliakas C D, Hoffman J M, Kanatzidis M G, Wu Z X. J. Am. Chem. Soc., 2020, 142(46): 19705.

[43]	Ren H, Yu S D, Chao L F, Xia Y D, Sun Y H, Zuo S W, Li F, Niu T T, Yang Y G, Ju H X, Li B X, Du H Y, Gao X Y, Zhang J, Wang J P, Zhang L J, Chen Y H, Huang W. Nat. Photonics, 2020, 14(3): 154.

[44]	Wu G B, Yang T H, Li X, Ahmad N, Zhang X N, Yue S L, Zhou J, Li Y X, Wang H, Shi X H, Liu S Z, Zhao K, Zhou H Q, Zhang Y. Matter, 2021, 4(2): 582.

[45]	Wei Y, Chen B Q, Zhang F, Tian Y Y, Yang X C, Cai B, Zhao J J. Sol. RRL, 2021, 5(4): 2000661.

[46]	Huang Y W, Li Y H, Lim E L, Kong T F, Zhang Y, Song J, Hagfeldt A, Bi D Q. J. Am. Chem. Soc., 2021, 143(10): 3911.

[47]	Quan L N, Yuan M J, Comin R, Voznyy O, Beauregard E M, Hoogland S, Buin A, Kirmani A R, Zhao K, Amassian A, Kim D H, Sargent E H. J. Am. Chem. Soc., 2016, 138(8): 2649.

[48]	Wang Z P, Lin Q Q, Chmiel F P, Sakai N, Herz L M, Snaith H J. Nat. Energy, 2017, 2(9): 17135.

[49]	Ye T, Bruno A, Han G F, Koh T M, Li J, Jamaludin N F, Soci C, Mhaisalkar S G, Leong W L. Adv. Funct. Mater., 2018, 28(30): 1801654.

[50]	Lee J W, Dai Z H, Han T H, Choi C, Chang S Y, Lee S J, de Marco N, Zhao H X, Sun P Y, Huang Y, Yang Y. Nat. Commun., 2018, 9: 3021.

[51]	Mao L L, Ke W J, Pedesseau L, Wu Y L, Katan C, Even J, Wasielewski M R, Stoumpos C C, Kanatzidis M G. J. Am. Chem. Soc., 2018, 140(10): 3775.

[52]	Li X T, Ke W J, TraorÉ B, Guo P J, Hadar I, Kepenekian M, Even J, Katan C, Stoumpos C C, Schaller R D, Kanatzidis M G. J. Am. Chem. Soc., 2019, 141(32): 12880.

[53]	Ke W J, Mao L L, Stoumpos C C, Hoffman J, Spanopoulos I, Mohite A D, Kanatzidis M G. Adv. Energy Mater., 2019, 9(10): 1803384.

[54]	Yin J, Naphade R, Maity P, GutiÉrrez-Arzaluz L, Almalawi D, Roqan I S, BrÉdas J L, Bakr O M, Mohammed O F. Nat. Commun., 2021, 12: 3995.

[55]	Ahmad S, Fu P, Yu S W, Yang Q, Liu X, Wang X C, Wang X L, Guo X, Li C. Joule, 2019, 3(3): 794.

[56]	Cohen B E, Li Y M, Meng Q B, Etgar L. Nano Lett., 2019, 19(4): 2588.

[57]	Li F Z, Zhang J, Jo S, Qin M C, Li Z, Liu T T, Lu X H, Zhu Z L, Jen A K Y. Small Methods, 2020, 4(5): 1900831.

[58]	Lu D, Lv G W, Xu Z Y, Dong Y X, Ji X F, Liu Y S. J. Am. Chem. Soc., 2020, 142(25): 11114.

[59]	Zhang X, Yang T H, Ren X D, Zhang L, Zhao K, Liu S F. Adv. Energy Mater., 2021, 11(19): 2002733.

[60]	Gao L L, Li X T, TraorÉ B, Zhang Y L, Fang J J, Han Y, Even J, Katan C, Zhao K, Liu S Z, Kanatzidis M G. J. Am. Chem. Soc., 2021, 143(31): 12063.

[61]	Cheng L, Liu Z, Li S D, Zhai Y F, Wang X, Qiao Z, Xu Q F, Meng K, Zhu Z Y, Chen G. Angew. Chem. Int. Ed., 2021, 60(2): 856.

[62]	Su P Y, Bai L, Bi H, Liu B B, He D M, Wang W Q, Cao X B, Chen S, Lee D, Yang H, Zang Z G, Chen J Z. ACS Appl. Mater. Interfaces, 2021, 13(25): 29567.

[63]	Su P Y, Bai L, Bi H, Liu B B, Chen S, Lee D, Yang H, Chen C, Zang Z G, Chen J Z. J. Power Sources, 2021, 506: 230213.

[64]	Soe C M M, Stoumpos C C, Kepenekian M, TraorÉ B, Tsai H, Nie W Y, Wang B H, Katan C, Seshadri R, Mohite A D, Even J, Marks T J, Kanatzidis M G. J. Am. Chem. Soc., 2017, 139(45): 16297.

[65]	Zhang Y L, Wang P J, Tang M C, Barrit D, Ke W J, Liu J X, Luo T, Liu Y C, Niu T Q, Smilgies D M, Yang Z, Liu Z K, Jin S Y, Kanatzidis M G, Amassian A, Liu S F, Zhao K. J. Am. Chem. Soc., 2019, 141(6): 2684.

[66]	Gu H, Liang C, Xia Y D, Wei Q, Liu T H, Yang Y G, Hui W, Chen H R, Niu T T, Chao L F, Wu Z H, Xie X J, Qiu J, Shao G S, Gao X Y, Xing G C, Chen Y H, Huang W. Nano Energy, 2019, 65: 104050.