Home

Search

Collections

Journals

About

Contact us

My IOPscience

Recent progress and challenges of organometal halide perovskite solar cells

This content has been downloaded from IOPscience. Please scroll down to see the full text. 2016 Rep. Prog. Phys. 79 026501 (http://iopscience.iop.org/0034-4885/79/2/026501) View the table of contents for this issue, or go to the journal homepage for more

Download details: IP Address: 149.171.67.164 This content was downloaded on 30/01/2016 at 06:03

Please note that terms and conditions apply.

Reports on Progress in Physics Rep. Prog. Phys. 79 (2016) 026501 (26pp)

doi:10.1088/0034-4885/79/2/026501

Review

Recent progress and challenges of organometal halide perovskite solar cells Liyan Yang1, Alexander T Barrows2, David G Lidzey2 and Tao Wang1 1

  School of Materials Science and Engineering, Wuhan University of Technology, Wuhan, 430070, People’s Republic of China 2   Department of Physics and Astronomy, University of Sheffield, Sheffield, S3 7RH, UK E-mail: [email protected] and [email protected] Invited by Mark Geoghegan Received 16 July 2015, revised 11 November 2015 Accepted for publication 23 November 2015 Published 28 January 2016 Abstract

We review recent progress in the development of organometal halide perovskite solar cells. We discuss different compounds used to construct perovskite photoactive layers, as well as the optoelectronic properties of this system. The factors that affect the morphology of the perovskite active layer are explored, e.g. material composition, film deposition methods, casting solvent and various post-treatments. Different strategies are reviewed that have recently emerged to prepare high performing perovskite films, creating polycrystalline films having either large or small grain size. Devices that are constructed using mesosuperstructured and planar architectures are summarized and the impact of the fabrication process on operational efficiency is discussed. Finally, important research challenges (hysteresis, thermal and moisture instability, mechanical flexibility, as well as the development of lead-free materials) in the development of perovskite solar cells are outlined and their potential solutions are discussed. Keywords: perovskite, solar cells, photovoltaics, organometal halide, energy harvesting (Some figures may appear in colour only in the online journal)

1. Introduction

solar cell technology, with the state-of-the-art power conversion efficiency (PCE) currently being over 20% as certificated by NREL [5]. Early research in perovskite solar cells is closely associated with work on dye-sensitized solar cells (DSSCs), with perovskite materials absorbed on mesoporous TiO2 (mp-TiO2) employed as a dye-sensitizer in DSSCs. Liquid DSSCs using CH3NH3PbI3 dyes gave an initial efficiency of 3.8% [4]. An evolutionary jump then followed, as perovskite absorbers were used as the primary photoactive layer to fabricate solid-state meso-superstructured perovskite solar cells [6]. This happened in 2012 when Kim, Grätzel and Park et al [6] used spiro-MeOTAD and mp-TiO2 as the hole transport and electron transport materials respectively (HTM/ETM) and obtained a PCE of 9.7% for the first reported perovskite

The world’s increasing consumption of energy, together with the decreasing supply of fossil fuels have driven the development of energy from renewable sources. Photovoltaic solar cells directly convert the energy of sunlight into electricity based on the photovoltaic effect and for this reason have been regarded as a very promising energy generation source. A number of different types of solar cell technologies have been explored, including silicon solar cells, III-V solar cells, quantum dots solar cells, dye-sensitized solar cells, organic solar cells and most recently perovskite solar cells [1, 2]. Since initial reports of the use of CH3NH3PbBr3 [3] and CH3NH3PbI3 [4] as a sensitizer in solar cells, great interest has been shown in the development of this perovskite-based 0034-4885/16/026501+26$33.00

1

© 2016 IOP Publishing Ltd  Printed in the UK

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 1.  Cross-section SEM images of (a) a meso-superstructured perovskite solar cell (scale bar is 500 nm). Figure reprinted from

[17]. Copyright 2013 American Chemical Society. (b) A normal planar perovskite solar cells with the presence of an HTL and an ETL. Figure reprinted from [59]. Copyright 2014 the Royal Society of Chemistry. (c) An inverted planar perovskite solar cell. Figure reprinted with permission from [28]. Copyright 2014 American Chemical Society.

layers in perovskite photovoltaic devices most often leads to higher power conversion efficiency due to enhanced charge transport away from the photoactive layer and thus reduced charge recombination. In a typical device, the substrate mat­ erial is comprised of rigid glass or a flexible polymer (e.g. polyethylene terephthalate PET [28–30], polyethylene naphthalate PEN [31]), which has been coated with a semi-transparent conductive film of ITO, FTO or IZO. Typical materials used for the opaque rear electrode in devices include Au, Ag, Al and carbon based materials [32–35]. In this review, we start with an introduction to the composition and optoelectronic properties of perovskite solar cells. The morphology of the photoactive layer is described, and the role of material composition, solution deposition techniques and subsequent post-treatments on device efficiency are also reviewed. The effects of different device structures—for example interface engineering or using mesoporous or planar layers—will also be discussed. The last part of this review will explore current challenges in the development of perovskite solar cells, and will address topics such as hysteresis, devicestability, mechanical flexibility and the realization of lead-free material systems.

based solid-state mesoscopic heterojunction solar cell. A further significant step forward was the use of a layer of insulating Al2O3 as meso-structured scaffold [7–9] instead of the semiconductor TiO2 [10]. The fact that this type of solar cell (incorporating either a semiconducting or an insulating scaffold layer) could deliver very high device efficiency indicates excellent ambipolar charge transport efficiency within the perovskite photoactive layer. A characteristic of meso-superstructured solar cells is that the perovskite layer is deposited on a mesoporous layer, e.g. TiO2 or Al2O3. A subsequent report by Snaith et al showed that planar heterojunction perovskite solar cells incorporating vapour-deposited perovskite layer could be used to generate devices with a PCE over 15% without the need for any mesoporous layers [11]. Since then, planar solar cells using a solution-processed photoactive layer have reached PCEs close to 20% using compositional [12] or interfacial [13] engineering. More information on the evol­ ution and development of perovskite solar cells can be found in other review papers [2, 14]. A perovskite solar cell (PSC) is a structure that usually comprises a substrate, electrodes, a perovskite photoactive layer, together with necessary charge transport layers (i.e. a hole transport layer (HTL) [15] and an electron transport layer (ETL) [16]). PSCs fall in two main types: meso-superstructured perovskite solar cells (MPSCs) [17] (see figure 1(a)) that incorporate a mesoporous layer, and planar perovskite solar cell (PPSCs) (see figures 1(b) and (c)) in which all layers are planar. Upon absorption of light, the low binding energy of the photo-excited states in the perovskite result in the formation of free charge-carriers that can be effectively transported to the device electrodes for extraction [18, 19]. Recent reports have evidenced excitons or electron-hole pairs in a perovskite [20–23], however it is not clear whether such excitons are generated directly upon the absorption of light or following recombination from electron-hole pairs during charge recombination [24]. By studying exciton dissociation within perovskite and conventional excitonic semiconductors, Huang et al demonstrated that MAPbI3 and MAPbBr3 are nonexcitonic materials [25]. This indicates that perovskites should be treated as traditional inorganic thin film solar cells. As perovskites can have ambipolar charge transport properties (they can conduct both electrons and holes) [24], HTL and ETL layers are not necessary to fabricate a working perovskite solar cell [26, 27]. However, the use of HTL and ETL

2.  Composition of the perovskite photoactive layer A ‘perovskite’ usually refers to any material that has the same crystal structure of calcium titanate [2] (ABY3 [11]), as shown in figure 2. Most often, the perovskites used in PSCs photoactive layers are composed of an organohalide-metal hybrid. Here, typical organic cations (A) include CH3NH+ 3 (methylammonium ‘MA’) and HC(NH2)+ (formamidinium ‘FA’), 2 with a range of metal cations (B) being utilized, including the divalent metal ions Pb2+, Sn2+, Eu2+, Cu2+, Ge2+ etc [18, 36, 37]. These cations have been combined with halide anions (Y) including Cl−, Br− and I−. By combining such components in various ratios, a range of perovskite materials can be realized, including MAPbI(3−x)Clx [38–41], MAPbI(3−x)Brx [42, 43], MAPbBr(3−x)Clx [44], FAPbI(3−x)Clx [45], (MA)x(FA)1−xPbI3 [46, 47], Csx(MA)1−xPbI3 [48], MASnxPb(1−x)I3 [49] and (FAPbI3)1−x(MAPbBr3)x [12]. This powerful approach opens up a large parameter-space for materials development, and critically allows the bandgap of the perovskite photoactive layer to be tuned between 1.30 to 2.30 eV [14]. In the 2

Review

Rep. Prog. Phys. 79 (2016) 026501

detectable in the final perovskite film [54]. It was proposed that an intermediate MAPbCl3 phase formed during solution casting, but later disappeared upon thermal annealing (see diagram in figure 3). The presence of chlorine in MAPbI3−xClx was evidenced by Williams et al through the detection of HCl gas during thermal decomposition of MAPbI3−xClx [55]. Using x-ray fluorescence measurements, Unger et al [56] found that chlorine exists both within the crystal lattice and at the grain boundaries or the interface with TiO2 layer upon thermal annealing. Chlorine was thus proposed to act as a dopant and surface passivator in the perovskite MAPbI3−xClx. Based on angleresolved x-ray photoelectron spectroscopy and first-principles calculation, Colella et al [57] concluded that chlorine was located close to the perovskite/TiO2 interface after the perovskite layer had been thermally annealed at 100 °C and speculated that such vertical stratification can induce band bending at the interface and enhance electron diffusion and collection (the arrow in figure  4). Recently, Li et al [58] studied MAPbI3−xClx (formed by first spin-coated the mixed solution of PbI2:PbCl2 at 1 : 1 followed with simultaneous annealing and exposure to MAI vapor at 120 °C) and concluded that doped-chloride existed in the final perovskite film but it only affected perovskite structure rather than Fermi level or the photo-carrier lifetime. It was shown that MAPbI3−xClx and MAPbI3 have similar absorption and photoluminescence (PL) decay-lifetimes, bandgaps and Fermi levels, however devices with a MAPbI3−xClx photoactive layer generally result in higher efficiency PSCs.

Figure 2.  Crystal structure of a perovskite. Figure reprinted with permission from [11]. Copyright 2013 Nature Publishing Group.

following sections, the three main perovskite compounds that have received most intensive studies (MAPbI3, MAPbI3−xClx and FAPbI3) will be discussed in more detail. 2.1. MAPbI3

The most widely studied perovskite material system is methylammonium lead triiodide (MAPbI3), with the state-of-theart PCE of a MAPbI3 perovskite PSC being as high as 18.1% [50]. The MAPbI3 perovskite was initially incorporated in the liquid electrolyte of DSSCs, however, it was found to be very unstable and gradually dissolved in the liquid electrolyte, with the device degrading after 10 min continuous illumination. In 2011, MAPbI3 quantum dot (with a diameter see 2 ~ 3 nm) sensitized solar cells reached the highest efficiency (6.5%) reported inorganic quantum dot sensitized solar cells at that time [51]. In 2012, a perovskite based solid-state device (FTO/compact-TiO2/MAPbI3  +  mp-TiO2  +  HTM/HTM/Au) was then demonstrated that showed both improved efficiency and stability, and gave impetus for the development of new types of PSC devices [6]. In 2013, a sequential deposition process using MAPbI3 precursors was demonstrated for the first time, resulting in films having fewer morphological defects and enhanced device efficiency of 15.0% [52]. This paved the way for the development of sequential deposition techniques using other perovskite materials.

2.3. FAPbI3

Formamidinium lead halide perovskites CH(NH2)2PbI3 (FAPbI3) have been used to create PSCs having enhanced efficiency and greater stability when compared to their more widely used methylammonium counterparts. Compared with MAPbI3 and CsPbI3, the increased ionic radius of FA results in FAPbI3 having a broader optical absorption band and reduced band-gap (1.48 eV compared to 1.57 eV of MAPbI3 and 1.73 eV of CsPbI3 as shown in figure 5) [59]. FAPbI3 is also more stable as it does not undergo a structural phase transition within the operational temperature-range of the solar cell. Lee et al [60] explored a FAPbI3 film (of thickness ~300 nm) that was capped by a thin MAPbI3 layer on top of mesoporous TiO2. Such PSCs were shown to have a PCE of 16.01%, with a short circuit current (Jsc) of 20.97 mA cm−2, an open circuit voltage (Voc) of 1.032 V and a fill-factor (FF) of 74%. Here, the thin MAPbI3 capping layer absorbed light at wavelengths longer than 700 nm (a spectral region where absorption by FAPbI3 is weak) and thus improved Jsc and Voc. The high FF observed was attributed to a decreased series resistance (Rs) and an increased shunt resistance (Rsh). FAPbI3 is usually synthesized from a combination of FAI and PbI2, however Wang et al [61] utilized a reaction between FAI and the precursor HPbI3, with the HPbI3 being synthesized by mixing PbI2 with HI at a molar ratio of 1 : 1.5. The PSCs incorporating FAPbI3 prepared using this new approach achieved a solar cell efficiency of 17.5%. In addition, the

2.2. MAPbI3−xClx

The mixed halide perovskite MAPbI3−xClx can be created by using both Cl− and I− as the halide anions. The perovskite MAPbI3−xClx has been found to have a long charge diffusion length and improved stability in air compared to MAPbI3. However the presence and role played by chlorine in MAPbI3−xClx is controversial. Grätzel [53] suggested that the chloride-doped state would not exist in MAPbI3−xClx because chloride ions—being part of an amorphous lead-containing phase—would melt at 103 °C. Other work found that chlorine affects the nucleation dynamics of perovskites but was not 3

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 3.  (a) A schematic representation of nucleation during deposition (top), phase evolution and growth during annealing (middle), and final morphology (bottom) of a perovskite film cast from a MAI:PbCl2 (3 : 1) precursor solution. (b-d) shows SEM images of three representative morphological components (scale bars: 5 μm). Figure reprinted with permission from [54]. Copyright 2014 American Chemical Society.

rates and high device efficiency. It has also known that the hole and electron diffusion lengths are not necessarily balanced. For example, Eperon et al [59] reported that the hole diffusion length for FAPbI3 is (813  ±  72) nm; a value much longer than its electron diffusion length of (177  ±  20) nm (see table 1). A higher LD is associated with reduced total recombination rates (Rtotal), as well as longer carrier lifetime (τe) [65] as expressed by LD = D /Rtotal (n) [66], where D is a materialdependent diffusion constant. A reduced recombination rate is known to increase FF, Voc, and PCE, external quantum efficiency (EQE) and Rs. Very recently Dong et al [67] synthesized millimeter-sized MAPbI3 single crystals via a lowtemperature solution process and reported an electron-hole diffusion length exceeding 175 μm under AM 1.5 irradiation; a value that increased to 3 mm under weak light illumination.

FAPbI3 films fabricated from the FAI/HPbI3 precursor were found to be more uniform and dense and had higher crystallinity and improved orientation compared with their counterparts prepared from FAI/PbI2 thus resulting in better carrier transport and improved device efficiency. These enhanced properties were attributed to the fact that it was not necessary to add HI to the precursor solution in order to produce a uniform, smooth perovskite film, unlike previous work using formamidinium [59]. Here, the use of HI—whilst being beneficial— involves the introduction of water to the system and can have an adverse effect on perovskite crystallization. 3.  Carrier diffusion length and device metrics 3.1.  Carrier diffusion length

The charge carrier diffusion length (LD) within both mixed halide and triiodide perovskite can be significant, with stateof-the-art values of LD being enhanced through control of film composition and morphology, with grain size being sensitive to deposition methods as well as processing conditions [62]. It was initially reported that the mixed-halide perovskite (MAPbI3−xClx) had a longer diffusion length (LD  >  1 μm) than the pure triiodide perovskite (MAPbI3) whose diffusion length was measured to be ~100 nm [63]. In subsequent work, Zhao et al [64] reported an electron diffusion length longer than 1 μm for the perovskite MAPbI3 in a meso-structured device configuration that had a 650 nm thick TiO2 layer. Such diffusion lengths are greater than the thickness of the perovskite layer (typically around 300–500 nm); a property consistent with, long carrier lifetimes (τe) and low recombination

3.2.  Device metrics

A key objective of perovskite solar cells research is to improve device power conversion efficiency (PCE); a parameter given by PCE = Jsc × Voc × FF /Pin where Pin is the input optical power. Currently, the best literature values of Jsc, Voc and FF reported in perovskite solar cells are 27.4mA cm−2 [68], 1.5 V [69] and 86.7% [70] respectively. We emphasize that so far such high-value parameters have not been simultaneously obtained in a single perovskite solar cell. The Jsc is related to light absorption and photocurrent conversion. It is thus clearly dependent on the thickness of the active perovskite layer, and is also detrimentally affected by incomplete film coverage owing to reduced absorption. 4

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 4.  Isodensity plot of the integrated DOS as a function of the distance from TiO2 surface for MAPbI3 and interfacial MAPbI3−xClx. A blue to red color variation indicates an increase of the DOS value. The red circles highlight energy range where occupied chloride states are found in MAPbI3−xClx but not in MAPbI3. Figure reprinted with permission from [57]. Copyright 2014 American Chemical Society.

[68]. This improvement originates from the improved electrical conductivity of PCBM compared to TiO2 [68], as well as the passivation effect of PCBM at the TiO2/MAPbI3 interface [71, 72]. The Voc is proportional to the difference (Δ) between the quasi-Fermi levels of the electron (Efn) and hole (Efe) as shown in figure 6 [73]. When using MAPbI3 as the photoactive layer, the expected maximum value of Voc ranges from 1.32 to 1.34 V in a device in which there is only radiative recombination; a value that is expected to be reduced further by nonradiative losses, e.g. by adopting a TiO2 scaffold into which electrons are injected [74]. MAPbBr3, having a wider bandgap (E.g.  =  2.3 eV) than MAPbI3 (E.g.  =  1.55 eV), results in a larger value of Δ and thus increases Voc to 1.40 V [73]. In the device FTO/c-TiO2/mp-Al2O3/MAPbBr3−xClx/4,4′bis(N-carbazolyl)-1,1′-biphenyl (CPB)/Au, a record value of Voc of 1.5 V was observed as a result of the HTL used having a deep HOMO level, as well as a reduced recombination rate in the photoactive layer [69]. Better charge transport also contributes to higher values of Voc [75]. Moreover, metal oxide HTLs (e.g. Cu:NiOx) that have better electrical conductivity than well explored HTL materials such as PEDOT:PSS, result in larger values of Voc. Finally, a better energy level alignment between charge transport layers and the perovskite will also enhance the Voc [76].

Figure 5.  UV-Vis spectra of MAPbI3, FAPbI3 and CsPbI3 perovskite films. Figure reprinted with permission from [59]. Copyright 2014 the Royal Society of Chemistry.

The Jsc is also affected by the efficiency of charge extraction. For example following the introduction of the water-soluble fullerene derivative (C60-Ac10) and [6]-phenyl-C61-butyric acid methyl ester (PCBM) between a perovskite layer and a compact TiO2 layer, electron extraction was enhanced and resulted in an improvement in Jsc from 20.9 to 27.4 mAcm−2 5

Review

Rep. Prog. Phys. 79 (2016) 026501

Table 1.  Summary of diffusion length (LD) in different perovskite material systems.

Perovskite photoactive layer LD MAPbI3−xClx MAPbI3 MAPbI3 FAPbI3 MAPbI3 MAPbI3 MAPbI3 MAPbI3 MAPbI3

>1 μm 100 nm 1 μm hole (813  ±  72) nm (L D )

) (177  ±  20) nm (L electron D 886 nm 282 nm >3 mm (light intensity  =  0.003 mW cm−2) (175  ±  25) μm (light intensity  =  100 mW cm−2) >3.5 μm 721 nm 390 nm

Average grain size

Ref.

N/A N/A N/A N/A

[63] [64] [59]

N/A 400 nm 260 nm 3.3 mm

[111] [67]

250–2250 nm N/A N/A

[95] [166] [50]

Note: The LD in table 1 is not a conclusive summary from the literature.

Fullerene layers, e.g. C60-substituted benzoic acid self-assembled monolayer (C60-SAM), indene-C60 bisadduct (ICBA) and PCBM etc. have been used to passivate the charge-transport layer/perovskite interface and suppress pin-hole defects and reduce recombination rates [71, 72, 79–81]. To summarize this section, we have shown that careful design of the perovskite photoactive layer and the device structure can result in an improvement of Jsc, Voc and FF; and permits the power conversion efficiency of a PSC device to be optimized. 4.  Correlating perovskite morphology with device efficiency The morphology of the perovskite active layer has a significant impact on the final performance of the device. High coverage, and uniform, crystalline domains are beneficial to achieve higher power conversion efficiencies [78, 82]. Incomplete surface coverage may result in direct contact between electrodes or charge transport layers, thereby leading to increased cur­ rent leakage and decreased efficiency as well as large variations in performance. Generally, morphology is influenced by the composition of the perovskite layer, solvent, depositionmethod and any subsequent post-treatments. A number of recent reviews address the impact of morphology on device performance [54, 83, 84]. Figure 6.  Schematic of the proportion between Voc and Δ

4.1.  The impact of film composition

(difference between Efn and Efp) in bilayered device architecture. Figure reprinted with permission from [73]. Copyright 2014 the Royal Society of Chemistry.

Since their introduction by Snaith et al mixed-halide perovskite photoactive layers have been studied extensively for their potential to increase PSC efficiency through their modified optoelectronic properties as well as improved film morph­ology [7]. For example, a PSC having a (FAPbI3)1−x(MAPbBr3)x layer was fabricated having a structure of (FTO/TiO2/ [(FAPbI3)1−x(MAPbBr3)x:mp-TiO2]/PTAA/Au) and had an average PCE of 17.3% (maximum PCE  =  20.3%) at steadystate current (i.e. with a stabilized power output) [12]. Here, the proportion of MAPbBr3 in the FAPbI3 was shown to influence the PCE, surface roughness as well as the device hysteresis.

Device FF is mainly affected by charge carrier transport and the recombination rate. Usually the use of a thinner perovskite layer will increase FF due to efficient charge collection and transport [73] and reduced recombination [50, 71]. Morphological and energetic defects are also known to affect the recombination rate. Morphological defects include pinholes at mesoscale and crystal defects at nanometer length scale. Pin-holes will reduce Rsh and bring the ETM and HTM into direct contact and result a reduction of the FF [77, 78]. 6

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 7.  SEM images of (FAPbI3)1−x(MAPbBr3)x films with x  =  0, 0.05 and 0.15. Figure reprinted with permission from [12]. Copyright 2015 Nature Publishing Group.

as uniform thickness. However, thermal evaporation consumes a significant amount of energy and necessitates the use of high vacuum. This technique therefore increases the cost of device fabrication and will increase the embodied energy within the device. Yang et al [90] reported a low-temperature vaporassisted solution process by in situ reaction of as-deposited PbI2 film with MAI vapor and obtained an efficiency of 12.1% for PPSCs. The VASP enables sufficient conversion of PbI2 into MAPbI3 with high film coverage and low roughness. Solution-casting however, can be used to create uniform perovskite layers having high device efficiency and can be used as the basis of a low-cost, low-energy manufacturing process. A number of different solution deposition techniques have been developed [37]. In the so-called ‘one-step’ deposition processes, the perovskite precursors are dissolved in a common solution before they are cast into a film and solar cells having an efficiency of around 20% have been created using this technique [12, 13]. In a two-step (sequential) deposition processes, a solid film of one precursor (usually a metal halide, e.g. PbI2) is first cast onto a substrate, with the second precursor typically introduced by dipping the film into a solution (see figure 8(a)) [91]. Alternately a two-step process can be realized by creating a multilayer structure of precursors that are then mixed using thermally-assisted diffusion (see figure 8(b)) [92]. Sometimes, both one-step and two-step solution casting methods can be employed to reduce defects. For example, Li and Ching et al [93] created a kind of bilayer perovskite structure using a two-step method to form a perovskite on an mp-TiO2 surface. A second perovskite capping layer was then spin coated via a one-step process onto the surface to improve film coverage and reduce film roughness. The capping layer was extremely compact and filled the pores of the meso-structured layer, leading to a device having a PCE of 14.4%; a value higher than that of a representative device (9.9%) prepared from a one-step casting method. More sophisticated morphology manipulation can be carried out during the solution casting process. We give specific examples of the control of perovskite grain size later in this review. We note, however, that other techniques apart from the well-explored spin-coating process have been used to prepare perovskite films. For example, Barrows et al [94] reported planar perovskite solar cells deposited by spray coating a precursor solution in air, with a maximum PCE of 11% achieved.

For values of x  =  0.15, i.e. (FAPbI3)0.85(MAPBr3)0.15, the perovskite was characterized by an extremely smooth surface without observable pin-holes (see figure 7), that—when fabricated into a PSC device—has favorable PCE and an EQE  >  80% over the range 400 nm to 750 nm [12]. Williams et al made a comprehensive study of the role of chloride concentration on the morphology of MAPbI3−xClx films prepared from a range of different solutions [54]. It was shown that the presence of chloride changed the nucleation dynamics and lead to the rapid formation of MAPbCl3 intermediate phase which generated pin-holes after thermal annealing (see figure  3). However others have found that the use of PbCl2 rather than PbI2 as a precursor in a one-step process leads to an improved morphology (fewer pin-holes) as well as increased grain size, improved crystallinity and an increase in preferential orientation. This has been attributed to the MA+ rich environment which is typical with a PbCl2 precursor, since standard molar ratios in the precursor solutions are 1 : 1 MAI:PbI2 for the fabrication of MAPbI3 and 3 : 1 MAI:PbCl2 for the fabrication of MAPbI3−xClx. This environment slows the crystallization, ensuring the formation of a dense and uniform perovskite layer [85, 86]. In addition, the incorporation of chlorine facilitates easier removal of the organic component during transformation to the perovskite, since MACl is removed at a lower temper­ ature than MAI. This allows for the use of a lower temperature thermal anneal which further reduces the formation of pinholes and voids [56, 85, 87]. PbCl2 has also been proposed to be important as nucleation sites for crystallization [88]. Carnie et al demonstrated that pin-holes in perovskite films can be passivated by adding up to 0.75 wt% of 3-aminopropyl (3-oxobutanoic acid) functionalized silica nanoparticles (f-SiO2) as additives to the precursor solution during processing [89]. 4.2.  Film deposition methods

Dual-source vacuum thermal evaporation (DSVD) and vaporassisted solution process (VASP) are effective techniques to control crystal growth and have been used to create smooth and uniform perovskite films. In DSVD, the organic and inorganic vapors are separately controlled with two sources, and their deposition rates can be varied to tune the composition. The PPSCs fabricated by Snaith et al [11] using this method showed flat photoactive layer without apparent holes, as well 7

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 8.  Diagram of two-step solution casting processes to form the perovskite photoactive layer. In part (a) MAPbI3−xBrx is formed by dipping a PbI2/PbBr2 film into a MABr:MAI solution. Figure reprinted from [91]. Copyright 2014 American Chemical Society. In part (b), MAPbI3 is formed by the interdiffusion of PbI2 and MAI films driven by thermal annealing. Figure reprinted with permission from [92]. Copyright 2014 The Royal Society of Chemistry.

Figure 9.  (a) Formation of intermediate phases during the two-step solution casting process of a perovskite. Figure reprinted from [98]. Copyright 2014 American Chemical Society. (b) XRD spectra of MAI-PbI2-DMSO intermediate phase powder as a function of annealing temperature. The intermediate phase changes completely at 130 °C, with MAI-PbI2-DMSO and perovskite phases coexisting at 100 °C. Figure reprinted with permission from [99]. Copyright 2014 Nature Publishing Group.

Deng et al [95] reported the fabrication triiodide perovskite solar cells using a doctor blade, with a PCE of 15.1% reported. Hwang et al [96] used slot-die coating process to fabricate perovskite solar cells (excluding the electrodes) having a maximum PCE of 11.96%. Recently, perovskite solar cells fabricated by slot-die coating under ambient conditions achieved a PCE of 4.9% [97]. Here flexible ITO-PET was employed as the device substrate with a printed Ag layer as the back electrode. It is apparent that such deposition methods can be easily scaled up to prepare large-area devices on flexible or rigid device substrates, offering significant potential for commercialization.

and can be reduced by thermal annealing above a certain temper­ature, as shown in figure  9(b) [99]. It is thought that the volume occupied by DMF and H2O turns into voids or holes when the solvent molecules are driven out of the crystal during the thermal annealing process. Recent work has demonstrated that whereas spin coating a FAI/PbI2 precursor from a DMF solution resulted in discontinuous perovskite film, the presence of a small amount of hydroiodic acid into the DMF solution added just before spin coating leads to extremely uniform and continuous films, as shown in figure 10. When using a volatile solvent such as DMF (having a boiling point see 153 °C [100]), it can be difficult to control the crystal structure of the perovskite. Higher boiling point solvents (e.g. dimethylsulfoxide (DMSO) or γ-butyrolactone (GBL)) are usually able to create more uniform crystal domains and smoother film surfaces due to their control over the crystallinity of the inorganic PbI2 and the subsequent reaction of PbI2 with MAI [101, 102]. Using a mixture of GBL and DMF (DMF:GBL 97:3 [vol%]), Kim et al [101] created a uniform and smooth perovskite film having an efficiency of 6.16%. The use of mixed solvents containing a small amount of GBL can induce the formation of small and uniform crystallites that can result in improved interfacial contact and reduced formation of voids. This in turn can lead to enhanced Jsc and FF as a result of efficient exciton dissociation and reduced recombination at the interface.

4.3.  Effects of casting solvent and solvent additives

As different solvents can have different evaporation rates and can solubilize precursors with different efficiency, various solvents and solvent additives can be used to modify the structural uniformity and crystallinity within a perovskite film. Perovskite films cast from a DMF solvent do not usually completely cover a surface, but instead contain voids and pinholes present at the surface or in the bulk of the active layer. Such voids or pin-holes are thought to originate from the formation of intermediate compounds, e.g. MAPbI3·DMSO, MAPbI3·DMF or MAPbI3·H2O (figure 9(a)) [98]. These intermediate phases have been studied by x-ray diffraction (XRD) 8

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 10.  Surface SEM images of perovskite FAPbI3 films formed from spin-coating the precursor solution: (a) without hydroiodic acid, and (b) with a small amount of hydroiodic acid before spin-coating. Figure reprinted with permission from [59]. Copyright 2014 the Royal Society of Chemistry.

The solvent DMSO has also been used to improve the morphology and crystallization of perovskite films prepared using a two-step process. MAPbI3 films in which the initial PbI2 layer was deposited from DMSO rather than DMF were found to be comparatively smoother and contained more uniformly sized grains with reduced PbI2 residue than those prepared using DMF. This resulted in higher efficiency PSCs having a reduced standard deviation (SD) in their efficiency (specifically, average PCEDMSO  =  12.5%, SDDMSO  =  0.25, average PCEDMF  =  9.7%, SDDMF  =  2.47) [103]. Here the DMSO can retard the crystallization of PbI2, forming amorphous PbI2 that is mainly in the PbI2-DMSO intermediate phase [99, 103]. Whilst the presence of DMSO molecules initially reduces the reaction rate between PbI2 and MAI during the second (dipping) step, the large PbI2 crystals formed from the DMF precursor mean that full conversion is considerably slower than with the amorphous layer from the DMSO precursor. This lead to a high degree of non-uniformity and low reproducibility in devices prepared using the DMF precursor, and in the case of thick PbI2 films even resulted in the perovskite film peeling off the substrate during the prolonged dipping process. Small molecule additives have also been introduced to the precursor to improve the quality of perovskite films. For example, a two-step method was explored to fabricate a perovskite film from PbI2 and MAI using PEDOT:PSS as the HTL. Here 2 wt% of H2O was used as an additive to the PbI2/DMF precursor, and was found to improve the solubility of PbI2 in DMF, and to dramatically improve the uniformity of the PbI2 film by reducing its roughness and the density of voids. The presence of H2O was thought to enhance the PbI2 adhesion to the hydrophilic PEDOT:PSS surface and improve the surface coverage after spin casting by matching the interfacial energies. Such a uniform PbI2 layer is then likely to help the creation of perovskite films having less defects after casting the MAI. Devices prepared using this technique had champion PCEs of 18%, with a FF as high as 85% [104] In other work, hydrochloric acid was added to the PbI2 and MAPbI3 precursors to assist the two- and one- step solution deposition processes respectively, with devices created having a PCE over 14% [105].

Figure 11.  In situ GIWAXS showing that PbI2 crystals start to form once the MAPbI3−xClxfilm has been held at 100 °C for 10mins. Note that the time series alongside the temperature labels are the periods from time T  =  0 to T  =  T ′, at which point the temperature of the film was raised to the value in the brackets. Figure reprinted from [106], with permission from authors.

4.4.  Thermal annealing

A thermal annealing step is sometimes necessary to fully convert the precursor film into perovskite after solution casting. This thermal annealing process can change the crystal structure, the grain size and the film uniformity. Excessively high annealing temperatures are able to grow large-size perovskite grains, but tend to reduce the surface coverage and most often result in devices having reduced efficiency. It was observed that PbI2 crystals start to form when the perovskite film (specifically MAPbI3−xClx) is annealed for 10 min at 100 °C (see figure 11) [106]. This result has been confirmed by both XRD and SEM studies [62, 106]. When the annealing temperature is increased above 110 °C, the growth of pin-holes is increased and the continuous film then splits into a series of isolated islands (see figure 12) [63, 78]. The large gaps between the islands can result in the formation of direct contacts between the HTL and ETL and result in reduced device efficiency [107]. 9

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 12.  SEM images of perovskite films (a) in an as-cast state (the unannealed precursor). Reprinted with permission from [63].

Copyright 2013 American Association for the Advancement of Science. Perovskite film (b-d) upon thermal annealing at 90 °C, 130 °C and 170 °C, respectively. Figure reprinted with permission from [78]. Copyright 2014 Wiley-VCH.

Figure 13.  (A) Schematic illustration of methylamine induced defect healing of MAPbI3 films. SEM images of (B) raw and (C) healed MAPbI3 films (insets: higher magnification SEM images). AFM images of (D) raw and (E) healed MAPbI3 films. Reprinted with permission from [110]. Copyright 2015 Wiley-VCH.

Two-step annealing techniques have been used to improve film morphology and PCE. Here annealing at 90 °C for 30 min followed by a second anneal at 100 °C for 2 min was found to create a more continuous and crystalline film than annealing at 100 °C for 5 min [108]. Similarly, a slow ramp to 100 oC, followed by an isothermal anneal at 100 °C for 45 min was compared against a process that involved annealing the films at 100 °C for 5 min followed by a fast ramp to 130 °C for 5 min (dubbed ‘flash annealing’). Whilst the flash annealing proved detrimental for devices based on an Al2O3 scaffold due to the formation of pores in the perovskite capping layer, it improved the performance of planar devices due to

the formation of more uniform and larger-sized perovskite domains [82]. An even faster annealing time of 2.5 s for perovskite crystallization (film thickness 450–500 nm) was realized using near-infrared (NIR) radiation, producing devices with a PCE of 10.0% [109]. 4.5.  Gas-induced defect healing

Gas-induced defect healing has been recently demonstrated to heal voids and pin-holes in perovskite films [110]. Methylamine (CH3NH2) gas was found to be particularly efficient for this purpose. Here, the MAPbI3 absorbs CH3NH2 10

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 14.  (a) A schematic showing an interdiffusion and solvent annealing process. Cross-sectional SEM images of perovskite films

formed by (b) thermal annealing and (c) solvent annealing. Reprinted with permission from [111]. Copyright 2014 Wiley-VCH.

molecules, forming an intermediate MAPbI3·xCH3NH2 liquid phase. The collapse of the perovskite structure into a clear liquid then undergoes spreading; a process that can reduce the presence of voids and pin-holes, forming an ultra-smooth film in a short time. The intermediate MAPbI3·xCH3NH2 liquid can revert to a crystalline MAPbI3 film after the removal of CH3NH2 gas, as depicted in figure 13(A). This CH3NH2 gasinduced defect-healing behavior was found to reduce surface roughness from see 153 nm in a ‘raw’ film to see 6 nm in a ‘healed’ film. This technique was used to prepare dense and smooth MAPbI3 films by healing dendrite-like MAPbI3 crystals (see figures  13(B)–(E)), and thereby improving device performance, with PCEs increasing from 5.7% to 15.1% as a result of the defect healing treatment. This gas-induced defect healing process can be used to heal films cast from one-step solution deposition using different precursors as well as films prepared by sequential solution deposition. The short reaction time and easy processing techniques used mean that this method can be employed to heal large-area perovskite films deposited on flexible substrates.

the crystallinity, grain-size and reduced defect density. As shown in figure 14, large grains can provide a contact between anode and cathode, permitting photogenerated charges to be extracted through the device electrodes without having to pass through any grain boundaries during their transport in the out of plane direction. When both thermal and solvent annealing process were used, MAPbI3 devices were demonstrated to have higher PCEs. Similarly, Lian et al found that introducing either dichlorobenzene or DMSO vapors during both the spincoating and annealing steps lead to the formation of larger, more orientated grains and an increased PCE [112]. A recrystallization procedure involving a series of exposures to a DMF vapor followed by a re-annealing step has also yielded significant enhancements in performance [113]. The effect of grain size, including other approaches to manipulate the grain sizes in perovskite films will be discussed in the following section. 4.7.  Effect of grain size

Recent work on the control of the solution casting process used to create perovskite films has demonstrated two different routes to prepare the photoactive layers with different grain sizes that can operate in PSCs with high efficiency. The first approach reduces the crystallization rate by using solvents (e.g. DMSO) that coordinate with PbI2 and therefore slow down the intercalation process of MAI into PbI2 during a two-step casting process [99, 103]. This results in perovskite crystals having a small but uniform grain size of ~200 nm, rather than having a size distribution between 50 and 330 nm as occurs when using a DMF solvent that does not undergo such coordination

4.6.  Solvent vapor annealing

Exposure to a solvent vapor has been found to be able to improve the crystallinity, grain size, and uniformity of a perovskite film. The solvent vapor provides a wet environ­ ment so that the precursor ions and molecules can diffuse and react, resulting in a reorganization of film morphology (see figure  14(a)). Xiao and Dong et al [111] reported that solvent annealing of MAPbI3 with DMF vapor increased 11

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 15.  SEM images of perovskite obtained from (a) DMF and (b) DMSO based PbI2 films showing the different distribution of grain size obtained, (c) Schematic of the retarded crystallization of PbI2 using the solvent DMSO, (d) efficiency distribution with data shown for grains of around (200  ±  20) nm (group A) or from 50 to 330 nm (group B) prepared from DMSO and DMF based PbI2 films, respectively. Figure reprinted with permission from [103]. Copyright 2014 The Royal Society of Chemistry.

interactions. Here, the improved film quality results in PSCs having an average PCE of 12.5% with a standard deviation of 0.57 (see figure  15). Such small grain sizes however increase the volume of grain boundaries within the perovskite layer which can act as charge traps and facilitate trap assistant recombination, thereby limiting device efficiency. A second approach to control film morphology accelerates the dynamics of crystallization using solvent [100, 114] or temperature control [115], creating crystals of micron-size or larger. Such large perovskite crystal grains improve carrier diffusion and transport compared to crystalline grains that are tens or hundreds of nanometers in size [116, 117].As has been briefly mentioned, solvent annealing has been shown to be an efficient approach to produce perovskites with larger grains compared to that using thermal annealing alone. Manipulation of crystal growth process using solvent exposure during the solution casting process has been demonstrated to allow control of grain size. For example, Xiao et al [100] controlled perovskite grain size and uniformity by spin-coating a second solvent chlorobenzene (CB) onto a still wet MAPbI3 film, that was itself spin cast from a DMF solution (see figure 16). The uniform perovskite film formed had micron-sized grains with

PSCs having an average PCE of 13.9% with an SD of 0.7%; a value significant higher than that of devices made from a conventional spin casting process (1.5%). The solubility of the second solvent used with the MAI precursor is able to control the structure of the perovskite film created. For solvents in which MAI has a high solubility (e.g. methanol, ethanol and ethylene glycol) a yellow film is formed, containing with a large fraction of PbI2 rather than MAPbI3 crystals. By using solvents in which MAPbI3 has a low solubility (e.g. chlorobenzene, benzene, xylene, toluene, 2-propanol and chloroform) MAPbI3 films are rapidly formed that are composed of micron-sized grains that have high surface coverage. However the point at which the second solvent is added to the stillwet perovskite film is critical; a fast crystallization forming a dense and uniform film can only be achieved when the wet film reaches a supersaturated state following the evaporation of a certain amount of solvent. This period was found to be 4 to 6 s after spin coating the wet perovskite film from DMF. This sensitivity brings a number of processing challenges, as DMF can evaporate quickly during spin coating. In other recent work, Yang et al [118] fabricated FAPbI3 films using a direct intramolecular exchange process (IEP) 12

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 16.  (a) Schematic illustration of a conventional spin-coating process used to create a perovskite film (top) and a solvent

treatment process during the spin coating used to induce fast crystallization of the perovskite film into uniform crystallites (named as fast crystallization-deposition (FDC)) (bottom). (b) Surface SEM images of FDC. Figure reprinted with permission from [100]. Copyright 2014 Wiley-VCH.

Figure 17.  (A) Schematics of FAPbI3 perovskite crystallization involving the direct intramolecular exchange of DMSO molecules intercalated in PbI2 with FAI. (B) Surface SEM images showing different grain sizes of FAPbI3 based perovskite films formed on mp-TiO2 by (a) intramolecular exchange (IEP) and (b) conventional methods. Representative J–V curves (C) and histogram of device efficiency (D) of FAPbI3 based device fabricated by IEP and conventional methods. Figure reprinted with permission from [119]. Copyright 2015 American Association for the Advancement of Science.

Guo et al [121] reported that the cooling rate of perovskites following thermal anneal could affect grain-size. For example using rapid cooling, large-size grains (~2 μm) could be created having a high surface-coverage of (99.55  ±  0.24)% as shown in figure 18(b). Slowly cooling the films seemed however to induce void formation (see figure  18(a)), with films having a surface coverage of 82.2%. Devices fabricated from the rapidly cooled films had a higher PCE of 12.4% compared with that of 6 to 8% obtained using slowly cooled films. Nie et al [115] fabricated perovskite films having very large (1 to 2 mm) crystalline grains using a one-step hot spincasting technique (see figure  19), in which a solution held at a temperature of 70 °C was spin-cast on device substrates held at different temperatures. This technique was used to create devices having a PCE of 17.7% without any noticeable hysteresis in the J–V curve. A dichloromethane dip during a hot drying process has also been shown to form a more uniform perovskite film characterized by large crystal grains compared with those formed by drying at 100 °C or drying at room temper­ature [122]. In general the charge diffusion length within large crystal domains is much increased, resulting in improved charge transport. Furthermore, the volume of grain boundaries within perovskite films having large crystallite grains is significantly reduced, thus reducing the density of defects available to facilitate charge-recombination. The

whereby FAI replaced DMSO molecules that were intercalated in PbI2. In this work, a PbI2 (DMSO) precursor was first synthesized and used to cast a film onto which a FAI(MABr) solution was then spin cast, allowing intramolecular exchange (the MABr was added to enhance stability) of FAI with DMSO (see figure 17(A)). This process did not need further thermal annealing and produced a flat FAPbI3 film without forming residual PbI2. The FAPbI3 films had a (1 1 1) preferred crystal orientation, and contained larger grains compared with their counterparts prepared using conventional PbI2 (figure 17(B)). PSC devices fabricated had a maximum PCE exceeding 20% and an average PCE of over 19%; values that were enhanced compared to those prepared from a sequential reaction using conventional PbI2 in which an averaged PCE around 15% was achieved. Perovskite grains can also grow though well-controlled thermal annealing. Zhu et al [119] developed controlled nucleation/crystallization and grain-growth processes by using a room temperature solvent-solvent extraction approach [120] for MAI:PbI2 precursor (MAI:PbI2  =  1.2 : 1) followed by thermal annealing at 150 °C for 15 min in air. It was found that using such techniques, the in-plane grain size of perovskite grew from see 30–150 nm to see 1–2 μm; a length that was between three and six times the thickness of the perovskite layer (see 350 nm). 13

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 18.  SEM image of a perovskite layer coated on ITO/TiOx after heating at 95 °C and (a) slowly cooled, and (b) rapidly cooled. The inset shows high magnification images with the red line showing grain boundaries. Figure reprinted from [121], with permission from authors.

Figure 19.  Optical micrographs illustrating grain formation as a function of substrate temperature with the casting solution maintained at

70 °C. Figure reprinted with permission from [115]. Copyright 2015 American Association for the Advancement of Science.

generation of perovskite films having large crystal grains thus appears to be a promising strategy to fabricate perovskite solar cells having high operational efficiency.

and extraction is of course required in order to produce high efficiency solar cells. A recent review on different hole transport materials for perovskite solar cells can be found in the literature [128]. HTM and ETM layers can be used to create perovskite solar cells having higher operational efficiency than devices in which these layers are absent, albeit at a greater economic cost. This is because interfacial barriers within a perovskite device can affect charge transport efficiency and thus the charge recombination rate. Perovskite solar cells are most often comprised of multiple layers separated by physical and electronic interfaces at which various defects can exist, including voids, incomplete coverage, energetic barriers etc. Such interfaces can, however be engineered to facilitate carrier extraction and transport. For example, Zhang et al [129] increased the average PCE from 8.52% to 11.28% and 12.01% by intercalating ultra-thin poly[3(6-trimethylammoniumhexyl)thiophene] (P3TMAHT) and polyethylenimine ethoxylated (PEIE) layers respectively in a ITO/ PEDOT:PSS/MAPbI3−xClx/PCBM/Interlayer/Ag device. Here, the PEIE and P3TMAHT layers effectively reduced the work function of the Ag and facilitated carrier transport across the electrode-junction barrier that helped to increase device efficiency. Yang et al [13] modified the electronic barrier at the ITO/TiO2 interface in an ITO/PEIE/PCBM/Y:TiO2/ MAPbI3/spiro-OMeTAD/Au device using a PEIE layer. This layer reduced the work function of the ITO from 4.6 eV to 4.0 eV, thereby increasing the Jsc from 18.9 to 19.9 mA cm−2 and increasing the FF from 65.25% to 73.28% (see figure 20). Dedicated reviews on interfacial engineering of perovskite solar cells can be found in the literature [130, 131].

5.  Influence of device architecture 5.1.  Interfacial engineering

In order to extract charges from a perovskite device, they must be transported across a series of interfaces, including between the perovskite and the HTL or ETL and between the HTL (or ETL) and the device electrodes. In devices in which no HTL or ETL is employed, it is clearly necessary to transport charge directly between the perovskite and the electrode. As a result of its ambipolar nature, perovskites can theoretically act as both electron and hole conductors. Indeed, HTL-free [46, 123–125] or ETL-free [126] perovskite solar cells have been reported, having PCEs between 10 and 15%. However, Edri et al [127] revealed that efficient mesoporous perovskite solar cells based on MAPbI3 require mp-TiO2 acting as an ETL but not necessarily as an HTL. This is because electrons have a comparatively shorter diffusion length in a perovskite than holes, and thus an mp-TiO2 layer is needed to efficiently transport electrons to the electrode. Mei et al [33] developed an HTL-free perovskite solar cell incorporating a mesoporous layer made of TiO2/ZrO2/C, resulting in PSCs having an efficiency of 12.8%, a lowered series resistance and high electric conductivity. This demonstrated that an HTL is not a necessary component of a perovskite device, although ensuring that the resulting device has effective hole and electron transport 14

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 20.  (a) SEM image of device cross-section. (b) Diagram of energy levels (relative to the vacuum level) of each layer.

Figure reprinted with permission from [13]. Copyright 2014 American Association for the Advancement of Science.

that incorporated TiO2 nanotubes (PCEs of 9.0% and 6.2% respectively). By using TiO2-coated ZnO nanowires and ZnO nanowires, device efficiencies of 4.0% and 3.0% respectively were obtained [145]. PSCs based on a nanostructured doublelayer ZnO film in which vertically aligned nanosheets covered with an overlayer of horizontally aligned nanorods displayed a champion PCE of 10.35% and had good photostability [151]. PPSCs can be fabricated using either a normal structure (figure 1(b)) or with an inverted structure (see figure 1(c)). The normal structure is similar to that of MPSCs, except that it does not utilize mesoporous materials. Inverted structures typically use organic-semiconductors as both HTL (e.g. PEDOT:PSS) [152–155] and ETL (e.g. PCBM, ICBA, C60-SAM or polyelectrolyte poly[(9,9-bis(3’-(N,N-dimethylamino)propyl)-2,7fluorene)-alt-2,7-(9,9-dioctylfluorene)] (PFN)), and form the basis of PPSCs devices [156]. Heo et al [50] compared the inverted structure (ITO/PEDOT:PSS/MAPbI3/PCBM/Au) with a normal structure (FTO/TiO2/MAPbI3/PTAA/Au), and found that the inverted structure had a higher efficiency. You et al [157] fabricated an inverted structure consisting of ITO/PEDOT:PSS/ MAPbI3−xClx/PCBM/PFN/Al in ambient air with humidity of (35  ±  5)% and achieved an average PCE of 15.4% (maximum PCE  =  17.1%). The hydrophilic nature of PEDOT:PSS is a notable challenge for inverted planar devices as moisture can be detrimental to the perovskite. The organic-semiconductors used in planar inverted structure do, however offer the advantage of relatively simple compatibility with the low temper­ ature deposition processes that are essential for utilization of flexible substrates, as well as offering more general compatibility with many roll-to-roll deposition methods.

5.2.  Meso-superstructured (MPSC) versus planar (PPSC) perovskite solar cells

The key distinction between MPSCs and PPSCs comes through the use of mesoporous ‘scaffold’. In an MPSC, a compact layer (e.g. compact TiO2) is usually deposited onto a fluorine-doped tin-oxide layer (FTO) which is used to extract electrons and block holes. The compact TiO2 layer is usually deposited by spin coating or spray pyrolysis and typically requires a sintering or calcination step performed at around 500 °C in air. This high temperature prevents the fabrication of MPSCs using plastic-based substrate materials that undergo rapid degradation at high temperatures. However, low temper­ ature sintering approaches to prepare an effective TiO2 layer in perovskite solar cells have been recently reported [132, 133]. Ideally, the compact TiO2 layer should contain few nano-scale pinholes, with TiO2 compact layers produced by atomic layer deposition usually characterized by fewer pinholes and thus improved device efficiency [134]. A mesoporous layer (such as mp-TiO2, mp-Al2O3, mp-ZnO, mp-ZrO2 or mp-NiO) is then deposited on top of the compact layer [135–139]. Such mesoporous layers can be deposited using a range of techniques, including spin coating, screen printing [140], magnetron sputtering [141] and electrospinning [142]. The perovskite is then deposited on the mesoporous layer, which is finally capped with an HTM and an electrode to complete the device. When preparing MPSCs, a full infiltration of the perovskite into the mesoporous layer is required to suppress charge recombination [10, 143], and a perovskite capping layer is generally employed to guarantee a full coverage of the mesoporous layer. The thickness and structure of the mesoporous layer both impact on the performance of the MPSC; decreasing the thickness of the mp-TiO2 layer can help the pore-filling process and thus result in improved efficiency [10]. One advantage of a mesoporous layer is to better suppress hysteresis [144], and we present more details on hysteresis later in this review. A number of structures for the mesoporous scaffold have been explored, including TiO2 nanowires [145], TiO2 nanotubes [145], TiO2 nanorods [146], TiO2 nanofibers [142], TiO2 nanosheets [137, 147] and TiO2 nanoparticles [148, 149] with the different structures having been shown to have a different influence on the efficiency of the final device [150]. For example, it was found that 600 nm mp-TiO2 nanowires could be used to create devices that performed better than those

6.  Challenges for PSCs Although significant progress has been made over the past five years in the development of perovskite solar cells, significant challenges still exist. These challenges include increasing the reproducibility of efficient devices, eliminating hysteresis during operation, reducing degradation on exposure to moisture, reducing poor adhesion between different layers, developing mechanically flexible perovskite solar cells, as well as reducing the amount of toxic (Pb-containing) material used in devices. We discuss such issues in the next section and outline current research that starts to address the challenges raised. 15

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 21.  Surface SEM images of two-step casting MAPbI3 with different MAI concentration of (a-1) 41.94, (b-1) 52.42, and (c-1) 62.91 mM, leading to crystallites having an average dimension of 440, 170, and 130 nm, respectively. J  −  V curves measured on a forward scan (solid line) and reverse scan (dashed line) for MAPbI3 with size of (a-2) 440, (b-2) 170, and (c-2) 130 nm. Cross-sectional SEM images with (d-1) 220 and (e-1) 110 nm thick mp-TiO2 and (f-1) without mp-TiO2. J  −  V curves measured for devices with (d-2) 220 and (e-2) 110 thick mp-TiO2 and (f-2) without mp-TiO2. The voltage settling time was 200 ms and light intensity was AM 1.5 simulated sunlight. Figure reprinted with permission from [144]. Copyright 2014 American Chemical Society.

has been found to passivate charge-traps and therefore reduce device hysteresis. The presence of a mp-TiO2 layer has also been found to reduce hysteresis in MPSCs [158]. This is shown in figure 21, where it can be seen that thicker mp-TiO2 layers and larger grain size are correlated with reduced hysteresis. For example, a 3D mp-TiO2 nanowire architecture was found to inhibit hysteresis due to better charge carrier conduction than 2D mpTiO2 [145]. The testing conditions has also been reported to affect hysteresis, with humidity [167] being shown to have a negative effect on hysteresis. Encouragingly, highly efficient PPSCs (MAPbI3) PSCs have been shown to have a discrepancy of 0.1% (absolute) (forward scan PCE  =  18.10%, reverse scan PCE  =  18.2%) [50] and 0.45% (absolute) (forward scan PCE   =  14.30%, reverse scan PCE  =  14.75%) [168] as a result of balanced electron flux and hole current and reduced surface traps. Unger and Hoke et al [161] carried out a series of studies to investigate the origin and contributory factors of hysteresis, and suggested methods to reduce it. It was suggested that hysteresis results from slow transient effects that relate to the perovskite material, its crystallinity, its crystalline domain size and the degree of interfacial contact. Experimental conditions that can influence hysteresis include the delay time of testing after device fabrication and light and voltage conditions prior to measurement. Both fast and slow scan rates can be hysteresis-free as long as the device is tested at a quasi-steadystate condition. To quantify the degree of hysteresis, the J–V hysteresis index (HI) [164] was introduced and defined as

6.1.  Hysteresis in J–V test

Operational efficiency is the most important characteristic of a solar cell device. It has been widely reported that efficiency can vary significantly depending upon the scan direction (from reverse to forward bias or vice-versa) and scan rate. This phenomenon is termed as hysteresis and hinders the accurate determination of a steady-state device efficiency, as is relevant in real world scenarios. A high efficiency of 20.3% has been reported, however this was only obtained via reverse bias scan [12]. Usually, the PCE tested on a forward bias scan is closer to its steady-state value [158]. Some perovskite solar cells have been reported to be hysteresis-free or close to being hysteresis-free. For example, Lin et al reported hysteresis-free perovskite solar cells when scanning at 0.05 V s−1, 0.1 V s−1 and 0.2 V s−1 [159]. The origin behind device hysteresis is only now starting to be understood. A number of explanations were first advanced to explain observed hysteresis including a defect-trapping effect [158], a ferroelectric effect [160], capacitive effects [144, 158], transient effects [161], and ionic motion [162, 163]. It is generally agreed that the scan rate has a direct effect on the degree of the hysteresis. It was reported by Snaith, et al that reducing the scan rate increases rather than reduces the degree of hysteresis (scan rate from 0.3 V s−1 to 0.044 V s−1) [158]. However according to a study by Sanchez et al hysteresis is more severe at high scan rates for MAPbI3 devices fabricated using either a one-step or sequential casting, as well as FAPbI3 devices fabricated by sequential casting [164]. It has also been reported that increasing the size of a perovskite crystal and adding a perovskite capping layer [143] can reduce hysteresis. In contrast, small perovskite crystals and planar structures have been proposed to store charge and result in more significant hysteresis [12, 144]. It has been shown that PCBM can be introduced into the grain boundaries within a perovskite film through the use of a perovskite-PCBM mixed solution [165] or by thermal-driven PCBM diffusion into the perovskite through a bilayer structure [166]. This technique

J (V / 2) − J (V / 2)

HI = rs oc J (V / fs2) oc , where Jfs(Voc/2) and Jrs(Voc/2) are fs oc photocurrents at half Voc for forward and reverse scans respectively. When the discrepancy between scans increases, the HI parameter tends towards unity. It was suggested that device efficiency should be tested at a stabilized power output near the maximum power point to ensure accuracy [158]. It was also recommended that hysteresis should be reported with both J–V [158] and EQE [161] measurements. 16

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 22.  PCE variation of MAPb(I1−xBrx)3 (x  =  0, 0.06, 0.20, 0.29) stored in air at room temperature without encapsulation. The cells

was maintained at RH  =  35%, and then exposed to a humidity of 55% for one day on the fourth day. Figure reprinted with permission from [49]. Copyright 2013 American Chemical Society.

Snaith et al first proposed that ionic motion was the likely mechanism that caused hysteresis [162, 169]. Key to this proposal is the excess iodide and methylammonium ions that may be present in a perovskite film. Upon the application of an external forward bias, such ionic species can migrate towards the device electrodes and oppose the electric field in regions close to the electrodes (positive ions migrate to the anode while negative ions migrate to the cathode). When the external field is removed, the displaced ions enable the stabilization of a high density of space charge near each of the contacts when charges are photogenerated. However such mobile ions will eventually diffuse away from the electrodes and will once more assume a homogenous distribution throughout the film, and thus the device will no longer benefit from the temporary polarization near the electrodes. This mechanism indicates that device efficiency is often over-estimated by pre-biasing before measurement, and that actual power output can only be determined by applying a steady-state bias. Very recently, by modeling the hysteresis using a numerical drift-diffusion model, Snaith et al showed that the reduction of either mobile ionic species or interfacial charge traps can remove the hysteresis in a perovskite solar cell [170]. It is clear however that further research is required to gain deeper insight into the process and mechanisms underpinning PSC hysteresis.

The reported decomposition temperature for perovskite CH3NH3PbI3 has been reported as being between ~100 °C [173] and 140 °C [174] and also at a much higher temperature of ~300oC54. An additional concern for solar cells involving a titania scaffold is that TiO2 is sensitive to sunlight and can be unstable under illumination. For example, under UV light, mp-TiO2-based perovskite solar cells exhibit serious light instability with the efficiency of encapsulated mp-TiO2-based solar cells degrading to approximately 85% of their original efficiency after 4 h UV illumination [175]. However, mp-TiO2 free solar cells (based on mp-Al2O3) exhibited quite stable performance even over 1000 h of continuous illumination at one sun [175]. Perovskite solar cells based on Al-doped TiO2 showed enhanced efficiency and light stability as doped Al in the TiO2 lattice is believed to permanently passivate electronic trapping sites [176]. Recent work lead by Haque et al shows that the electrons that are generated in the MAPbI3 perovskite layer upon light exposure can react with free oxygen in air to form highly reactive superoxide O− 2 [177]. These superoxides attack the perovskite layer by reaction with the methylamonium moiety of the perovskite and break down the device. The by-product of this reaction is water which further speeds up the degradation. This indicates that the degradation can be minimized by rapidly extracting the electrons after their generation to prevent the formation of superoxide, or by replacing the MA comp­ onent with species without acid protons that are involved in the reaction. The degradation in the mesoporous TiO2 based system is much less compared with mp-Al2O3 or planar systems, highlighting an advantage to retaining the mesoporous superstructure in the design of perovskite solar cells [178]. The origin of moisture instability results partially from the hygroscopic nature of amine salts [179]. Both MAPbI3 and

6.2. Stability

Another significant challenge in perovskite research is the observed (sometimes rapid) degradation that occurs when devices are exposed to thermal-stress, light, oxygen or moisture [171, 172]. The degradation that occurs in MAPbI3 as a result of thermal stress is summarized by the following expression: CH3NH3PbI3  →  PbI2  +  CH3NH2↑  +HI↑ [107]. 17

Review

Rep. Prog. Phys. 79 (2016) 026501

comparable devices fabricated by spin coating in air, and retained 95% of their original PCE after 10 d storage in air without sealing [188]. Layer-by-layer deposition on a flexible PET substrate has also been used to increase air-stability [189]. Mei et al [33] capped a (5-AVA)x(MA)1−xPbI3 perovskite layer with a 10 μm thick layer of carbon (mp-TiO2/ ZrO2/C) that acted as a barrier to moisture. The HTM-free perovskite devices had a PCE of 12.8% and demonstrated high stability, retaining almost 100% of their original efficiency after 1008 h in ambient air under full AM 1.5 simulated sunlight (see figure 23). The stability of perovskite devices are often characterized by their T80 time—the time required for the efficiency of an unencapsulated solar cell to decay to 80% of its original efficiency [182, 190]. A T80 of around 100 h has been achieved at T  =  23 °C and RH  =  50% by Yu et al [182] for the device ITO/PEDOT:PSS/MAPbI3/PC61BM/TiO2/Al. Although the stability of PSCs under exposure to moisture has improved significantly, they are still far away for commercialization. Furthermore, the mechanisms by which water degrades the perovskite layer is not completely clear [191].

MAPbI3−xClx undergo a similar moisture-assisted degradation process in which methylamine group is lost via sublimation and PbI2 is formed [173]. The formation of PbI2 is evidenced through the discoloration of MAPbI3 after being heated ~30 min at 150 °C in air. FAPbI3 however has better thermal stability as no discoloration was observed when heated for ~60 min at 150 °C in air, although when exposed to moist air for 15 min (RH  =  100%), FAPbI3 degrades at a similar rate to that of MAPbI3 [59]. A number of approaches have been explored to reduce the moisture-sensitivity of perovskites, including the use of mixed-halide perovskites, the formation of crystal crosslinking in perovskite with alkylphosphonic acid ω-ammonium chlorides [180], the adoption of protective transporting layers or electrodes and the use of new deposition methods. It has been found that perovksites having different compositions vary in their sensitivity to moisture. For example, it was found that a MAPb(I1−xBrx)3 solar cell underwent negligible reduction in efficiency under ambient conditions (relative humidity (RH) less than 50%) [49]. However when the humidity was increased above 50%, the efficiency reduces dramatically when x is less than 0.2. It is found that the cubic phase MAPb(I1−xBrx)3 (x  ⩾  0.2) is much more stable than the tetragonal (pseudo-cubic) phase (x  1 cm2 and maintained more than 90% of its initial efficiency after 1000 h light soaking [187]. Unencapsulated MAPbI3 solar cells incorporating an NiOx layer doped with 5% Cu and have also been shown to maintain 90% of their original efficiency after storage for 240 h in air; a stability far greater than that obtained using a hydrophilic PEDOT:PSS layer [76]. MAPbI3 devices incorporating the hydrophobic conjugated polymer poly[2,5bis(2-decyldodecyl)pyrrolo[3,4-c]pyrrole-1,4(2H,5H)dione-(E)-1,2-di(2,2′-bithiophen-5-yl)ethene] (PDPPDBTE) showed excellent moisture stability and were able to retain 91.3% of their original efficiency after 1000 h storage in ambient air (room temperature, RH  =  20% without sealing) [179]. To facilitate PSC manufacture, it is advantageous if devices could be manufactured in air, necessitating perovskite stability even when exposed to, or fabricated under conditions of high humidity. Using a preheated sequential deposition technique, perovskite solar cells have been fabricated in air at RH  =  50% with a PCE of 15.76% achieved [143]. Perovskite films have been deposited by blade-coating in both air and N2. It was found that the blade-coated MAPbI3 device fabricated in air exhibited enhanced air stability than

6.3.  PSCs on flexible substrates

Flexible PSCs have been developed using either polymer substrates (e.g. PET [192, 193], PEN [31] or polyethyleneimine (PEI) [194]), metal foils [195, 196] and fiber substrates [197]. We note that flexible PSCs fabricated on polymer substrates offer the prospect of creating devices for new applications, such as bendable or wearable power sources [31] (see figure  24(a)). The casting of films on flexible polymer substrates can be made using spin-coating, or more preferably by spray coating, inkjet printing or slot-die coating; processes that can all be scaled up for mass production [198]. However as polymer films can easily undergo deformation when heated above their glass transition temperature (Tg), this defines a maximum temperature for thin-film processing. For this reason, structures that require the sintering of TiO2 at temperatures around 500 °C are incompatible with polymer substrates. So far, devices have been fabricated on PET using a self-organized HTL composed of PEDOT:PSS and tetrafluoroethylene-perfluoro-3,6-dioxa-4-methyl-7-octenesulfonic acid co-polymer, and achieved an efficiency of 8.0% (Jsc  =  15.5mA cm−2, Voc  =  1.04 V, FF  =  49.9%) compared to comparable devices made on glass that had an efficiency of 11.7% (Jsc  =  16.7mA cm−2, Voc  =  0.982 V, FF  =  70.5%) [199]. It can be seen that the efficiency of devices on the PET substrate were limited by a much lower FF, which usually results from inefficient charge transport and collection. Importantly, flexible PSCs fabricated on metal foil substrates can withstand the high temperature calcination process used to fabricate TiO2. For example, PSCs fabricated on titanium foil (Ti foil/c-TiO2/MAPbI3−xClx:mp-Al2O3/Spiro-OMeTAD/ PEDOT:PS/transparent conductive adhesive/PET embedded with Ni grid) reached a PCE of 10.3% [195]. A second key challenge for flexible PSCs is the reduction of structural damage after repeated bending cycles. Recent work in which PSCs based on a PET substrate deposited 18

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 23.  Stability test of an unsealed (5-AVA)x(MA)1−xPbI3 perovskite solar cells in ambient air over 1008 h, the perovskite being

protected by a carbon layer. Figure reprinted with permission from [33]. Copyright 2014 American Association for the Advancement of Science.

organohalide lead perovskite solar cells. Chen and Qi et al [202] reported PSCs using lead from recycled car batteries and reached comparable efficiency of 9.37% compared with a PCE of 9.73% by using PbI2 purchased from Sigma-Aldrich. A single lead-acid car battery has been estimated to supply enough lead material for ~710 m2 of PSCs. Nevertheless, if high-efficiency lead-free PSCs are to be manufactured at scale, it would be desirable to eliminate the use of toxic Pb-containing materials. For this reason lead-free perovskites (e.g. MASnY3 and CsSnY3) [203] have been explored as PSC materials. In general, the development of lead-free PSCs is still at an early stage and thus the efficiency and stability of these materials when fabricated into a device are not comparable with more established lead-containing materials. The first lead-free PSCs were explored by Hao et al [204] and had a structure FTO/blocking TiO2 layer/MASnI3−xBrx  +  mpTiO2/spiro-OMeTAD/Au, and an efficiency of 5.73%. In further work, PSC efficiency was then increased to 6.4% using MASnI3 [205]. Unfortunately MASnI3 has a short charge carrier lifetime (~200 ps [205]) which has been partially ascribed to spontaneous hole-doping during crystallization [206]. More problematic however is the fact that Sn-based perovskites have significantly increased sensitivity to oxygen, with Sn2+ being easily oxidized to Sn4+ when exposed to oxygen and moisture. The presence of excessive concentrations of Sn4+

using an argon-assisted one-step solution casting method had an efficiency of 12.3% and maintained 80% of the original PCE after being bent 500 times at frequency of 20 rpm around a curvature radius (R) of 15.8 mm [200]. Kim et al [31] also reported the fabrication of flexible PSCs on a PEN substrate (PEN/ITO/ALD-TiO2 layer/MASnI3−xClx/spiro-OMeTAD/ Ag) and obtained an efficiency of 12.2% and maintained almost 100%, 95%, and 50% of their original efficiency after being bent 1000 times around a radius of 400, 10 and 4 mm respectively (see figure  24(b)). Repeated bending results in cracks in the film which cause structural damage, result in large increases in electrical resistance and consequently reduce device efficiency (see figure  25). The reason for this degradation originates from the brittleness of ITO [140] and the different Young’s modulus of ITO and flexible PEN [31]. Indeed devices bent around a smaller radius undergo greater reductions in efficiency [140] with the bending radius having a larger impact on degradation than the number of applied bending cycles. The interested reader is directed to a recent review regarding the fabrication of flexible PSCs [201]. 6.4.  Lead-free PSCs

The toxicity of lead on humans and the environment has been a challenge for the development and future applications of 19

Review

Rep. Prog. Phys. 79 (2016) 026501

Figure 24.  (a) PCE measured after bending from radius 400 mm to 1 mm. The inset shows images of a device attached to a person’s neck,

wrist and finger, corresponding to bending radii 400 nm (R400), 10 nm (R10) and 4 nm (R4), respectively. (b) PCE measured at R400, R10 and R4 mm. The inset shows images taken during the bending tests. Figure reprinted with permission from [31]. Copyright 2014 The Royal Society of Chemistry.

Figure 25. (a) In situ resistance change (ΔR/R0(%)) of PEN/ITO/TiOx/perovskite and PEN/TiOx/perovskite, respectively. The lower panel

shows a magnified plot of the upper panel (yellow squared region). (b-1) and (c-1) SEM images of PEN/ITO/TiOx/perovskite and PEN/ TiOx/perovskite after 300 bending cycles. Scale bar: 100 mm. (b-2), (b-3), (c-2), and (c-3) SEM images corresponding to the green and red regions in (b-1) and (c-1). Scale bars: 5 mm. Figure reprinted with permission from [31]. Copyright 2014 The Royal Society of Chemistry.

20

Review

Rep. Prog. Phys. 79 (2016) 026501

Acknowledgments

in the perovskite may then result in reduced carrier mobility and enhanced recombination [204]. Indeed, MASnI3 degrades completely within minutes in air [205]. The perovskite CsSnX3 has also been explored for application in PSCs, however its efficiency is lower than that of MASnI3. To address this, it has been doped with SnF2 to suppress defects and increase Jsc and Voc. At a 20% SnF2 doping level, Kumar et al [207] created a champion CsSnI3 PSCs having a PCE of 2.02%, Jsc  =  22.7 mA cm−2, Voc  =  0.24 V, and FF  =  37%. This should be compared with undoped CsSnI3 based PSCs in which PCE was 3  ×  10−4%, with Jsc  =  0.19 mA cm−2, Voc  =  0.01 V and FF  =  21%. Again, the Sn2+ component is unstable and thus the CsSnI3 phase coexists with Cs2SnF6 and SnI4. Recently, devices based on a CsSnI3−xBrxSnF2 perovskite were fabricated and had a Voc of 0.411 V and a FF of 58%, although the Jsc was reduced to 3.99 mA cm−2, leading to a PCE of 0.95% [208]. It is clear that although the lead-free PSCs have an efficiency that is much lower than those based on lead-containing compounds, the development of environmentally friendly materials is an important area of research.

T.W. acknowledges funding support from the Recruitment Program of Global Experts (1000 Talents Plan) of China, the National Natural Science Foundation of China (Grant No. 21504065), and the Fundamental Research Funds for the Central Universities (WUT: 2015III018 and 2015III029) of China. D.G.L. gratefully thank the UK EPSRC for funding this research through research grants EP/J017361/1 and EP/ M014797/1 (SUPERGEN Supersolar Solar Energy Hub) and EP/I028641/1 (Polymer/fullerene photovoltaic devices: new materials and innovative processes for high volume manufacture). A.B. also thanks the EPSRC for a studentship via the E-Futures Doctoral Training Centre in Interdisciplinary Energy Research EP/G037477/1. References [1] Yin W J, Shi T and Yan Y 2014 Unique properties of halide perovskites as possible origins of the superior solar cell performance Adv. Mater. 26 4653–8 [2] Snaith H J 2013 Perovskites: the emergence of a new era for low-cost, high-efficiency solar cells J. Phys. Chem. Lett. 4 3623–30 [3] Kojima A et al 2006 Novel photoelectrochemical cell with mesoscopic electrodes sensitized by lead-halide compounds (2) 210th ECS Meeting Abstracts [4] Kojima. A et al 2009 Organometal halide perovskites as visible-light sensitizers for photovoltaic cells J. Am. Chem. Soc. 131 6050–1 [5] Best Research-Cell Efficiencies www.nrel.gov/ncpv/images/ efficiency_chart.jpg (last accessed on Nov. 10th 2015) [6] Kim H-S et al 2012 Lead iodide perovskite sensitized allsolid-state submicron thin film mesoscopic solar cell with efficiency exceeding 9% Sci. Rep. 2 591 [7] Lee M M et al 2012 Efficient hybrid solar cells based on meso-superstructured organometal halide perovskites Science 338 643–7 [8] Ball J M et al 2013 Low-temperature processed mesosuperstructured to thin-film perovskite solar cells Energy Environ. Sci. 6 1739–43 [9] Guarnera S et al 2015 Improving the long-term stability of perovskite solar cells with a porous Al2O3 buffer layer J. Phys. Chem. Lett. 6 432–7 [10] Leijtens T et al 2014 The importance of perovskite pore filling in organometal mixed halide sensitized TiO2-based solar cells J. Phys. Chem. Lett. 5 1096–102 [11] Liu M, Johnston M B and Snaith H J 2013 Efficient planar heterojunction perovskite solar cells by vapour deposition Nature 501 395–8 [12] Jeon N J et al 2015 Compositional engineering of perovskite materials for high-performance solar cells Nature 517 476–80 [13] Zhou H et al 2014 Photovoltaics: Interface engineering of highly efficient perovskite solar cells Science 345 542–6 [14] Jung H S and Park N G 2015 Perovskite solar cells: from materials to devices Small 11 10–25 [15] Swetha T and Singh S P 2015 Perovskite solar cells based on small molecules hole transporting materials J. Mater. Chem. A 3 18329–44 [16] Liu H et al 2016 Nano-structured electron transporting materials for perovskite solar cells Nanoscale 10.1039/c5nr05207f [17] Jeon N J et al 2013 Efficient inorganic-organic hybrid perovskite solar cells based on pyrene arylamine

7. Outlook We have reviewed the most recent research progress and have discussed a number of the existing challenges for perovskite solar cells. Such materials appear to hold considerable promise for photovoltaic applications, due to their large carrier diffusion lengths, high charge-carrier mobility, and low exciton binding energy. At the time of writing (2015) the state-ofthe-art PSCs have a PCE in excess of 20%. In this review, we have shown that the nano- and meso-scale morphology of the photoactive layer prepared from solution depends on the details of the preparation process and plays a significant role in determining the eventual PCE of the device. It is expected that perovskite photoactive layers containing large crystalline grains that achieve full surface-coverage, together with careful control of structural and electronic properties of interfaces within the device will help further increase the efficiency of PSCs. The low exciton binding energy and ambipolar charge transport properties of perovskite films make them suitable materials from which to prepare tandem cells using other PV-applicable semiconductors, including organic photovoltaics (OPVs), silicon, copper indium gallium diselenide (CIGS), copper indium diselenide (CIS) or copper zinc tinsulfide/selenide (CZTSSe) [2, 27, 209]. It is expected that such a combination of materials will enable the creation of low cost and high efficiency (>25%) solar cell devices and products [210]. Research on PSCs also faces challenges, including hysteresis in device operation, sensitivity to moisture, a lack of mechanical flexibility, low operational lifetimes and the toxicity of lead. Determining the mechanisms behind the degradation of such devices is a rich area for study. Clearly agreed standards need to be introduced to specify the testing devices to include the device hysteresis, mechanical stability, stability under thermal stress and exposure to light, oxygen and moisture to ensure the continued rapid and consistent progress of the field. 21

Review

Rep. Prog. Phys. 79 (2016) 026501

[41] Hwang S H et al 2014 Size-controlled SiO2 nanoparticles as scaffold layers in thin-film perovskite solar cells J. Mater. Chem. A 2 16429–33 [42] Wang Q-K et al 2015 Energy level offsets at lead halide perovskite/organic hybrid interfaces and their impacts on charge separation Adv. Mater. Interfaces 2 1400528 [43] Qiu J et al 2013 All-solid-state hybrid solar cells based on a new organometal halide perovskite sensitizer and 1D TiO2 nanowire arrays Nanoscale 5 3245–8 [44] Zhang M et al 2014 Composition-dependent photoluminescence intensity and prolonged recombination lifetime of perovskite CH3NH3PbBr(3−x)Cl(x) films Chem. Commun. 50 11727–30 [45] Lv S et al 2014 One-step, solution-processed formamidinium lead trihalide (FAPbI(3−x)Cl(x)) for mesoscopic perovskitepolymer solar cells Phys. Chem. Chem. Phys. 16 19206–11 [46] Hu M et al 2014 Efficient hole-conductor-free, fully printable mesoscopic perovskite solar cells with a broad light harvester NH2CH=NH2PbI3 J. Mater. Chem. A 2 17115–21 [47] Pellet N et al 2014 Mixed-organic-cation perovskite photovoltaics for enhanced solar-light harvesting Angew. Chem. Int. Ed. 53 3151–7 [48] Choi H et al 2014 Cesium-doped methylammonium lead iodide perovskite light absorber for hybrid solar cells Nano Energy 7 80–5 [49] Noh J H et al 2013 Chemical management for colorful, efficient, and stable inorganic-organic hybrid nanostructured solar cells Nano Lett. 13 1764–9 [50] Heo J H et al 2015 Hysteresis-less inverted CH3NH3PbI3 planar perovskite hybrid solar cells with 18.1% power conversion efficiency Energy Environ. Sci. 8 1602–8 [51] Im J-H et al 2011 6.5% efficient perovskite quantumdot-sensitized solar cell Nanoscale 3 4088–93 [52] Burschka J et al 2013 Sequential deposition as a route to high-performance perovskite-sensitized solar cells Nature 499 316–9 [53] Grätzel M 2014 The light and shade of perovskite solar cells Nat. Mater. 13 838–42 [54] Williams S T et al 2014 Role of chloride in the morphological evolution of organo-lead halide perovskite thin films ACS Nano 8 10640–54 [55] Williams A E et al 2014 Perovskite processing for photovoltaics: a spectro-thermal evaluation J. Mater. Chem. A 2 19338–46 [56] Unger E L et al 2014 Chloride in lead chloride-derived organo-metal halides for perovskite-absorber solar cells Chem. Mater. 26 7158–65 [57] Colella S et al 2014 Elusive presence of chloride in mixed halide perovskite solar cells J. Phys. Chem. Lett. 5 3532–8 [58] Li Y et al 2015 Fabrication of planar heterojunction perovskite solar cells by controlled low-pressure vapor annealing J. Phys. Chem. Lett. 6 493–9 [59] Eperon G E et al 2014 Formamidinium lead trihalide: a broadly tunable perovskite for efficient planar heterojunction solar cells Energy Environ. Sci. 7 982–8 [60] Lee J-W et al 2014 High-efficiency perovskite solar cells based on the black polymorph of HC(NH2)2PbI3 Adv. Mater. 26 4991–8 [61] Wang F et al 2015 HPbI3: a new precursor compound for highly efficient solution-processed perovskite solar cells Adv. Funct. Mater. 25 1120–6 [62] Bai S et al 2014 High-performance planar heterojunction perovskite solar cells: Preserving long charge carrier diffusion lengths and interfacial engineering Nano Res. 7 1749–58 [63] Stranks S D et al 2013 Electron-hole diffusion lengths exceeding 1 micrometer in an organometal trihalide perovskite absorber Science 342 341–4

derivatives as hole-transporting materials J. Am. Chem. Soc. 135 19087–90 [18] Sum T C and Mathews N 2014 Advancements in perovskite solar cells: photophysics behind the photovoltaics Energy Environ. Sci. 7 2518–34 [19] Marchioro A et al 2014 Unravelling the mechanism of photoinduced charge transfer processes in lead iodide perovskite solar cells Nat. Photonics 8 250–5 [20] Zhang C et al 2015 Magnetic field effects in hybrid perovskite devices Nat. Phys. 11 427–34 [21] Sheng C et al 2015 Exciton versus free carrier photogeneration in organometal trihalide perovskites probed by broadband ultrafast polarization memory dynamics Phys. Rev. Lett. 114 116601 [22] Hsiao Y-C et al 2015 Magneto-optical studies on spindependent charge recombination and dissociation in perovskite solar cells Adv. Mater. 27 2899–906 [23] Saba M et al 2014 Correlated electron-hole plasma in organometal perovskites Nat. Commun. 5 5049 [24] Hsiao Y-C et al 2015 Fundamental physics behind highefficiency organo-metal halide perovskite solar cells J. Mater. Chem. A 3 15372–85 [25] Hu M et al 2015 Distinct exciton dissociation behavior of organolead trihalide perovskite and excitonic semiconductors studied in the same system Small 11 2164–9 [26] Frost J M et al 2014 Atomistic origins of high-performance in hybrid halide perovskite solar cells Nano Lett. 14 2584–90 [27] Liu Y et al 2015 Integrated perovskite/bulk-heterojunction toward efficient solar cells Nano Lett. 15 662–8 [28] You J et al 2014 Low-Temperature Solution-Processed Perovskite Solar Cells with High Efficiency and Flexibility ACS Nano 8 1674–80 [29] Docampo P et al 2013 Efficient organometal trihalide perovskite planar-heterojunction solar cells on flexible polymer substrates Nat. Commun. 4 2761 [30] Ameen S et al 2015 Exclusion of metal oxide by an RF sputtered Ti layer in flexible perovskite solar cells: energetic interface between a Ti layer and an organic charge transporting layer Dalton Trans. 44 6439–48 [31] Kim B J et al 2015 Highly efficient and bending durable perovskite solar cells: toward a wearable power source Energy Environ. Sci. 8 916–21 [32] Batmunkh M et al 2015 Nanocarbons for mesoscopic perovskite solar cells J. Mater. Chem. A 3 9020–31 [33] Mei A et al 2014 A hole-conductor–free, fully printable mesoscopic perovskite solar cell with high stability Science 345 295–8 [34] Habisreutinger S N et al 2014 Carbon nanotube/polymer composites as a highly stable hole collection layer in perovskite solar cells Nano Lett. 14 5561–8 [35] Ku Z et al 2013 Full printable processed mesoscopic CH3NH3PbI3/TiO2 heterojunction solar cells with carbon counter electrode Sci. Rep. 3 3132 [36] Gao P, Grätzel M and Nazeeruddin M K 2014 Organohalide lead perovskites for photovoltaic applications Energy Environ. Sci. 7 2448–63 [37] Boix P P et al 2015 Perovskite solar cells: beyond methylammonium lead iodide J. Phys. Chem. Lett. 6 898–907 [38] Polander L E et al 2014 Hole-transport material variation in fully vacuum deposited perovskite solar cells APL Mat. 2 081503 [39] Wu Z et al 2014 Efficient planar heterojunction perovskite solar cells employing graphene oxide as hole conductor Nanoscale 6 10505–10 [40] Docampo P et al 2014 Influence of the orientation of methylammonium lead iodide perovskite crystals on solar cell performance APL Mat. 2 081508 22

Review

Rep. Prog. Phys. 79 (2016) 026501

[85] Yantara N et al 2015 Unravelling the effects of Cl addition in single step CH3NH3PbI3 perovskite solar cells Chem. Mater. 27 2309–14 [86] Yu H et al 2014 The role of chlorine in the formation process of CH3NH3PbI3−xClx Perovskite Adv. Funct. Mater. 24 7102–8 [87] Zhang W et al 2015 Ultrasmooth organic-inorganic perovskite thin-film formation and crystallization for efficient planar heterojunction solar cells Nat. Commun. 6 6142 [88] Grancini G et al 2014 The impact of the crystallization processes on the structural and optical properties of hybrid perovskite films for photovoltaics J. Phys. Chem. Lett. 5 3836–42 [89] Carnie M J et al 2014 Performance enhancement of solution processed perovskite solar cells incorporating functionalized silica nanoparticles J. Mater. Chem. A 2 17077–84 [90] Chen Q et al 2014 Planar heterojunction perovskite solar cells via vapor-assisted solution process J Am Chem Soc 136 622–5 [91] Aharon S, Cohen B E and Etgar L 2014 Hybrid Lead halide iodide and lead halide bromide in efficient hole conductor free perovskite solar cell J. Phys. Chem. C 118 17160–5 [92] Xiao Z et al 2014 Efficient, high yield perovskite photovoltaic devices grown by interdiffusion of solutionprocessed precursor stacking layers Energy Environ. Sci. 7 2619–23 [93] Li G et al 2015 Identifying the optimum morphology in highperformance perovskite solar cells Adv. Energy Mater. 5 1401775 [94] Barrows A T et al 2014 Efficient planar heterojunction mixed-halide perovskite solar cells deposited via spraydeposition Energy Environ. Sci. 7 2944–50 [95] Deng Y et al 2015 Scalable fabrication of efficient organolead trihalide perovskite solar cells with doctorbladed active layers Energy Environ. Sci. 8 1544–50 [96] Hwang K et al 2015 Toward large scale roll-to-roll production of fully printed perovskite solar cells Adv. Mater. 27 1241–7 [97] Schmidt T M et al 2015 Upscaling of perovskite solar cells: fully ambient roll processing of flexible perovskite solar cells with printed back electrodes Adv. Energy Mater. 5 1500569 [98] Hao F et al 2014 Controllable perovskite crystallization at a gas-solid interface for hole conductor-free solar cells with steady power conversion efficiency over 10% J. Am. Chem. Soc. 136 16411–9 [99] Jeon N J et al 2014 Solvent engineering for highperformance inorganic-organic hybrid perovskite solar cells Nat. Mater. 13 897–903 [100] Xiao M et al 2014 A fast deposition-crystallization procedure for highly efficient lead iodide perovskite thin-film solar cells Angew. Chem. Int. Ed. 126 10056–61 [101] Kim H-B et al 2014 Mixed solvents for the optimization of morphology in solution-processed, inverted-type perovskite/fullerene hybrid solar cells Nanoscale 6 6679–83 [102] Seo J et al 2014 Benefits of very thin PCBM and LiF layers for solution-processed p-i-n perovskite solar cells Energy Environ. Sci. 7 2642–6 [103] Wu Y et al 2014 Retarding the crystallization of PbI2 for highly reproducible planar-structured perovskite solar cells via sequential deposition Energy Environ. Sci. 7 2934–8 [104] Wu C-G et al 2015 High efficiency stable inverted perovskite solar cells without current hysteresis Energy Environ. Sci. 8 2725–33 [105] Li G, Zhang T and Zhao Y 2015 Hydrochloric acid accelerated formation of planar CH3NH3PbI3 perovskite with high humidity tolerance J. Mater. Chem. A 3 19674–8

[64] Zhao Y, Nardes A M and Zhu K 2014 Solid-state mesostructured perovskite ch3nh3pbi3 solar cells: charge transport, recombination, and diffusion length J. Phys. Chem. Lett. 5 490–4 [65] Docampo P et al 2014 Solution deposition-conversion for planar heterojunction mixed halide perovskite solar cells Adv. Energy Mater. 4 1400355 [66] Wehrenfennig C et al 2014 Charge-carrier dynamics in vapour-deposited films of the organolead halide perovskite CH3NH3PbI3−xClx Energy Environ. Sci. 7 2269–75 [67] Dong Q et al 2015 Electron-hole diffusion lengths  >175 μm in solution-grown CH3NH3PbI3 single crystals Science 347 967–70 [68] Liu C et al 2015 High performance planar heterojunction perovskite solar cells with fullerene derivatives as the electron transport layer ACS Appl. Mater. Interfaces 7 1153–9 [69] Edri E et al 2014 Chloride inclusion and hole transport material doping to improve methyl ammonium lead bromide perovskite-based high open-circuit voltage solar cells J. Phys. Chem. Lett. 5 429–33 [70] Wang K et al 2015 Bulk heterojunction perovskite hybrid solar cells with large fill factor Energy Environ. Sci. 8 1245–55 [71] Wang Q et al 2014 Large fill-factor bilayer iodine perovskite solar cells fabricated by a low-temperature solution-process Energy Environ. Sci. 7 2359–65 [72] Bi C et al 2015 Low-temperature fabrication of efficient widebandgap organolead trihalide perovskite solar cells Adv. Energy Mater. 5 1401616 [73] Ryu S et al 2014 Voltage output of efficient perovskite solar cells with high open-circuit voltage and fill factor Energy Environ. Sci. 7 2614–8 [74] Tress W et al 2015 Predicting the open-circuit voltage of CH3NH3PbI3 perovskite solar cells using electroluminescence and photovoltaic quantum efficiency spectra: the role of radiative and non-radiative recombination Adv. Energy Mater. 5 1400812 [75] Edri E et al 2013 High Open-circuit voltage solar cells based on organic-inorganic lead bromide perovskite J. Phys. Chem. Lett. 4 897–902 [76] Kim J H et al 2015 High-performance and environmentally stable planar heterojunction perovskite solar cells based on a solution-processed copper-doped nickel oxide holetransporting layer Adv. Mater. 27 695–701 [77] Conings B et al 2014 Perovskite-based hybrid solar cells exceeding 10% efficiency with high reproducibility using a thin film sandwich approach Adv. Mater. 26 2041–6 [78] Eperon G E et al 2014 Morphological control for high performance, solution-processed planar heterojunction perovskite solar cells Adv. Funct. Mater. 24 151–7 [79] Wojciechowski K et al 2014 Heterojunction modification for highly efficient organic-inorganic perovskite solar cells ACS Nano 8 12701–9 [80] Liang P-W et al 2015 Roles of fullerene-based interlayers in enhancing the performance of organometal perovskite thinfilm solar cells Adv. Energy Mater. 5 1402321 [81] Hau S K et al 2010 Effect of chemical modification of fullerene-based self-assembled monolayers on the performance of inverted polymer solar cells ACS Appl. Mater. Interfaces 2 1892–902 [82] Saliba M et al 2014 Influence of thermal processing protocol upon the crystallization and photovoltaic performance of organic–inorganic lead trihalide perovskites J. Phys. Chem. C 118 17171–7 [83] Salim T et al 2015 Perovskite-based solar cells: impact of morphology and device architecture on device performance J. Mater. Chem. A 3 8943–69 [84] Song T-B et al 2015 Perovskite solar cells: film formation and properties J. Mater. Chem. A 3 9032–50 23

Review

Rep. Prog. Phys. 79 (2016) 026501

[106] Tan K W et al 2014 Thermally induced structural evolution and performance of mesoporous block copolymer-directed alumina perovskite solar cells ACS Nano 8 4730–9 [107] Dualeh A et al 2014 Effect of annealing temperature on film morphology of organic-inorganic hybrid pervoskite solidstate solar cells Adv. Funct. Mater. 24 3250–8 [108] Hsu H-L et al 2014 Two-step thermal annealing improves the morphology of spin-coated films for highly efficient perovskite hybrid photovoltaics Nanoscale 6 10281–8 [109] Troughton J et al 2015 Rapid processing of perovskite solar cells in under 2.5 s J. Mater. Chem. A 3 9123–7 [110] Zhou Z et al 2015 Methylamine-gas induced defect-healing behavior of CH3NH3PbI3 thin films for perovskite solar cells Angew. Chem. Int. Ed. 54 9705–9 [111] Xiao Z et al 2014 Solvent annealing of perovskite-induced crystal growth for photovoltaic-device efficiency enhancement Adv. Mater. 26 6503–9 [112] Lian J et al 2015 Organic solvent vapor sensitive methylammonium lead trihalide film formation for efficient hybrid perovskite solar cells J. Mater. Chem. A 3 9146–51 [113] Zhu W et al 2015 A facile, solvent vapor-fumigationinduced, self-repair recrystallization of CH3NH3PbI3 films for high-performance perovskite solar cells Nanoscale 7 5427–34 [114] Shi D et al 2015 Low trap-state density and long carrier diffusion in organolead trihalide perovskite single crystals Science 347 519–22 [115] Nie W et al 2015 High-efficiency solution-processed perovskite solar cells with millimeter-scale grains Science 347 522–5 [116] Chen C-W et al 2014 Efficient and uniform planar-type perovskite solar cells by simple sequential vacuum deposition Adv. Mater. 26 6647–52 [117] Fu Y et al 2015 Solution growth of single crystal methylammonium lead halide perovskite nanostructures for optoelectronic and photovoltaic applications J. Am. Chem. Soc. 137 5810–8 [118] Yang W S et al 2015 High-performance photovoltaic perovskite layers fabricated through intramolecular exchange Science 348 1234–7 [119] Yang M et al 2015 Square-centimeter solution-processed planar CH3NH3PbI3 perovskite solar cells with efficiency exceeding 15% Adv. Mater. 27 6363–70 [120] Zhou Y et al 2015 Room-temperature crystallization of hybrid-perovskite thin films via solvent–solvent extraction for high-performance solar cells J. Mater. Chem. A 3 8178–84 [121] Guo Y et al 2014 Enhancement in the efficiency of an organic–inorganic hybrid solar cell with a doped P3HT hole-transporting layer on a void-free perovskite active layer J. Mater. Chem. A 2 13827–30 [122] Bi D et al 2014 Improved morphology control using a modified two-step method for efficient perovskite solar cells ACS Appl. Mater. Interfaces 6 18751–7 [123] Shi J et al 2014 Hole-conductor-free perovskite organic lead iodide heterojunction thin-film solar cells: High efficiency and junction property Appl. Phys. Lett. 104 063901 [124] Yang Y et al 2014 An all-carbon counter electrode for highly efficient hole-conductor-free organo-metal perovskite solar cells RSC Adv. 4 52825–30 [125] Tsai K-W et al 2015 High-performance hole-transporting layer-free conventional perovskite/fullerene heterojunction thin-film solar cells J. Mater. Chem. A 3 9128–32 [126] Ke W et al 2015 Efficient hole-blocking layer-free planar halide perovskite thin-film solar cells Nat. Commun. 6 6700 [127] Edri E et al 2014 Why lead methylammonium tri-iodide perovskite-based solar cells require a mesoporous electron

transporting scaffold (but not necessarily a hole conductor) Nano Lett. 14 1000–4 [128] Yu Z and Sun L 2015 Recent progress on hole-transporting materials for emerging organometal halide perovskite solar cells Adv. Energy Mater. 5 1500213 [129] Zhang H et al 2014 Improved High-Efficiency perovskite planar heterojunction solar cells via incorporation of a polyelectrolyte interlayer Chem. Mater. 26 5190–3 [130] Chueh C-C, Li C-Z and Jen A K Y 2015 Recent progress and perspective in solution-processed Interfacial materials for efficient and stable polymer and organometal perovskite solar cells Energy Environ. Sci. 8 1160–89 [131] Shi J et al 2015 Interfaces in perovskite solar cells Small 11 2472–86 [132] Wang J T et al 2014 Low-temperature processed electron collection layers of graphene/TiO2 nanocomposites in thin film perovskite solar cells Nano Lett. 14 724–30 [133] Yella A et al 2014 Nanocrystalline rutile electron extraction layer enables low-temperature solution processed perovskite photovoltaics with 13.7% efficiency Nano Lett. 14 2591–6 [134] Wu Y et al 2014 Highly compact TiO2 layer for efficient hole-blocking in perovskite solar cells Appl. Phys. Express 7 052301 [135] Wang K-C et al 2014 Low-temperature sputtered nickel oxide compact thin film as effective electron blocking layer for mesoscopic NiO/CH3NH3PbI3 perovskite heterojunction solar cells ACS Appl. Mater. Interfaces 6 11851–8 [136] Xu X et al 2015 Hole selective NiO contact for efficient perovskite solar cells with carbon electrode Nano Lett. 15 2402–8 [137] Rong Y et al 2014 Hole-conductor-free mesoscopic TiO2/ CH3NH3PbI3 heterojunction solar cells based on anatase nanosheets and carbon counter electrodes J. Phys. Chem. Lett. 5 2160–4 [138] Liu Z et al 2015 p-Type mesoscopic NiO as an active interfacial layer for carbon counter electrode based perovskite solar cells Dalton Trans. 44 3967–73 [139] Cao K et al 2015 Efficient mesoscopic perovskite solar cells based on the CH3NH3PbI2Br light absorber J. Mater. Chem. A 3 9116–22 [140] Giacomo F D et al 2015 Flexible perovskite photovoltaic modules and solar cells based on atomic layer deposited compact layers and UV-irradiated TiO2 scaffolds on plastic substrates Adv. Energy Mater. 5 1401808 [141] Liang L et al 2014 Magnetron sputtered zinc oxide nanorods as thickness-insensitive cathode interlayer for perovskite planar-heterojunction solar cells ACS Appl. Mater. Interfaces 6 20585–9 [142] Dharani S et al 2014 High efficiency electrospun TiO2 nanofiber based hybrid organic-inorganic perovskite solar cell Nanoscale 6 1675–9 [143] Ko H-S, Lee J-W and Park N-G 2015 15.76% efficiency perovskite solar cells prepared under high relative humidity: importance of PbI2 morphology in two-step deposition of CH3NH3PbI3 J. Mater. Chem. A 3 8808–15 [144] Kim H-S and Park N-G 2014 Parameters affecting I–V hysteresis of CH3NH3PbI3 perovskite solar cells: effects of perovskite crystal size and mesoporous TiO2 layer J. Phys. Chem. Lett. 5 2927–34 [145] Yu. Y et al 2015 Development of lead iodide perovskite solar cells using 3D titanium dioxide nanowire architectures ACS Nano 9 564–72 [146] Kim H-S et al 2013 High efficiency solid-state sensitized solar cell-based on submicrometer rutile TiO2 nanorod and CH3NH3PbI3 perovskite sensitizer Nano Lett. 13 2412–7 [147] Etgar L et al 2012 Mesoscopic CH3NH3PbI3/TiO2 heterojunction solar cells J. Am. Chem. Soc. 134 17396–9 24

Review

Rep. Prog. Phys. 79 (2016) 026501

[171] Leijtens T et al 2015 Stability of metal halide perovskite solar cells Adv. Energy Mater. 5 1500963 [172] Rong Y et al 2015 Beyond efficiency: the challenge of stability in mesoscopic perovskite solar cells Adv. Energy Mater. 5 1501066 [173] Philippe B et al 2015 Chemical and electronic structure characterization of lead halide perovskites and stability behavior under different exposures-a photoelectron spectroscopy investigation Chem. Mater. 27 1720–31 [174] Supasai T et al 2013 Formation of a passivating CH3NH3PbI3/PbI2 interface during moderate heating of CH3NH3PbI3 layers Appl. Phys. Lett. 103 183906 [175] Leijtens T et al 2013 Overcoming ultraviolet light instability of sensitized TiO2 with meso-superstructured organometal tri-halide perovskite solar cells Nat. Commun. 4 2885 [176] Pathak S K et al 2014 Performance and stability enhancement of dye-sensitized and perovskite solar cells by al doping of TiO2 Adv. Funct. Mater. 24 6046–55 [177] Aristidou N et al 2015 The role of oxygen in the degradation of methylammonium lead trihalide perovskite photoactive layers Angew. Chem. Int. Ed. 54 8208–12 [178] O’Mahony F T F et al 2015 Improved environmental stability of organic lead trihalide perovskite-based photoactivelayers in the presence of mesoporous TiO2 J. Mater. Chem. A 3 7219–23 [179] Kwon Y S et al 2014 A diketopyrrolopyrrole-containing hole transporting conjugated polymer for use in efficient stable organic–inorganic hybrid solar cells based on a perovskite Energy Environ. Sci. 7 1454–60 [180] Li X et al 2015 Improved performance and stability of perovskite solar cells by crystal crosslinking with alkylphosphonic acid ω-ammonium chlorides Nat. Chem. 7 703–11 [181] Wei Z et al 2014 Cost-efficient clamping solar cells using candle soot for hole extraction from ambipolar perovskites Energy Environ. Sci. 7 3326–33 [182] Yu Y-Y et al 2014 Perovskite photovoltaics featuring solution-processable TiO2 as an interfacial electrontransporting layer display to improve performance and stability Nanoscale 6 11403–10 [183] Subbiah A S et al 2014 Inorganic hole conducting layers for perovskite-based solar cells J. Phys. Chem. Lett. 5 1748–53 [184] Li M-H et al 2015 Inorganic p-type contact materials for perovskite-based solar cells J. Mater. Chem. A 3 9011–9 [185] Zheng L et al 2014 A hydrophobic hole transporting oligothiophene for planar perovskite solar cells with improved stability Chem. Commun. 50 11196–9 [186] Xiao J et al 2015 Efficient CH3NH3PbI3 perovskite solar cells based on graphdiyne (GD)-modified P3HT holetransporting material Adv. Energy Mater. 5 1401943 [187] Chen W et al 2015 Efficient and stable large-area perovskite solar cells with inorganic charge extraction layers Science 350 944–8 [188] Kim J H et al 2015 Enhanced Environmental Stability of Planar Heterojunction Perovskite Solar Cells Based on Blade-Coating Adv. Energy Mater. 5 1401229 [189] Chen Y, Chen T and Dai L 2015 Layer-by-layer growth of CH3NH3PbI3−xClx for highly efficient planar heterojunction perovskite solar cells Adv. Mater. 27 1053–9 [190] Reese M O et al 2011 Consensus stability testing protocols for organic photovoltaic materials and devices Sol. Energy Mat. Sol. C 95 1253–67 [191] Niu G, Guo X and Wang L 2015 Review of recent progress in chemical stability of perovskite solar cells J. Mater. Chem. A 3 8970–80

[148] Chandiran A K et al 2014 Sub-nanometer conformal TiO2 blocking layer for high efficiency solid-state perovskite absorber solar cells Adv. Mater. 26 4309–12 [149] Wojciechowski K et al 2014 Sub-150 °C processed mesosuperstructured perovskite solar cells with enhanced efficiency Energy Environ. Sci. 7 1142–7 [150] Luo S and Daoud W A 2015 Recent progress in organic– inorganic halide perovskite solar cells: mechanisms and material design J. Mater. Chem. A 3 8992–9010 [151] Mahmood K, Swain B S and Amassian A 2014 Doublelayered ZnO nanostructures for efficient perovskite solar cells Nanoscale 6 14674–8 [152] Sun K et al 2015 Efficiency enhancement of planar perovskite solar cells by adding zwitterion/LiF double interlayers for electron collection Nanoscale 7 896–900 [153] Min J et al 2015 Interface engineering of perovskite hybrid solar cells with solution-processed perylene-diimide heterojunctions toward high performance Chem. Mater. 27 227–34 [154] Azimi H et al 2015 A Universal interface layer based on an amine-functionalized fullerene derivative with dual functionality for efficient solution processed organic and perovskite solar cells Adv. Energy Mater. 5 1401692 [155] Liang P-W et al 2014 Additive enhanced crystallization of solution-processed perovskite for highly efficient planarheterojunction solar cells Adv. Mater. 26 3748–54 [156] Shi S et al 2015 Advancements in all-solid-state hybrid solar cells based on organometal halide perovskites Mater. Horiz. 2 378–405 [157] You J et al 2014 Moisture assisted perovskite film growth for high performance solar cells Appl. Phys. Lett. 105 183902 [158] Snaith H J et al 2014 Anomalous hysteresis in perovskite solar cells J. Phys. Chem. Lett. 5 1511–5 [159] Lin Q et al 2014 Electro-optics of perovskite solar cells Nat. Photonics 9 106–12 [160] Kim H-S, Im S H and Park N-G 2014 Organolead halide perovskite: new horizons in solar cell research J. Phys. Chem. C 118 5615–25 [161] Unger E L et al 2014 Hysteresis and transient behavior in current-voltage measurements of hybrid-perovskite absorber solar cells Energy Environ. Sci. 7 3690–8 [162] Zhang Y et al 2015 Charge selective contacts, mobile ions and anomalous hysteresis in organic-inorganic perovskite solar cells Mater. Horiz. 2 315–22 [163] Zhao C et al 2015 Revealing underlying processes involved in light soaking effects and hysteresis phenomena in perovskite solar cells Adv. Energy Mater. 5 1500279 [164] Sanchez R S et al 2014 Slow Dynamic processes in lead halide perovskite solar cells. Characteristic times and hysteresis J. Phys. Chem. Lett. 5 2357–63 [165] Xu J et al 2015 Perovskite-fullerene hybrid materials suppress hysteresis in planar diodes Nat. Commun. 6 7081 [166] Shao Y et al 2014 Origin and elimination of photocurrent hysteresis by fullerene passivation in CH3NH3PbI3 planar heterojunction solar cells Nat. Commun. 5 5784 [167] Christians J A, Miranda Herrera P A and Kamat P V 2015 Transformation of the excited state and photovoltaic efficiency of CH3NH3PbI3 perovskite upon controlled exposure to humidified air J. Am. Chem. Soc. 137 1530–8 [168] Xie F X et al 2015 Vacuum-assisted thermal annealing of CH3NH3PbI3 for Highly stable and efficient perovskite solar cells ACS Nano 9 639–46 [169] Beilsten-Edmands. J et al 2015 Non-ferroelectric nature of the conductance hysteresis in CH3NH3PbI3 perovskitebased photovoltaic devices Appl. Phys. Lett. 106 173502 [170] Reenen S V, Kemerink M and Snaith H J 2015 Modeling anomalous hysteresis in perovskite solar cells J. Phys. Chem. Lett. 6 3808–14 25

Review

Rep. Prog. Phys. 79 (2016) 026501

[202] Chen P-Y et al 2014 Environmentally responsible fabrication of efficient perovskite solar cells from recycled car batteries Energy Environ. Sci. 7 3659–65 [203] Marshall K P, Walton R I and Hatton R A 2015 Tin perovskite/fullerene planar layer photovoltaics: improving the efficiency and stability of lead-free devices J. Mater. Chem. A 3 11631–40 [204] Hao F et al 2014 Lead-free solid-state organic-inorganic halide perovskite solar cells Nat. Photonics 8 489–94 [205] Noel N K et al 2014 Lead-free organic-inorganic tin halide perovskites for photovoltaic applications Energy Environ. Sci. 7 3061–8 [206] Takahashi Y et al 2011 Charge-transport in tin-iodide perovskite CH3NH3SnI3: origin of high conductivity Dalton Trans. 40 5563–8 [207] Kumar M H et al 2014 Lead-free halide perovskite solar cells with high photocurrents realized through vacancy modulation Adv. Mater. 26 7122–7 [208] Sabba D et al 2015 Impact of anionic Br− substitution on open circuit voltage in lead free perovskite (CsSnI3−xBrx) solar cells J. Phys. Chem. C 119 1763–7 [209] Yin W-J et al 2015 Halide perovskite materials for solar cells: a theoretical review J. Mater. Chem. A 3 8926–42 [210] Bailie C D et al 2015 Semi-transparent perovskite solar cells for tandems with silicon and CIGS Energy Environ. Sci. 8 956–63

[192] Kaltenbrunner M et al 2015 Flexible high power-per-weight perovskite solar cells with chromium oxide-metal contacts for improved stability in air Nat. Mater. 14 1032–9 [193] Xu X et al 2015 Working mechanism for flexible perovskite solar cells with simplified architecture Nano Lett. 15 6514–20 [194] Ryu S et al 2015 Fabrication of metal-oxide-free CH3NH3PbI3 perovskite solar cells processed at low temperature J. Mater. Chem. A 3 3271–5 [195] Troughton J et al 2015 Highly efficient, flexible, indium-free perovskite solar cells employing metallic substrates J. Mater. Chem. A 3 9141–5 [196] Wang X et al 2015 TiO2 nanotube arrays based flexible perovskite solar cells with transparent carbon nanotube electrode Nano Energy 11 728–35 [197] Qiu L et al 2014 Integrating perovskite solar cells into a flexible fiber Angew. Chem. Int. Ed. 53 10425–8 [198] Roldán-Carmona C et al 2014 Flexible high efficiency perovskite solar cells Energy Environ. Sci. 7 994–7 [199] Lim K-G et al 2014 Boosting the power conversion efficiency of perovskite solar cells using self-organized polymeric hole extraction layers with high work function Adv. Mater. 26 6461–6 [200] Dkhissi Y et al 2015 Low temperature processing of flexible planar perovskite solar cells with efficiency over 10% J. Power Sources 278 325–31 [201] Susrutha B, Giribabu L and Singh S P 2015 Recent advances in flexible perovskite solar cells Chem. Commun. 51 14696–707

26

Recent progress and challenges of organometal halide perovskite solar cells.

We review recent progress in the development of organometal halide perovskite solar cells. We discuss different compounds used to construct perovskite...
5MB Sizes 6 Downloads 19 Views