Jiahui
Cheng
,
Huijie
Cao
,
Shuming
Zhang
,
Fang
Yue
* and
Zhongmin
Zhou
*
College of Chemistry and Molecular Engineering, Qingdao University of Science and Technology, Qingdao 266042, China. E-mail: fangyue@qust.edu.cn; zhouzm@qust.edu.cn
First published on 28th November 2023
Perovskite solar cells (PSCs) have gained significant attention in recent years due to their low fabrication cost, solution-based processing, potential for flexibility and large-scale solar energy conversion. The performance of PSCs depends on effectively separating and collecting the photogenerated charge carriers before they recombine and lose excess energy. Therefore, efficient carrier extraction is crucial for achieving high-performance PSCs. The built-in electric field (BEF) within PSCs serves as the driving force for extracting carriers to their corresponding electrodes. Reinforcing the BEF can reduce carrier recombination in the bulk phase/interface, facilitate carrier separation and extraction, and minimize energy losses in PSCs. Up to now, numerous efforts have been made to enhance the BEF and further improve the photovoltaic performance of PSCs. This review first describes the carrier separation and extraction mechanism influenced by the BEF. It then summarizes various strategies used to enhance the BEF in PSCs, including forming semiconductor junctions, regulating energy levels through dipole layers and functional layer modifications, constructing a ferroelectricity polarization electric field (PEF) within PSCs, and mitigating the surface charge accumulation caused by ion migration. Finally, this review discusses the prospects for designing an enhanced BEF to promote the continuous development of high-performance PSCs. This review provides useful insights for further analyzing the mechanism of the BEF in PSCs and achieving breakthroughs in the efficiency of PSC devices through field effect passivation.
In principle, the high efficiency of PSCs strongly depends on the efficient separation and extraction of photogenerated charge carriers to the corresponding electrodes. The built-in asymmetry arising from appropriate contacts in photovoltaic devices creates a driving force for carrier extraction. Commonly, the perovskite photoactive absorbing layer absorbs the energy of light and then produces electron–hole pairs (e–h pairs). Owing to the weak exciton binding energy, the e–h pairs are easily separated and migrate directionally under the driving force, and are subsequently extracted at the negative and positive poles through the electron transport layer (ETL) and the hole transport layer (HTL), respectively.10 There are two main mechanisms for charge extraction in solar cells, including diffusion and drift.11 The charge extraction mechanism based on diffusion is reliant on the establishment of kinetic selectivity at the contacts and long carrier diffusion length of the active layer.12 As for the drift-based mechanism, the charge extraction is driven by the built-in electric field (BEF). It is necessary to establish a drift–diffusion model to study the carrier separation mechanism in PSCs. In the diffusion-dominated PSC model, the BEF is not the main driving force for carrier separation. The selective contact of interface can ensure the effective extraction of the diffusion-driven carriers, but the BEF is still an important driving force for the directional transport of carriers.13,14 Therefore, an in-depth understanding of the source and mechanism of the BEF in PSCs is a prerequisite for obtaining high-performance PSCs. This BEF is mainly determined by the difference in work functions (WFs) between two electrodes or charge transport layers.15 In general, PSCs are composed of a perovskite layer sandwiched between the HTL and the ETL. The difference in the majority carrier concentration results in the formation of a space charge region (depletion region) near the interface between the perovskite layer and the transport layer, which produces a BEF. Under the action of the BEF, the photogenerated carriers can realize directional drift and give rise to a photocurrent. Moreover, a built-in potential arising from the BEF could avoid forming a reverse electric field inside the active layer, which will ensure efficient charge extraction through the ETL/HTL.12
It has been well recognized that the serious charge recombination at the perovskite bulk and interface hinders the efficient charge extraction and thus restricts further improvement of the device performance.16 Generally, perovskite films prepared by a simple solution method are polycrystalline, which induces inevitable structural defects at grain boundaries (GBs). These structural defects can serve as non-radiative recombination centers for photogenerated carriers, which will cause severe non-radiative recombination and energy loss. Maintaining the advantage of facile fabrication of PSCs and minimizing the energy loss require more effort. Besides the defect passivation strategy, regulating the internal driving force for carrier separation and transmission has been developed to solve this problem.17–23 For PSCs, the existence of the BEF can reduce the possibility of charge carriers being trapped by defects and help the trapped charge carriers escape.24,25
In view of the important role of the BEF in the process of carrier separation, extraction and transport in PSCs, the intensity of the BEF is directly related to the performance of PSCs.26 In previous studies, many strategies to enhance the BEF have been proposed. For example, constructing p–n junctions or heterojunctions in the perovskite layer or at the perovskite/transport layer interface can increase additional depletion regions, resulting in an enhanced BEF in PSCs.27,28 Introducing a dipole layer at the interface to regulate the WF of the functional layer also can be used to increase the BEF.29 Furthermore, designing a ferroelectric intermediate layer of two-dimensional (2D) perovskites or introducing ferroelectric materials can form an additional polarization electric field (PEF) which promotes charge extraction and thus improves the performance of PSCs.30
Although significant effort has been made in the design of an enhanced BEF and high-performance PSCs, there is still a lack of a comprehensive review that summarizes these strategies through the core point of the BEF. In this review, we briefly introduce the carrier separation and extraction mechanism affected by the BEF and the distribution of the BEF in PSCs. Then, we summarize the strategies proposed in the past few years to enhance the BEF (Fig. 1), which are divided into four parts: constructing junctions in PSCs, regulating the energy level, constructing the PEF and mitigating charge accumulation caused by ion migration. Finally, we discuss the problems existing in the BEF regulation strategy and look forward to the prospect of BEF enhancement for the development of high-performance PSCs.
The typical device structures of PSCs can be categorized into two main types: mesoporous and planar structures. The latter also includes regular n–i–p and inverted p–i–n devices, where n, i, and p represent the ETL, the perovskite layer, and the HTL, respectively.32 The sandwich structure of PSCs endows them with excellent interface selective contact. The BEF is formed near the interface, which can drive the directional movement of charge carriers. In fact, the BEF arises from the difference of WFs between the two contacts. This established BEF would provide a driving force for interfacial electron–hole separation and their extraction to the corresponding electrodes to create photocurrent. Furthermore, the deep-level defects in the bulk perovskite layer and the interface between the perovskite layer and the transport layer can form harmful non-radiative recombination centers, leading to serious energy loss. A strong BEF is sufficient to support the escape of photogenerated excitons in deep-level defects. In addition, the BEF is often directly related to the VOC of PSCs.35 Therefore, the related fundamental understanding of the BEF is necessary for elucidating the working principle of PSCs and further optimizing their performance.36–38
To study the separation and extraction mechanism of carriers affected by the BEF, it is crucial to pinpoint the location of BEF distribution and its intensity in PSCs. Eddi et al. used electron beam-induced current (EBIC) technology and cross-sectional SEM to plot the changes of the current amplitude generated in the device in cross-section, which is useful for locating the driving force for charge separation and thus analyzing the working principle of PSC devices.31 In principle, e–h pairs are generated when a scanning beam of electrons is directed toward the light absorbing layer. In the presence of a driving force for charge separation and contacts to collect the charge carriers, a current is generated in the form of an EBIC signal. The intensity of the EBIC signal indicates the efficiency of charge separation and collection. The results show that there exist two main efficient carrier separation positions (Fig. 2): one near the absorber/hole barrier layer and the other near the absorber/electron barrier layer interface. This “bimodal” structure is a feature of p–i–n solar cells, which contain a semiconductor absorber layer with low doping and high electron quality between the p- and n-layers.
Fig. 2 (a) EBIC characterization of a flat CH3NH3PbI3−xClx device with hole transporting material (HTM). SE image and corresponding EBIC image of a cross-section of a solar cell without mesoporous alumina. Left, SE image; right, EBIC image. (b) EBIC characterization of a CH3NH3PbI3−xClx device with a thin mesoporous-Al2O3 without HTM. SE image and corresponding EBIC image of a cross-section of a solar cell with a thin mesoporous-Al2O3 and without HTM. Left, SE image; right, EBIC image.31 Copyright 2014, Springer Nature. |
Since EBIC measurement is based on the local excitation and extraction of charge carriers by an electron beam, it cannot truly simulate solar cells under normal operating or dark conditions. The potential distribution of PSCs measured based on EBIC is inaccurate and speculative. The potential distribution in PSCs can be directly detected by Kelvin probe force microscopy (KPFM) by detecting the contact potential difference (CPD) between the needle tip and the sample using a probe. In addition, the KPFM measurement method obtains nano-resolution imaging through a local contact, which is more advantageous in exploring the potential distribution of PSCs.39–43 The mechanisms of EBIC and KPFM are described in Fig. 3.
Fig. 3 (a) Illustration of an EBIC experiment on a cross-section of a thin-film solar cell. The electron beam generates e–h pairs in the absorber material, which, after separation, creates a current generation profile inside the cell. TCO, transparent conducting oxide; ETM, electron transporting materials.31 Copyright 2014, Springer Nature. (b) Schematic diagram of the KPFM measurement principle.42 Copyright 2014, Springer Nature. |
Based on the KPFM measurement, Bergmann et al. studied the real spatial cross-sectional distribution of internal potential in high-efficiency mesoporous perovskite solar cells, which proved that the electric field in PSCs is uniform. However, the charge transfer is imbalanced under lighting conditions. Electron extraction is faster than hole extraction, leading to hole accumulation at the interface between the perovskite layer and the HTL. Such hole accumulation results in the formation of a potential barrier in the front of the HTL, which will reduce JSC in PSCs.42
Furthermore, Cai et al. studied the control of the potential distribution in high-performance PSCs through KPFM technology.44 They prepared p-type (rich-MAI) and n-type (rich-PbI2) perovskite devices to investigate the effect of semiconductor doping type on the formation of diode junctions in PSCs. As shown in KPFM (Fig. 4), there is a diode junction and significant potential drop between mesoporous-TiO2 and perovskite film in mesoporous-TiO2 structured PSCs, where the main charge separation occurred. Meanwhile, this diode junction is almost unaffected by the self-doping of the perovskite layer. In addition, there are two diode junctions near the perovskite/HTL and perovskite/ETL interfaces in planar PSCs, which are affected by the doping of perovskite layers. The mechanism of the BEF in PSCs can be better understood by determining the number, position, and intensity of junctions in PSCs through measuring the potential distribution and potential changes.
Fig. 4 (a) KPFM measurements of planar PSCs. (b) and (c) KPFM characterization of the mesoporous type PSCs fabricated using MAI- and PbI2-rich precursors. (d) and (e) KPFM characterization of the planar type PSCs fabricated using MAI- and PbI2-rich precursors.44 Copyright 2018, Elsevier. |
The p–n junction is a frequently mentioned concept in semiconductor physics and is widely applied in conventional solar cells. Constructing additional p–n junctions in PSCs is a common strategy to enhance the BEF. In general, the majority of carriers of n-type semiconductors are electrons, and the majority of carriers of p-type semiconductors are holes. Thus, the difference in concentration would drive the majority of carriers to the low-concentration region when an n-type semiconductor is in contact with a p-type semiconductor. Under the influence of diffusion, the electrons go from the n-type region into the p-type region and the holes go from the p-type region into the n-type region. As for the p-type region, after the holes leave, the immovable ionized acceptors are no longer screened by the majority carriers (holes in the p-type material). Therefore, a negative charge region appears on the p-type material near the p–n junction. Similarly, a positive charge region composed of ionized donors is formed on the side of the n-region. The charges carried by these ionized donors and acceptors are usually called space charges.46 The transition region where the space charges exist is called the space-charge region, which is also called the depletion region. As a result, an electric field is produced in the space-charge region, which points from positive space charges to negative space charges. In addition to the p–n junction, the other semiconductor heterojunctions are also used for enhancing the BEF in PSCs.
Constructing additional junctions is an effective strategy to enhance the BEF of PSCs, which will increase the driving force for photogenerated charge extraction and thus improve the performance of PSCs. In this chapter, we summarize the recently proposed methods for constructing extra junctions, which comprise fabricating a perovskite–perovskite structure, the forming of gradient heterojunctions, and obtaining junctions directly inside PSCs through additives and other means.
Fig. 5 Cubic perovskite crystal structure.32 Copyright 2014, Springer Nature. |
Direct deposition of perovskite layers to achieve a double-layer structure is a common strategy to construct a p–n junction in PSCs. Combining a p-type perovskite and an n-type perovskite to form a p–n homojunction can generate an additional electric field and enhance the overall BEF in PSCs,27,47 which is beneficial for the extraction and transmission of charge carriers within PSCs. Since the perovskite layer is often obtained by a solution method in PSCs, it is difficult to avoid the first deposited perovskite layer being solvated by the second deposition of the perovskite layer. According to previous reports, vapor deposition can be used to solve this problem.53–55 Dänekamp et al. first prepared a perovskite–perovskite homojunction by a vacuum deposition method.47 By adjusting the relative deposition rates of MAI and PbI2, they obtained a perovskite junction containing p-type (MAI excess) and n-type (MAI defect) MAPbI3 (Fig. 6a and b). The CPD between two different types of perovskites was measured to be 250 mV (Fig. 6c and d), which indicates the obvious existence of potential change in the junction interior. The device based on two types of perovskite layers displays a high fill factor and high photovoltage, benefiting from efficient charge extraction and limited recombination in the presence of a perovskite homojunction. Furthermore, Li et al. utilized a combined deposition method to integrate a p–n homojunction into a planar structure (Fig. 6e–h).27 There is a significant potential change between the two layers of perovskite (Fig. 6i and g), which proves the formation of an internal electric field between them, resulting in an increase in the BEF of PSCs. The BEF in the homojunction induces the directional transport of photogenerated carriers, thereby reducing the carrier recombination loss.
Fig. 6 (a) IR relative transmittance spectra of the N–H stretching vibration for reference, n-type, and p-type MAPI films. (b) Scheme of the solar cell layout. (c) SKPM image of the cross section for a homojunction solar cell. (d) CPD profile of a p–n junction solar cell.47 Copyright 2018, American Chemical Society. (e) The bottom n-type perovskite layer is deposited by a one-step spin-coating method. The top p-type perovskite layer is deposited by vacuum vapour deposition followed by a dipping process. (f) The PSC with the homojunction composed by p-type and n-type perovskite layers. (g) and (h) surface potential maps (KPFM) of MAPbI3 films. (i) The energetic levels of the corresponding n-type (blue) and p-type (orange) perovskite films. (j) The carrier concentration (red symbols) and mobility (black symbols) for electrons (circle) and holes (square) in the perovskite films extracted by Hall measurements.27 Copyright 2019, Springer Nature. |
As for the low-dimension perovskite quantum dot films, the upper perovskite solvent does not technically dissolve the lower perovskite.56–59 During the preparation process, perovskite quantum dots are suspended in a colloidal solution through their surface ligands, and a layer of quantum dots is first deposited from the colloidal solution. The ligand is chemically removed and the resulting film is insoluble in a non-polar solvent. Thus, the subsequent quantum dot layer can be processed without affecting the lower layer film. Therefore, a perovskite double-layer structure can be prepared by quantum dot film deposition to construct a heterojunction and enhance the BEF. For example, Zhao et al. constructed an internal heterojunction by layer-by-layer deposition of perovskite quantum dots (Fig. 7a–c), which promoted charge extraction at the interface and improved photocarrier harvesting (Fig. 7d and e).59 Moreover, Li et al. prepared perovskite quantum dot solar cells with a double-layer structure (Fig. 7f).60 The combination of wide-bandgap α-CsPbI3 and narrow-bandgap FAPbI3 resulted in the formation of a heterojunction, which ensures more effective charge separation and extraction (Fig. 7g). The perovskite quantum dot deposition method provides a simple strategy for the preparation of double-layer perovskite PSCs, which is beneficial for the regulation of the BEF.
Fig. 7 (a)–(c) ToF–SIMS depth profile of samples with the structure and interface location. (d) and (e) Transient absorption of heterojunction QD films.59 Copyright 2019, Springer Nature. (f) Schematic illustration of the PSCs. (g) TA measurements for time-wavelength-dependent TA color maps and TA spectra at a different GSB as a function of delay time.60 Copyright 2019, American Chemical Society. |
The construction of 2D/3D heterojunctions in PSCs is also an efficient strategy to introduce an extra electric field for efficient charge extraction. As for 2D perovskite materials, the Goldsmith tolerance factor has relatively loose restrictions on organic cations. Therefore, 2D perovskite materials are usually achieved by combining large-sized organic cations, which broadens the band gap of 2D perovskite materials. The wide bandgap of the 2D perovskite determines its weak optical performance potential and is difficult to be used as a main light absorbing material. However, this structure makes the 2D perovskite have excellent humidity stability, so it has broad application prospects in PSCs.61–66 In the past few years, many high-performance PSCs have used two-dimensional (2D) or quasi-two-dimensional (quasi-2D) perovskites modified three-dimensional (3D) perovskites to construct 2D/3D perovskite heterojunction structures. This structure can effectively enhance the BEF and achieve defect passivation and effective band alignment. Therefore, the optical performance parameters and PCE of PSCs can be improved.
Many strategies have been reported for enhancing the BEF based on 2D/3D heterojunction structures. Grancini et al. innovatively constructed the ultra-stable 2D/3D (HOOC(CH2)4NH3)2PbI4/CH3NH3PbI3 perovskite junction of carbon-based structure.67 This work prepared a stable 2D/3D heterojunction in PSCs and improved the stability of PSCs. He et al. applied dodecyl ammonium bromide (DABr) to react with PbI2 in GBs, which formed a clear 2D/3D heterojunction and increased the intensity of the BEF (Fig. 8a).68 At the same time, the super-long chain organic molecules have a good hydrophobic effect, which can inhibit the penetration of water and effectively improve the humidity stability (Fig. 8b and c). Furthermore, Zhou et al. developed a novel amino-functionalized fullerene (C60) derivative added to the MAPbI3 perovskite as a spacer for a 2D/3D hybrid perovskite.69 The derivative of C60 increases the electron density in the 2D/3D heterojunction and induces additional BEF enhancement (Fig. 8d), which promotes electron transport in 2D/3D PSCs. In addition, the formation of a 2D/3D heterojunction effectively reduces the trap-state density (Fig. 8e and f), which is beneficial to avoid BEF loss.
Fig. 8 (a) Mott–Schottky plots of the pristine and 2D/3D devices. (b) Pristine and 2D/3D perovskite films exposed to ambient air with a relative humidity of 85%. Inset: contact angle images. (c) Normalized PCE of PSCs under wet conditions.68 Copyright 2022, Elsevier. (d) Schematic illustration of the BEF of the 2D/3D PSCs (inset) and Mott–Schottky curves. (e) DLTS spectra of 3D and 2D/3D PSCs. (f) Arrhenius plots obtained from DLTS signals.69 Copyright 2022, Wiley. |
The space charge region formed by the 2D/3D heterojunction can control the intensity of the BEF by adjusting the thickness of the 2D perovskite layer. However, the 2D perovskite obtained by the solution spin coating method will produce an additional metastable quasi-2D perovskite phase, which cannot effectively control the thickness.70 Jang et al. developed a solvent-free solid phase in plane growth (SIG) to deposit the 2D perovskite on a 3D perovskite, thus obtaining a stable and highly crystalline 2D (C4H9NH3)2PbI4 film (Fig. 9a) in PSCs and made the BEF can be designed through adjusting the width of the depletion region.71 The SIG-processed film consistently shows only one sharp peak that corresponds to a pure 2D perovskite with no formation of the quasi-2D perovskite phase (Fig. 9b), which implies that the 2D film can be grown on the 3D surface in any desired thickness by repetition of the SIG process. Finally, the BEF enhancement of PSCs (Fig. 9c) based on the 2D/3D heterojunction improves the separation and extraction of carriers. As a result, the device performance is improved, resulting in a VOC of 1.185 V and a certified steady-state efficiency of 24.35% (Fig. 9d). In this work, the thickness of the 2D perovskite layer can be approximated to the thickness of the space charge region according to the Anderson model,61 so the SIG method can directly control the BEF.
Fig. 9 (a) X-ray diffraction patterns of the control film (3D perovskite only), 2D/3D films developed by the SIG process under different temperature conditions, and a pure 2D perovskite film. (b) A cross-sectional SEM image of the SIG-processed 2D/3D device. (c) J–V curves of 3D/2D heterojunction PSCs prepared by the SIG method. (d) Motts–Schottky plots of PSCs prepared by the SIG method. (e) Estimated energy-level alignment in the 2D/3D heterojunction. (f) Schematic diagram of the 3D/CLP/2D perovskite device structure.71 Copyright 2021, Springer Nature. (g) AFM–IR images of 3D/2D and 3D/CLP/2D perovskite heterostructures before and after aging at 100 °C for 120 min. (h) Energy diagram of the 3D perovskite, 3D/CLP, 2D perovskite and spiro-OMeTAD. (i) Representative J–V curves of PSCs fabricated using 3D perovskite, 3D/2D, 3D/CLP, and 3D/CLP/2D perovskite heterostructures as the light absorbing layer.72 Copyright 2023, Springer Nature. |
Furthermore, the contact between the 2D and 3D layers will result in ion diffusion, leading to the gradual disappearance of the 2D layer. To ensure the stability of the 2D/3D structure, Luo et al. introduced cross-linked polymers (CLP) into the upper layer of the 3D perovskite layer, and then deposited the 2D perovskite layer through a vapor-assisted two-step method, forming a 3D/CLP/2D perovskite heterostructure (Fig. 9e and f). The CLP stabilizes the heterostructure by inhibiting the diffusion of cations between 2D and 3D perovskites (Fig. 9g). Meanwhile, the introduction of the CLP can optimize the energy level alignment (Fig. 9h). In this work, PSCs based on carbon electrodes achieved an efficiency of 21.2% (Fig. 9i).72
It is a challenge to construct gradient heterojunctions with precise controllable and ideal energy level distribution for enhancing the BEF and realizing high-performance PSCs.76 Various strategies have been reported to achieve high-quality gradient heterojunctions for PSCs. For example, Li et al. realized energy level gradient distribution by regulating the halogen ratio of CsPbX3 and formed a gradient heterojunction with MAPbI3 to prepare a hole-free device (Fig. 10a).77 The increase in the proportion of bromides will broaden the energy band and increase the valence band maximum (VBM) of the perovskite, which can expand the difference between the Fermi levels of the acceptor and the donor in the gradient heterostructure. Therefore, the multi-layer distribution meeting the gradient energy level orientation can be prepared (Fig. 10b). The existence of the gradient heterojunction enhances the BEF and provides an additional hole driving force, which is beneficial for the separation of electron–hole pairs in the perovskite light absorption layer.
Fig. 10 (a) Schematic diagram of solar cell configuration with graded CsPbBrxI3−x films. (b) Energy diagram of solar cells with graded CsPbBrxI3−x.77 Copyright 2017, Wiley. (c) Schematic diagram of the gradient structure of the 3D–2D–0D perovskite. (d) Mott–Schottky curves with/without dimensional gradient heterojunctions. (e) Energy-level diagram of multigrade CsPbBrI2 PSCs (top) and the carrier-transport mechanism in multigrade CsPbBrI2 PSCs (bottom).78 Copyright 2018, Wiley. |
Furthermore, gradient heterojunctions can be designed by low-dimensional perovskites. Zhang et al. prepared inorganic perovskite material CsPbBrI2 into bulk–nanosheet–quantum dots (3D–2D–0D) dimensional structures to obtain a multiple dimensional gradient heterojunction (Fig. 10c).78 The introduction of the 3D–2D–0D gradient junction optimized the energy level orientation (Fig. 10e) and enhanced the BEF of the PSC device (Fig. 10d). Zhu et al. introduced different Ln3+ (Yb3+, Ce3+, Eu3+, and Sm3+) doped CsPbBrCl2 perovskite quantum dots (PQDs) to prepare high-performance PSCs.79 In this work, Ln3+ was doped in CsPbBrCl2 PQDs and applied to PSCs to achieve gradient doping of Ln3+. The introduction of Ln3+ formed a stronger Ln–I bond than Pb–I, which can adjust the WF, optimize the band alignment, and thus enhance the BEF (Fig. 11a–c). In addition, the CsPbBrCl2:Sm3+ modified PSCs have the optimal band alignment that can reduce interface carrier recombination. The PSC device based on the MAPbI3 modified with CsPbBrCl2:Sm3+ showed a maximum PCE of 22.52% and a high VOC of 1.20 V, representing 91% of the S–Q limit. In addition, the strategy of constructing gradient heterojunctions to enhance the BEF is of great significance for the preparation of lead-free environment-friendly PSCs. Wu et al. prepared a FASnI3 perovskite layer with a vertical Sn2+ gradient (Fig. 11d–f) to construct a heterostructure. In this work, the Fermi level moves up from the bottom to the top with the increase of Sn2+ content, which generates an additional electric field, and thus enhances the BEF and minimizes the bulk recombination loss inside the tin perovskite.80
Fig. 11 (a) Bandgap alignment for layers in a PSC device. (b) Schematic diagrams of energy band alignment for control and CsPbBrCl2:Sm3+ devices. (c) Schematic of the MAPbI3 perovskite crystal with defects and schematic of the proposed CsPbBrCl2:Sm3+ modified perovskite crystal structure.79 Copyright 2022, Wiley. (d) Schematic illustration of the gradient FASnI3-based inverted PSCs. (e) TOF–SIMS 2D mapping profiles of Sn along the x–z plane for the FASnI3 control and FASnI3 gradient films. (f) Depth profiles of the Fermi level in FASnI3 control and FASnI3 gradient films derived from the valence-band XPS analysis.80 Copyright 2022, Springer Nature. |
Meanwhile, the design of heterojunctions is not limited to the perovskite layer. Wu et al. hetero-epitaxially grew rutile type TiO2 single crystal array films on the fluorine-doped tin oxide (FTO) layer to construct an efficient electron selective layer with gradient Sn-doped interface by the hydrothermal method and subsequent thermal treatment.81 The Sn4+ diffuses from the FTO layer to the TiO2 film, resulting in gradient distribution of Sn4+ dopants in the film. The gradient doping induced a BEF perpendicular to the FTO substrate, which promotes the collection of photoelectrons in the TiO2 film and thus leads to the overall improvement of device performance and stability. Tan et al. obtained an interface gradient heterojunction BTA+–CsPbI3−xBrx by low-temperature treatment of quaternary bromide (BTABr) in inorganic CsPbI3 PSCs.82 The Br− in BTABr diffused into CsPbI3 thin films, which passivate coordination deficient Pb2+ and inhibit the formation of Pb clusters, effectively reducing the generation of I vacancies and suppressing non-radiative recombination. Meanwhile, the gradient Br− distribution and organic cation surface termination help to optimize the energy level arrangement between the CsPbI3 perovskite and HTL through I/Br ion exchange, thereby further promoting charge transfer in the device.
The introduction of additives can realize the transformation of semiconductor properties in the upper layer of the perovskite, forming a similar-double-layer structure, which is classified as the p–n junction. Xiong et al. directly observed the complete transformation from p-type to n-type surface energetics of perovskite MAPbI3 in the process of defect passivation by using capsaicin as an additive.28 This leads to the spontaneous formation of a p–n homojunction located at about 100 nm below the perovskite surface (Fig. 12a–d). The presence of the p–n homogeneous junction in devices enhances the built-in potential, resulting in a high VOC of the device. Finally, an efficiency of 21.88%, a filling factor (FF) of 83.81%, and excellent device stability were achieved. Furthermore, Liu et al. adopted a double-layer modification strategy to modify the perovskite layer and the ETL by introducing different functionalized natural vitamins.86 Vitamin D2 is introduced into the perovskite layer to transform its surface energetics from n-type to p-type, forming a spontaneous n–p homojunction (Fig. 12e and f), which increases the BEF and the hole extraction efficiency of PSCs. Meanwhile, vitamin C with high conductivity is introduced into the SnO2 ETL, which improves the electron mobility of SnO2 films and reduces interface energy level offset. Therefore, the interface charge transfer ability is improved.
Fig. 12 (a) UPS spectra of the secondary electron cutoff region and valence band region. (b) Calculated electrostatic potential and Fermi level of the perovskite surface with and without capsaicin. (c) Cross-sectional AFM topographies, corresponding KPFM images, and potential profiles under zero-voltage bias. (d) Energy level alignment schematics of the PSC comprising capsaicin.28 Copyright 2021, Elsevier. (e) Hall effect measurement of perovskite films. The inset shows the Hall effect measurement setup. (f) UPS spectra and energy level alignment schematics.86 Copyright 2023, Wiley. |
Direct cation doping can also form a p–n junction across the perovskite. Yu et al. constructed a p–n homojunction in the Sn–Pb perovskite layer by a gradient doping of Ba2+.87 The suitable Ba2+ doping can effectively compensate for p-type doping of the Sn–Pb perovskite and transform it into n-type without a change of bandgap. Induced by the unique film formation process, the distribution of Ba2+ cations is heterogeneous in perovskite films. Most of the Ba2+ remains in the top 600 nm region of the perovskite film, which turns this region into a weak n-type. In contrast, the bottom of the film remains as p-type (Fig. 13a–c). Therefore, the p–n homojunction is generated from top to bottom in the perovskite layer. The BEFs arising from this p–n homojunction facilitate the extraction of photogenerated carriers, which can increase the carrier extraction length and enhance the carrier recombination lifetime. In addition, the increased carrier extraction length offsets the negative impact of the increase in the thickness of the Sn-Pb perovskite layer. Such optimization makes Sn-Pb perovskites more suitable as a narrow band gap part of the perovskite tandem solar cells (Fig. 13d and e). Ultimately, the Sn–Pb single junction PSCs and all-perovskite tandem solar cells achieved a PCE of 21.8% and 25.3% (5.9 mm2), respectively. Moreover, doping directly at the interface can also form a junction. Li et al. prepared a heterojunction by the method of surface vulcanization to construct inverted PSCs (Fig. 13f).88 The Pb–S bond formed by surface vulcanization can shift the Fermi energy level of the perovskite interface upward. The surface energy band bending can align the energy levels of the perovskite/ETL and form a back electric field to accelerate the extraction of electrons (Fig. 13g). Finally, the resultant inverted device exhibited a PCE of 24.3% and a high VOC of 1.19 V (Fig. 13h), corresponding to a low voltage loss of 0.36 V.
Fig. 13 (a) Time-dependent Hall voltages for the Sn–Pb perovskite. (b) and (c) DLCP and TOF–SIMS measurements. (d) Schematic diagram of the structure. (e) Cross-sectional scanning electron microscope (SEM) image.87 Copyright 2022, Wiley. (f) SEM images, SEM–EDX mapping of the perovskite film and cross-sectional SEM–EDX mapping. (g) Energy level of perovskite films and back-surface field formation at the perovskite surface with SST. (h) J–V curves of control and SST-based PSCs.88 Copyright 2022, AAAS. |
Fig. 14 (a) Schematic illustration of the functions of embedded p–n heterointerfaces. (b) TEM image of CdTe NCs (inset: size distribution diagram and Mie-scattering image). (c) HRTEM and corresponding fast Fourier transform of CdTe NCs. (d) Electron mobilities of different ETLs using the SCLC model.90 Copyright 2023, Wiley. (e) Cross-sectional SEM image of the cell and the schematic energy level alignment of the cell. (f) The distribution of surface potential measured by KPFM.91 Copyright 2019, AAAS. |
As is well known, the difference in the WF between two contacts strongly influences the formation of a BEF in PSCs, in which the WF arises from the difference between the vacuum level and the Fermi level.15 Thus, the regulation of the energy level for selective contact layers can be used to increase the BEF for maximizing charge extraction and VOC. The excellent energy level orientation between functional layers would effectively avoid the loss of the BEF. At the same time, on the premise of energy level alignment, expanding the WF difference between the two poles/transport layers can enhance the BEF.102 Tuning the energy level position of selective contact layers, including the ETL/HTL and perovskite layer, can optimize the WF difference in PSCs. The Fermi level of the selective contact materials can be tuned via vacancy defect doping, substitution doping, and molecule modification. In this chapter, we divide the strategy of energy level regulation into the construction of a dipole layer and functional layer modification. The former is a kind of interface engineering, which can efficiently and directly adjust the WF of the perovskite surface to achieve the enhancement of the BEF. The latter changes the WF by developing and modifying the HTL and the ETL.
Lee et al. applied p-doped poly(9,9-bis(4′-sulfonatobutyl)fluorene-alt-co-1,4-(2,5-dimethoxy)phenylene) (p-PFP-O) as a HTL for the anode and poly[(9,9-bis(3′-(N,N-dimethylamino)propyl)-2,7-fluorene)-alt-2,7(9,9-dioctylfluorene)] (PFN) as an ETL for the cathode (Fig. 15a and b).29 The introduction of this strong electric dipole expands the WF difference of the interface between the electrode and the transport layer, thus enhancing the BEF in PSCs. Wang et al. introduced electric dipole molecules PTFCN and CF3BACl to enhance the WF difference Δμ1 between the two electrodes and Δμ2 between the upper and lower surfaces of the perovskite (Fig. 15c).107 Ultimately, the maximization of the BEF can be achieved within the controllable range. Yu et al. developed Ni phthalocyanine (NiPc) modified by four methoxy group ethoxy units as hole transporting materials (HTMs).108 The molecular structure of NiPc effectively increases the intramolecular dipole, which builds an intramolecular electric field inside NiPc and accelerates charge transport. Ansari et al. applied quasi-3D aziridine lead iodide (AzPbI3) as the surface intermediate layer of the primary 3D perovskite film. The existence of AzPbI3 dipoles on the surface of the perovskite changes the WF, which provides an enhanced electric field and thus improves the photovoltaic parameters significantly.109
Fig. 15 (a) Chemical structures of EDL materials and schematic device architecture, and (b) corresponding energy level diagrams.29 Copyright 2018, RSC. (c) Structure of a PSC device.15 Copyright 2020, Wiley. (d) WF shift measured with a Kelvin probe (KP), Kelvin probe force microscopy (KPFM), ambient pressure photoelectron spectroscopy (APS) and cyclic voltammetry (CV); the order is upper left, upper right, lower left and lower right. (e) Comparison of the WF shift and VB shift related to the perovskite obtained with different techniques for the cases of a positive and negative dipole.110 Copyright 2021, RSC. |
Furthermore, Canil et al. induced stable dipoles on the surface of a perovskite using self-assembled monolayers (SAM) to regulate the WF of the perovskite.110 The “positive dipole” and “negative dipole” are defined in this work. The former is the case where the total dipole points to the perovskite surface, and the latter is the opposite. The perfluorodecyl iodide as a positive dipole will cause an upshift of the vacuum energy level, resulting in an increase in the WF of the functionalized material, and the effect of amyl sulfide as a negative dipole is the opposite. Meanwhile, a strategy that combines multiple characterization methods was proposed to ensure the measurement accuracy of the WF (Fig. 15d and e), which is beneficial for directly selecting the molecular type and deposition conditions to adjust the WF of the perovskite and then designing a controllable energy level structure to regulate the BEF. In addition, Liu et al. used organic–inorganic (OI) complexes (CL-CH3 and CL-CF3) as multifunctional interlayers to prepare high-performance PSCs by a two-step continuous method.111 The introduction of the intermediate layer not only passivates the defects of the perovskite film but also forms a vertical dipole moment, which enhances the BEF and facilitates the extraction of charge carriers.
For example, Kamarudin et al. replaced formamidinium iodide (FAI) with dimethylammonium iodide (DMAI) partly, which is a strategy of A-site cation substitution to improve the performance of lead-free (Sn-based) PSCs.112 The oxidation of Sn2+ is inhibited due to DMAI doping, resulting in a decrease of defect density and hole (p-type) carrier density. Finally, the beneficial effect of DMAI is demonstrated by BEF enhancement and VOC improvement. Moreover, Zhang et al. doped LiI into MAPbI3 to modulate the interface energy band alignment and facilitate the carrier transport.113 The non-intrinsic Li+ and I− were introduced as ion dopants in MAPbI3, which significantly improved the ionic conductivity in the perovskite (Fig. 16a). Meanwhile since both Li+ and I− can move within the perovskite structure, there is excessive I− aggregation at the Ag anode, and the Li+ is considered to concentrate on the FTO side due to its high mobility group (Fig. 16b). The aggregation of Li+/I− regulated the energy levels of the perovskite and induced n/p doping in MAPbI3. Li+ shifts the Fermi energy level upward to form n-type doping and I− causes the Fermi energy level to move down, thereby achieving p doping (Fig. 16c). Finally, the doping of LiI optimizes the energy level orientation of tin-based PSCs and enhances the BEF.
Fig. 16 (a) EDX linear scan of the element distribution. (b) σ–T plots of Li-0, 2, and 5% in Arrhenius coordinates; the inset is the device structure. (c) The surface potential profiles (line scan) at 0 V bias, 0.5 V bias, and −0.5 V bias.113 Copyright 2017, Wiley. (d) Schematic illustration of the energy level diagram of the presented PSCs. (e) Hysteresis curves.114 Copyright 2019, ACS. |
In addition to perovskite layers, the electrode and transport layers can be modified by additive doping to enhance energy level matching. Furthermore, other influencing factors, such as hole/electron mobility and electrical conductivity, are usually considered to achieve the best performance. Zhou et al. regulated the WF of the TiO2 ETL and ITO electrodes to improve interface contact by modifying them with Y-doping and polyethylenimine ethoxylate (PEIE), respectively.115 After regulating, the ITO/PEIE electrode and the Y-TiO2 ETL achieve energy level alignment, which is conducive to efficient transport of electrons between the ETL and ITO layers. Xiong et al. introduced Mg into the SnO2 ETL by high-temperature treatment to change its optical properties and adjust its energy level.116 Appropriate Mg doping improves the interface contact, reduces the free electron density, and increases the electron mobility, which increases the electron driving force and thus facilitates the electron transfer at the ETL/perovskite interface. Wang et al. doped Ru into the TiO2 ETL to improve the performance and conductivity of PSCs.117 Ru doping significantly reduced the electron trap density of the ETL, which avoids excessive charge accumulation and ensures effective electron injection at the ETL/perovskite interface. Akin used Ru-doped SnO2 as the ETL to improve the photoelectric properties of PSCs.114 The incorporation of Ru makes the Fermi level of the SnO2 ETL move up (Fig. 16d), which improves the energy level alignment of the ETL/perovskite interface and thus enhances the BEF of the device and improves VOC and electron extraction. In addition, the device prepared by this work has almost no hysteresis (Fig. 16e).
Furthermore, Liu et al. doped p-type dopant 2,3,5,6-tetrafluoro-7,7,8,8-tetracyanoquinodimethane (F4-TCNQ) into the HTL to improve the photovoltaic performance of inverted PSCs.118 The low concentration of F4-TCNQ doping can effectively adjust the electrical properties of PEDOT:PSS films and reduce the highest occupied molecular orbital (HOMO) energy level. Therefore, the enhanced conductivity and optimized energy level arrangement of doped thin films allow for effective carrier transport from the perovskite absorption layer to the HTL. Seo et al. used Zn-TFSI2 to replace Li-TFSI2 as a p-type dopant for HTL material spiro-MeOTAD in PSCs.119 Therefore, the hole mobility was increased by an order of magnitude and the BEF was significantly enhanced. Jiang et al. developed a Spiro-OMeTAD layer modified with MoS2, which plays an important role in improving the hole mobility of Spiro-OMeTAD and enhancing the stability of the film.120
Fig. 17 (a) Diagram of the non-centrosymmetric crystal classes and their ferroic properties.121 Copyright 2019, Wiley. (b) Schematic illustration of the hysteresis loop for electric displacement versus electric field.126 Copyright 2023, ACS. |
In order to apply the PEF to halide perovskite photovoltaics, the ferroelectricity of halide perovskites has been studied recently, including the transformation of the ferroelectric crystal phase, the existence of ferroelectric domains in the perovskite and the mechanism of the switchable photovoltaic effect. Walsh et al. established a computational model to study the lattice polarization affected by polar organic cations and verified the existence of spontaneous polarization in the MAPbI3 perovskite.131 The ferroelectric domains in the perovskite will lead to internal junctions, which is beneficial for the separation of e–h pairs and reduction of recombination. By using piezoelectric response force microscopy (PFM), Röhm et al. demonstrated that MAPbI3 possesses alternating polarization domains with a width of 90 nm, which were identified as polarized ferroelectric domains (Fig. 18a).132 This work directly observed the presence of ferroelectric domains in perovskites. Rossi et al. modeled different polarization configurations based on the alternating ferroelectric domain patterns measured by PFM (Fig. 18b). The polarization model was combined with drift–diffusion simulations to study the influence of ferroelectric domains on charge separation and recombination losses in PSCs.122 This work proved that the presence of ordered ferroelectric domains improves the VOC by reducing the Schockley–Read–Hall (SRH) recombination loss and forms a high conductive channel to improve the carrier extraction, thereby improving the PCE.
Fig. 18 (a) High-resolution PFM.132 Copyright 2017, RSC. (b) Piezoresponse force micrograph recorded on a typical MAPbI3(Cl) grain surface and three polarization patterns.122 Copyright 2018, Elsevier. (c) Layout of the sample architecture used for poling experiments. (d) Top-view optical microscope image of the MAPbI3 layer atop interdigitated gold electrodes. (e) LPFM images before polarization, after polarization, and after subsequent reverse polarization.133 Copyright 2020, Wiley. |
Furthermore, Röhm et al. demonstrated that MAPbI3 films exhibited spontaneous polarization during the transition from the cubic phase (above Tc) to the tetragonal phase (below Tc).121 The polarization is mainly in-plane orientation as probed by PFM. Drift–diffusion simulation on ferroelectric domains with alternating polarization indicated that the presence of a ferroelectric built-in electric field is beneficial for reducing the SRH recombination of charge carriers in perovskite grains. Subsequently, they proved that applying an EEF can achieve ferroelectric poling of MAPbI3.133 The microscopic modulation of the polar domains under the EEF and the concurrent persistent shift of the currents indicate that MAPbI3 is a ferroelectric semiconductor. However, since MAPbI3 has higher conductivity compared to conventional insulating ceramic ferroelectrics, applying an electric field can drive remarkable current through MAPbI3, which will damage the PSC device. To avoid this situation, this work designed a sample architecture (as shown in Fig. 18c and d), allowing in-plane poling of the MAPbI3 layer. Based on this method, a mild electric field can be applied to avoid sample damage, and the change of the polarization domain in the electric field can be directly observed by lateral PFM (LPFM) (Fig. 18e). Controlling the ferroelectric polarization by the EEF is beneficial to study the ferroelectric properties of perovskites, and then regulate the intensity of the PEF.
The influence of ferroelectric polarization on the performance of PSCs was further explored. Leonhard et al. presented experimental evidence to illustrate the correlation between the ferroelectric polarization of MAPbI3 thin-films and the performance of PSCs.134 The evolution and polarization orientation of ferroelectric domains during the fabrication of MAPbI3 thin-films were investigated through PFM and KPFM. Within several seconds to minutes of the thermal annealing process, MAPbI3 thin films undergo a transition from the coexistence of out-of-plane and in-plane polarization to dominant in-plane polarization. Solar cells with alternating lateral (in-plane) domains have the best FF and efficiency, while solar cells with vertical (out-of-plane) domains only exhibit moderate performance. Moreover, solar cells with out-of-plane polarization exhibit lower stability during the process of electrical characterization. Through this work, it can be seen that the evolution of ferroelectric domains in perovskite thin films is an important factor to affect the final device performance.
The ferroelectricity of 3D perovskites is still controversial in current reports. Edmands et al. measured the conductance hysteresis of CH3NH3PbI3 perovskite thin films by a double-wave method, which excluded the hysteresis effect caused by ferroelectricity in PSCs and emphasized the importance of ion migration.135 Coll et al. demonstrated by piezoelectric force microscopy (PFM) and macroscopic polarization methods that MAPbI3 does not exhibit permanent polarization at room temperature.136 This work suggests that the transient polarization at room temperature may be related to the photogenerated dipole induced by intrinsic charge transfer. Almora et al. showed that the enhancement of capacitance observed in a capacitance–voltage experiment for MAPbI3 at a specific frequency is related to the overlap effect of the interface cumulative capacitance, and the light-induced changes in the dielectric properties of the perovskite can be ignored.137 Tong et al. demonstrated that MAPbI3 may exhibit fluctuating polar nanoregions (especially below 190 K), and it is difficult to obtain a stable ferroelectric phase.138 These reports show that the abnormal photovoltaic behavior of 3D perovskites do not necessarily originate from their own ferroelectric properties. In addition, even if 3D perovskites exhibit ferroelectricity, there are often harsh constraints, such as specific temperatures. Therefore, the ferroelectricity of 3D perovskites cannot be well applied in photovoltaic devices.
Although the ferroelectricity of 3D perovskites has limitations, the positive role of the ferroelectricity PEF in PSCs is still worth looking forward to. The 3D structure of perovskite ABX3 has strict requirements for the size of each component. Exploring perovskites towards low dimensionality (including 2D, 1D, and 0D) can broaden the tunability of perovskites.139–142 In fact, the ferroelectric design of low-dimensional perovskites has fewer constraints on the A-site organic cation, so it is easy to obtain low-dimensional ferroelectric perovskites. For example, introducing large hydrophobic organic cations to construct 2D perovskites can induce ferroelectricity of perovskites.30 However, low-dimensional perovskites usually exhibit a large band gap and are difficult to directly apply as light absorption materials in PSCs. It is an effective strategy to combine low-dimensional perovskites with traditional 3D perovskites as a light absorber.
The 2D ferroelectric perovskites are usually used to construct a 2D intermediate layer structure in PSCs. The ferroelectricity of 2D perovskites can be controlled by modulating the dipole moment. Inducing the ferroelectricity of 2D perovskites based on 2D/3D heterostructures can provide a PEF and thus effectively enhance the BEF in PSCs.143–145 Yang et al. obtained 2D PbI2 perovskite ferroelectrics for the first time by introducing chiral cations (Fig. 19a).146 As shown in Fig. 19b and c, the 180° reversal of the PFM phase signal and the typical butterfly ring of the amplitude signal demonstrate the switching ability of ferroelectric domain polarization.
Fig. 19 (a) Temperature dependence of the SHG intensity. (b) and (c) Vertical PFM phase and amplitude signals as functions of the tip voltage for a selected point at room temperature, showing local PFM hysteresis loops.146 Copyright 2019, Wiley. (d) I–E curve and P–E curve of 2D, 3D and 2D/3D perovskite films. (e) Mott–Schottky plots of devices.30 Copyright 2023, Wiley. (f) HRTEM image of the 0.6% molar ratio 1D/3D mixed perovskite. Scale bar: 2 nm. (g) PFM phase, amplitude images, topographic image of the thin film and dependence of phase and amplitude signals with applied DC bias for a selected point.148 Copyright 2021, Wiley. |
Subsequently, Chen et al. reported a 2D ferroelectric perovskite (4-aminotetrahydropyran)2PbBr4[(ATHP)2PbBr4] with a huge piezoelectric voltage coefficient (g33).147 This molecular ferroelectric has a higher Tc and its g33 is twice that of polyvinylidene fluoride (PVDF). Han et al. designed a ferroelectric 2D perovskite based on a pyridine heterocyclic ring as the organic intermediate layer.30 The P–E loop of the sample exhibits the ferroelectric polarization inversion (Fig. 19d), indicating the obtained 2D perovskite has ferroelectricity. The incorporation of ferroelectric 2D materials into 3D perovskites leads to an increase in the BEF (Fig. 19e), thereby improving the exciton dissociation efficiency in a device. Furthermore, the applications of 1D ferroelectric perovskites have been reported. Zhang et al. combined 1D ferroelectric perovskite trimethyliodomethylammonium tris(μ2-iodo)-lead with 3D perovskite (FAPbI3)0.85(MAPbBr3)0.15 to prepare 1D/3D perovskite films with ferroelectric properties (Fig. 19f and g).148 The enhancement of the BEF caused by ferroelectric polarization significantly improves the performance of the PSC.
Many strategies have been proposed to generate a PEF through ferroelectric doping. While introducing the ferroelectric materials, the synergistic effect of the interface dipole layer and molecular passivation should also be considered. Zhang et al. added polyvinylidene fluoride-trifluoroethylene polymer P(VDF-TrFE) as a ferroelectric dopant in the PSCs, thus broadening the depletion region and reducing the overlap of electron and hole wave functions.151 The thin layer of the polarized ferroelectric polymer film was inserted between the perovskite layer and the HTL to act as an interface dipole layer. The existence of an interface dipole layer leads to obvious band bending, which aligns the energy levels of adjacent functional layers and enhances the BEF to provide strong charge transfer force for carrier separation. Meanwhile, the F atom in the ferroelectric molecule can combine with the H atom to form an F–H bond, thereby fixing the perovskite phase structure. Through synergistic effects, this work obtained a PCE of 21.38% and a VOC of 1.14 V, with almost no hysteresis.
Furthermore, Chen et al. developed a high polarizability organic ferroelectric polymer, poly(vinylidene fluoride):dabcoHReO4 (PVDF:DH), to enhance the BEF of the FA0.92MA0.08PbI3 perovskite.152 The combination of PVDF and DH caused β phase transition of PVDF. β-PVDF with an all-trans diode arrangement has better ferroelectric properties and the F atoms on the surface can provide excellent hydrophobicity. PVDF:DH is polarized by the BEF to generate an additional PEF, which is permanently maintained in the direction consistent with the BEF direction of PSCs. The superposition of the electric field can more fully drive the transport and extraction of charge carriers, thereby obtaining excellent PCE.
For example, Li et al. investigated the mechanism that mainly leads to hysteresis in planar FTO/compact TiO2/CH3NH3PbI3−xClx/Spiro-OMeTAD/Ag perovskite solar cells, and attributed the hysteresis effect of PSCs to the interface barrier formed by the migration of iodide ions/interstitials.160 Belisle et al. established a “wiggly band” model of mobile ion field screening and hysteresis to demonstrate that the BEF screening caused by ion migration is an important factor leading to the hysteresis effect.161 Kim et al. studied the built-in equilibrium charge region at the interface that considered the equilibrium distribution of both ionic and electronic carriers.162 The schematic of concentration and conductivity profiles indicated that there is obvious ion adsorption and accumulation at the interface contact (Fig. 20a), which leads to the redistribution of charge. Liu et al. established a mechanism that uses the BEF as a key factor to explain ion migration and abnormal behavior of the device (Fig. 20b).163
Fig. 20 (a) Schematic concentration and conductivity profiles at the interface.162 Copyright 2022, Wiley. (b) The anomalous device behaviors caused by electrochemical ion migration.163 Copyright 2021, ACS. |
Based on the above research, it can be concluded that unfavorable ion migration leads to the BEF screening, which will affect the efficiency of carrier separation. To reduce the shielding of the BEF caused by ion migration, many possible strategies have been proposed in the past. Designing new functional layer materials to improve the contact at the interface and increase ion migration activation energy is an efficient strategy.164 For example, Xu et al. reported the perovskite-phenyl-C61-butyric acid methyl ester (PCBM) hybrid solid, based on which PSCs exhibited low hysteresis and recombination loss.165 Meanwhile, this work links the migration of iodide ions with the BEF and investigates the possible mechanism. Iodide anions migrate under the action of an electric field to generate ion currents. Therefore, the existence of the BEF may cause anionic charge to move in the opposite direction of the field, resulting in the decrease of photocurrent over time. PCBM can combine with iodide-rich surface sites or unincorporated iodide anions, to reduce the migration of anions through defects at grain boundaries, thus compensating the BEF.166,167 Subsequently, Zhong et al. demonstrated that PCBM could inhibit the hysteresis effect caused by ion migration and weaken the shielding effect of ion accumulation on the BEF.168 In addition, the ion compensation strategy is an effective strategy to improve interface contact. Chen et al. synthesized CuScO2 as a mesoporous HTL in inverted PSCs and pretreated the CuScO2 HTL with formamidine salt, which compensated for the iodine vacancy of the buried perovskite film.169 This ion compensation strategy can repair the ion loss in situ and reduce the charge injection barrier, thus enhancing the BEF.
The directional arrangement of dipoles in the functional layer can also improve the charge accumulation at the interface and compensate for the BEF. Ahmadi et al. studied the HTL with significant photogenerated dipoles, which can be orientationally aligned under the action of the BEF (Fig. 21a).170 The photogenerated dipole in the transport layer can trigger the charge transfer through the interface dipole field, thereby reducing the charge accumulation at the interface and BEF screening. Yang et al. inserted an effective interface dipole trimethylamine oxide (TMAO) between the perovskite layer and the ETL, which reduced the interfacial charge density and enhanced the BEF (Fig. 21b).171 In addition, the decrease of low-frequency capacitance indicates that this change of the BEF is mainly affected by the inhibition of interface charge accumulation, rather than the polarization effect of the interface dipole (Fig. 21c).172
Fig. 21 (a) Electric field-induced photoluminescence spectroscopy.170 Copyright 2017, Wiley. (b) Mott–Schottky analysis. (c) C–F plots. (d) Capacitance transients from anion migration (drift).171 Copyright 2019, ACS. (e) Arrhenius fitting results of anion migration. (f) The function of the thermal-electrical poling voltage.173 Copyright 2022, RSC. |
In addition to directly inhibiting ion accumulation, constructing favorable ion accumulation can induce an electric field in line with the BEF, which will improve the charge extraction efficiency and improve the device performance significantly.173,174 Such favorable ion accumulation can be achieved by using the sensitivity of ion accumulation to light, heat, and bias. For example, Lian et al. prepared high-performance quasi-2D PCSs by using light-induced favorable ion accumulation.173 Compared with the 3D perovskite, the ions in the quasi-2D perovskite have lower mobility at room temperature and higher anion activation energy (Fig. 21d and e), which excludes the influence of some unfavorable ion migration in PCSs. The light-induced quasi-Fermi level splitting compensates for the BEF. Meanwhile, the uneven ion potential caused by uneven light intensity distribution can be used to enhance favorable ion accumulation, which generates an electric field that has the same polarity as the BEF of PSCs. Furthermore, the EEF polarization can amplify the effect of beneficial ion accumulation and improve the performance of PSCs (Fig. 21f). Additionally, Xiong et al. fabricated PSCs with vertical and lateral structures and switching the direction of photocurrent by controlling the application of a small electric field (<1 V μm−1).175 This field photovoltaic effect can be explained by the formation of a reversible p–i–n structure caused by ion drift in the perovskite layer. The optimal performance parameters of the polarized single-layer vertical structure device are close to that of the optimized multi-layer device with both ETL and HTL. The introduction of the EEF can effectively guide the ion migration direction to obtain the ion accumulation polarity consistent with the BEF direction, which can enhance the BEF and improve the performance of PSCs.
(1) Constructing junctions in PSCs. The construction of junctions in PSCs, such as perovskite–perovskite junctions, gradient heterojunctions, and extra junctions, can enhance the BEF of PSCs by creating a new space charge region, leading to efficient carrier extraction. However, there are still controversies regarding the definition and characterization of these junctions in PSCs, and further research is needed to establish a conceptual model that accurately represents the structure and working principles of PSCs. Methods to directly characterize the junctions and measure the electric field intensity generated by them need to be developed. In addition, constructing heterojunctions in perovskites is an effective means to reduce carrier interface recombination through field effect passivation. Tandem solar cell devices usually suffer from serious energy loss caused by multi-interface recombination. Therefore, constructing heterojunctions in tandem solar cells is a promising research direction to reduce the energy loss and break through the S–Q limit efficiency.
(2) Energy level regulation. Regulating the energy levels in PSCs has been another approach to enhance the BEF. Strategies such as constructing dipole layers and modifying functional layers have been effective in expanding the WF difference between layers. However, accurately determining the WF shift is a challenge, as current methods like UPS require a vacuum environment, which may affect the measurement accuracy. Developing characterization techniques under atmospheric pressure, such as KPFM, APS, and CV measurements, could improve the accuracy and repeatability of energy level structure characterization. The conductivity of functional layers should also be considered when designing energy level regulation to prevent reduced carrier mobility or insulating intermediate layers.
(3) Polarized electric field in PSCs. The introduction of the ferroelectric PEF into mixed halide PSCs shows promise in improving the performance of PSCs. However, there are challenges in applying 3D ferroelectric perovskite materials, as they often exhibit a paraelectric phase and do not have the condition of spontaneous polarization at room temperature. Increasing the Tc of 3D perovskites to achieve a ferroelectric phase at room temperature by designing a new type of 3D perovskite can be a future research direction. Furthermore, optimizing the PSC device model and designing ferroelectric materials with excellent and stable characteristics can fully exploit the potential benefits of the PEF, leading to higher photovoltage and overall performance in PSCs.
(4) Mitigating surface charge accumulation caused by ion migration. Ion migration in PSCs leads to charge accumulation at the interface, which in turn causes screening of the BEF. Although several strategies have been proposed to mitigate charge accumulation and reduce BEF screening, the negative effects of ion migration in PSCs cannot be ignored. One effective strategy is doping Sn metal in Pb-based PSCs, which greatly reduces the influence of ion migration and makes it even difficult to observe in Sn-based PSCs. Therefore, B-site substituted PSCs, such as Sn-doped PSCs, can effectively reduce the influence of charge accumulation at the interface due to ion migration, resulting in PSC devices with low BEF screening.
This journal is © the Partner Organisations 2024 |