Open access peer-reviewed chapter

Perovskite as Light Harvester: Prospects, Efficiency, Pitfalls and Roadmap

By Ruby Srivastava

Submitted: February 25th 2016Reviewed: July 29th 2016Published: February 22nd 2017

DOI: 10.5772/65052

Downloaded: 1099

Abstract

In the recent years, perovskite materials have attracted great attention due to their excellent light‐harvesting properties. The organic materials of these hybrid inorganic organic light harvesters are used as sensitizers and the inorganic materials have been used as light absorbers. The exceptional properties of these materials such as long diffusion length, high carrier mobility, affordable device fabrication, and adjustable adsorption range have created a new era in optoelectronic technologies. The perovskites have become promising materials due of their versatility in device architecture, flexibility in material growth, and ability to achieve the high efficiency through various processing techniques. The superior performance of silicon‐based tandems by achieving efficiency more than 40% has encouraged researchers to further expand the investigations to higher levels. The quest to transit the research curiosity to the market photovoltaic technology has given a new dimension to the remarkable ascension of perovskite solar cells. This chapter introduces the experimental and theoretical aspects, the electrical and optical properties, pitfalls, and a roadmap for the future prospects of perovskite materials.

Keywords

  • perovskite
  • power conversion efficiency
  • energy conversion and storage
  • toxicity
  • hysteresis

1. Introduction

The fast‐paced industrial development and population growth has increased the consumption of global energy to such an extent that it has become the ultimate necessity to use the renewable energy resources for long‐term sustainable development. Now it has become a challenge for both scientists and technologists to generate the cost‐effective and environmentally friendly renewable energy resources [1, 2].

Although solar cells based on the photovoltaic effect have attracted great attention due to the advantage of decentralization and sustainability, yet they suffer low cost effectiveness. Another emerging class of thin‐film energy devices based on amorphous silicon also tried to capture the market, making headway by processing of costs per unit area [35]. The manufacturing of inorganic thin‐films solar cells needs high‐temperature and high vacuum‐based techniques [6]. In addition, these techniques are limited and due to the inclusion of toxic elements, they are limited to large‐scale production and wide applications [7].

In 1991, a new breakthrough emerged in the form of dye‐sensitized solar cells (DSSCs) that have attracted considerable attention due to their potential application in low‐cost solar energy conversion [816]. A high efficiency exceeding 12% was obtained by using 10 μm mesoporous TiO2 film sensitized with a cobalt redox electrolyte and an organic dye [17]. Furthermore, solid‐state DSSCs were also investigated where the liquid electrolyte was replaced by a solid hole‐transporting material (HTM) [e.g., poly(3‐hexylth‐iophene)(P3HT),2,2′,7,7′‐tetrakis‐(N,N‐di‐p‐methoxyphenyl‐amine)‐9,9′spirobifluorene (spiro‐MeOTAD)], polyaniline, and polypyrrole [8] to increase the open circuit voltage and stability of solar cells [1822]. However, these ss‐DSSCs also suffer from faster electron recombination dynamics between electrons (TiO2) and holes (hole transporter), which results in the low efficiency of ss‐DSSCs [23]. So attempts were made to design various types of cells to increase the efficiency of solar cells [24].

This efficiency criterion was increased by the introduction of the perovskite sensitizer ABX3 (A = CH3NH3, B = Pb, Sn, and X = Cl, Br, I), introduced by Prof. Grätzel and team, which has opened a new era in the field of DSSCs due to the excellent light‐harvesting capabilities [2437]. These materials are composed of earth abundant materials, inexpensive, processable at low temperatures (printing techniques), generate charges freely (after absorption) in bulk materials, which qualify them as low energy‐loss charge generators and collectors [3840]. Methylammonium lead trihalide (CH3NH3PbX3, where X is a halogen ion such as I, Br, and Cl) have an optical bandgap between 2.3 and 1.6 eV depending on halide content, while formamidinum lead trihalide (H2NCHNH2PbX3) also have a bandgap between 2.2 and 1.5 eV. The minimum bandgap is closer to optimum for a single‐junction cell than methylammonium lead trihalide, which enhance to higher efficiencies [41]. The power conversion efficiency (PCE) of perovskite cells was improved from 7.2 to 15.9%, which is associated with the comparable optical absorption length and charge‐carrier diffusion lengths, making this device the most outperforming relative to the other third‐generation thin‐film solar cell technologies. Although two different configurations using CH3NH3PbI3 perovskite in a classical solid‐state DSSC and in a thin‐film planar configuration with CH3NH3PbI3−xCl x, having efficiency exceeding 16%, have been reported [26, 42], provided few issues related to the stability and hysteresis are to be solved effectively [43].

Here, it is necessary to mention that the lack of hysteresis that was an obstacle for stable operation in perovskite was observed recently using thin films of organometallic perovskites with millimeter‐scale crystalline grains with efficiencies approximately equal to 18% [44].

The three recent reports have given high hopes in the field of solar cells as EPFL scientists have developed a new hole‐transporting material FDT that can reduce the cost and achieve the power conversion efficiency of 20.2% [45]. Another study by Hong Kong University claims that they have achieved the highest efficiency of 25.5% by perovskite‐silicon tandem solar cells [46]. In the meantime, it has been claimed that the efficiency of more than 30% can be achieved by tandem solar cells based on silicon and perovskites [47].

2. Structure of perovskite materials

The basic structure of perovskite consists of a 3D network corner‐sharing BX6 octahedra, where A (e.g., A = Cs, CH3NH3, NH2CHNH2) cations are located in the larger 12‐fold coordinated holes between the octahedra [44]. It is composed of a metal cation (M = Sn, Pb, Ge, Cu) and its ligantanions (X = O2−, Cl, Br, I, or S2−). In the case of inorganic perovskite compounds, the structures can be distorted as a result of the cation displacements, which give rise to some useful properties of ferroelectricity and antiferroelectricity due to the stereochemically active pairs of A cations [48]. The simple cubic structure of CH3NH3PbI3 is given in Figure 1.

Figure 1.

The crystal structure of perovskites, ABX3, a large cation (A) at center together with metal cation (B) bonded to the surrounded halides (X). Color code: A (CH3NH3),blue; B (Pb), green; and X (I), pink.

These inorganic‐organic hybrid compounds have the advantages of inorganic components that include structural order and thermal stability with interesting characteristics of organic materials such as low cost, mechanical flexibility, and functional versatility [4953]. Numerous compounds have been reported by the covalent bonding between the inorganic and organic bonds [54]. Although the degree of interactions in organic‐inorganic systems with the van der Waals interacting system is relatively small, the reason for the small van der Waals interaction is the choice of organic cations, which is limited as the restricted dimension of the cuboctahedral hole formed by the 12 nearest‐neighbor X atoms. The synthesis of compounds CH3NH3 MX3 with M = Sn, Pb and X= Cl, Br, and I has been successfully carried out by some groups [5557]. These organic cations show orientational disorder at high temperature, while at lower temperature the cubic phase results in a structural phase transition as the tolerance factor is smaller than unity. Upon cooling, the structure distorts to lower its symmetry as there are many restrictions to the motion of methylammonium cations [57].

MA, FA, Pb, and Sn perovskite combinations to identify three distinct phase transitions that occur are classified as a high temperature α phase, an intermediate β temperature phase, and a low temperature γ phase [54]. These different phases are represented in Figure 2.

Figure 2.

Graphical representation of phase transitions of MA(Pb, Sn)X3 perovskite materials (a) α‐phase, (b) β‐phase, (c) γ‐phase. Precision images are taken at the [006] view. (d) The structural transformation of Br included in MAPbI3. Adapted with permission from reference [37].

The perovskites were first investigated by Goldschmidt in the 1920s [58] in work related to tolerance factors. The tolerance factor, t, with respect to the ionic radius of the actual ions is given in Eq. (1), where r A, r B, and r C are the ionic radius of the A, B, and C ions, respectively.

t=rA+rC2(rB+rC)E1

The tolerance factor of (0.9–1) is for an ideal cubic structure, for a cubic structure with the tolerance factor (0.7–0.9), the A ion is too small or the B ion is too large. This can be resulted in orthorhombic, rhombohedral, or tetragonal structure. For a large A cation, t becomes larger than one, which results in layered perovskite structures [59, 60]. The compiled results are given in Table 1 and the different forms of perovskite material CH3NH3PbI3 are given in Figure 3. The expected structure is also related to Pauling's rules (PRs) [61], given the expected coordination around a two‐component radii (cation/anion) system which is summarized in Table 2.

Figure 3.

The crystal structure of perovskites (CH3NH3PbI3) in different forms: (a) cubic, (b) tetragonal, (c) rhombohedral, and (d) orthorhombic. Color code: CH3NH3, pink; Pb, green and I, blue.

StructureTolerance factorComment for cation/anion
Tetragonal/rhombohedral/orthorhombic0.7–0.9Cation too large or anion too small
Cubic0.9–1.0Ideal perovskite
Layered structures>1.0Cation too large

Table 1.

Tolerance factors for the perovskite structures

Coordinationrc/raCoordination number
0.15–0.22Triangular3
0.22–0.41Tetrahedral4
0.41–0.73Octahedral6
0.73–1.0Cubic8

Table 2.

Coordination and ideal r c r a (Pauling's rules). r c and r a represent the cationic and anionic radii

The smaller tolerance factor is related to lower symmetry tetragonal or orthorhombic structures, whereas larger t (t > 1) could destabilize the three‐dimensional (3D) B‐X network.

The other important parameter is an octahedral factor that plays an important role in these materials, and is given by,

μ=RB/RAE2

where R B is the ionic radii of the B cation and R A is the ionic radii of A anion. If μ > 0.442, the formation of halide perovskite achieves, whereas below this value BX6 octahedron will become unstable and a perovskite structure will not form, although these factors provide a guidelines for the formation of halide perovskite, yet they are not sufficient to predict the structural formations within the perovskite family [62].

2.1. Experimental scenario

2.1.1. Origin of perovskite

Although these materials already possessed useful physical properties, organic‐like mobility, nonlinear optical properties, enhanced exciton binding energies, electroluminescence, magnetic properties, and conductivity, they have emerged as DSSCs only in 2009 [6368]. The performance of DSSCs is assessed by three major parameters: short‐circuit photocurrent (J SC), open‐circuit voltage (V OC), and fill factor (FF), which are further used to calculate the efficiency (PCE). V OC is proportional to the HOMO‐LUMO energy gap and J SC reflects the mobility, efficient light‐harvesting, and carrier generation. These values of different device structures are presented in Table 3.

PerovskitePhoto anodeHTMJ SC (mA/cm)V OC (v)FFPCE (%)References
CH3NH3PbI3mp (TiO2)Spiro17.60.880.629.7[69]
CH3NH3PbI3TiO2 NSSpiro16.10.630.575.5[30]
CH3NH3PbI3mp (TiO2)Spiro18.80.710.668[70]
CH3NH3PbI3mp (TiO2)Spiro18.30.870.6610.4[71]
CH3NH3PbI2Clmp (Al2O3)Spiro17.80.980.6310.9[26]
CH3NH3PbI3TiO2‐NWAsSpiro10.670.740.544.29[72]
CH3NH3PbI2BrTiO2‐NWAsSpiro10.120.820.594.87[72]
CH3NH3PbI3mp (TiO2)Spiro20.00.990.7315.0[73]
CH3NH3PbI3Rutile (TiO2)Spiro15.60.950.639.4[74]
CH3NH3PbI3mp‐ZrO2Spiro17.31.070.5910.8[75]
CH3NH3PbI3(TiO2)crystalSpiro12.860.790.707.29[76]
CH3NH3PbI3mp (Al2O3)Spiro18.01.020.6712.3[37]
CH3NH3PbI3CH3NH3PbI3Spiro21.51.070.6715.4[77]
CH3NH3PbI3mp (TiO2)P3HT12.60.730.736.7[42]
CH3NH3PbI3mp (TiO2)PCPDTBT10.30.770.675.3[42]
CH3NH3PbI3mp (TiO2)PCPDTBT10.50.920.434.2[42]
CH3NH3PbI3mp (TiO2)PTAA16.40.900.619.0[42]
CH3NH3PbI3mp (TiO2)PTAA19.30.910.7012.3[32]
CH3NH3Pb(I1-xBr x)3Mesoscopic and planar structuresPoly(triarylamine)19.641.1174.216.2[79]
CH3NH3PbI3Mesoscopic TiO2Spiro1.0221.277.616.7[80]
FAPbI3Mesoscopic TiO2Spiro1.0320.977416[78]
CH3NH3PbI3-xCl xPlanar heterojunctionSpiro1.1322.7575.0119.3[81]
CH3NH3PbI3FDT1.14822.70.7620.2[55]

Table 3.

comprehensive summary of the performance of perovskite solar cells, including the perovskite materials, photoanodes, hole‐transport materials (HTMs), J SC (mA/cm), V OC (v), FF and PCE (%)

Abbreviations: mp, mesoporous; spiro, spiro OMeTAD.


The first perovskite‐sensitized TiO2 solar cell gave the efficiency of 3.8 and 3.1%, respectively [13]. Later on the titania's surface and CH3NH3PbI3‐based iodide liquid electrolyte solar cell have increased the efficiency to 6.5% [25]. In 2012, the liquid electrolyte was replaced with a solid electrolyte and a PCE of 9.7% was achieved [69]. A sequential deposition method for the formation of the perovskite pigment within the porous metal oxide film was developed with a PCE of 15% in 2013 (short‐circuit current density J SC = 21.5 mA/cm2, open‐circuit voltage V OC = 1.02 V, and fill factor FF = 0.71) [27]. An efficiency of 20% at low temperature was achieved in a processed solar cell, through the end of 2013 [70, 71]. Further, it is reported that the achieved efficiency has above 30% in 2016.

2.1.2. Photoanodes

Mesoporous metal oxide films act as a working electrode for perovskite cells. The charge extraction rates are relatively faster for the perovskite solar cells than the conventional DSSCs [39]. Again the mesoporous TiO2 was replaced by Al2O3 with similar mesomorphology and it was seen that the PCE unexpectedly reached to 10.9% giving hopes for the future increase in efficiency. Furthermore, the DSSC efficiency has improved to 15.9% [27], yet there is the difficulty in pore filling because of the labyrinthine maze structure [72], which was alternatively substituted by a vertically aligned nanowire (NW) and nanotube (NT) structure. These nanotubes and nanowires can be used in pore filling due to their open porous structures. Moreover, they are reported to be better in electron transportation and recombination behavior and hole conductors presenting faster recombination than nanoparticulate films in liquid‐based DSSCs [7375].

As the absorption properties of perovskite are excellent, a possible decrease in the total surface area of the NWs/NTs compared to the nanoparticles does not stimulate the significant reduction of photocurrent. Later it was concluded that perovskite semiconductors in their simple architecture can exhibit sufficiently good ambipolar charge transport and the principal roles of photovoltaic operation, including charge generation, light absorption, and transport of both electrons and holes. Now the challenge is to determine whether mesostructure is essential or the thin‐film p‐i‐n can lead to a better performance [76].

2.1.3. Perovskite thin films

While using the methylammonium lead halide (CH3NH3PbX3, X‐halogen) and its mixed‐halide crystals, corresponding to the 3D perovskite structures as light harvesters in solar cells, it is observed that substituting the I with Cl/Br ions, bandgap tuning of MAPbX3 is achieved, which occurred due to the strong dependence of electronic energies on the effective exciton mass [76]. The entire visible region was controlled by tuning the bandgap. Apart from that, the addition of Cl/Br into an iodide‐based structure shows a drastic improvement in the charge transport, relative stability, and separation kinetics within the perovskite layer [77]. It was also observed that the bandgap is reduced (1.48–2.23 eV), leading to high short‐circuit currents of >23 mA/cm2 and a PCE of up to 14.2%, when the cation size of perovskite materials is increased [42].

There are a few solution‐based techniques that has been used for the fabrication of thin films, where a mixture of two precursors is used to form final absorber, but due to the lack of suitable solvents and high‐reaction rate of the perovskite component, the process results in thin film with pinhole formation and incomplete surface coverage, which deteriorates the film quality and thus effect the device performance.

The two‐step deposition technique that was used previously to prepare the films of organic‐inorganic systems has incompatible solubility characteristics where the organic component is difficult to evaporate. Devices based on the planar CH3NH3PbI3 thin film via the modified two‐step deposition technique have also achieved the efficiency of 12.1% [78].

Another technique that was developed was dual‐source vapor‐deposited organometallic trihalide perovskite solar cells based on a p‐i‐n thin‐film architecture with high efficiency. However, the deposition with the vacuum evaporation method will make it cost effective [26].

2.1.4. Hole‐transporting materials (HTMs)

The conductivity of perovskite is high, which requires a thick layer of HTM to avoid pinholes. Spiro‐OMeTAD, due to being less conductive, offers high resistance because of thick capping layers. A wide variety of polymer hole conductors are also used, which is shown in Figure 4. Protic ionic liquids (PILs) are used as effective p dopants in hybrid solar cells [78] based on triarylamine hole‐transporting materials. Further, the efficiency is improved by replacing the lithium salts, p‐dopants for spiro‐OMeTAD with PILs [79]. While using other HTMs as P3HT and DEH HTM, the efficiency of spiro‐OMeTAD is much slower than P3HT and DEH HTM, respectively. However, a recent synthesis based on the pyrene‐based derivative Py‐C also exhibited an overall PCE of 12.7% [76]. As the hole conductors, spiro‐OMeTAD and P3HT are costly, so an inexpensive, stable, solution‐processable inorganic CuI as the hole conductor has been demonstrated [80]. A solution‐processed p‐type direct bandgap semiconductor CsSnI3 with a perovskite structure can also be used for hole conduction replacing a liquid electrolyte [34]. Overall we can say that perovskite materials play both the role of light harvesters and hole conductors. Recently, a hole‐conductor‐free mesoscopic CH3NH3PbI3 perovskite/TiO2 heterojunction solar cell has reported with a PCE of 5.5% [30], yet the photovoltaic performance was inferior to that of HTM. The tuning of bandgap of perovskite materials plays an important role in photophysical properties. The energy bandgaps of different hybrid materials and the hole‐transporting materials are given in Figures 5 and 6.

Figure 4.

Structural representation of hole‐transporting materials (HTMs).

Figure 5.

Energy bandgap diagram of hybrid perovskite materials.

Figure 6.

Energy level diagram of hole transporting materials (HTMs).

2.1.5. Measurement of charge‐carrier mobility, lifetime and diffusion lengths

Regarding the exciton or the electron and hole diffusion length, it was observed that 100‐nm long range diffusion length was obtained in solution‐processed CH3NH3PbI3 by applying femtosecond transient optical spectroscopy to bilayers that interface this perovskite with either selective‐electron or selective‐hole extraction materials [38]. The higher efficiency of these materials is only due to the comparable optical absorption length and charge‐carrier diffusion lengths. Photoluminescence quenching measurements were performed to extract the electron‐hole diffusion lengths in triiodide (CH3NH3PbI3) and mixed‐halide (CH3NH3PbI3−xCl x) perovskite thin films [39]. In mixed‐halide perovskite cells, the larger diffusion length is due to the much longer recombination time, requires both low recombination rates and high charge‐carrier mobility; however, the mechanism causing the extended diffusion length is still unclear. Few other things that remain unclear is the relative fraction of free and bound charge pairs at room temperature, the nature of the excited state, and the role of the two species [81, 82].

2.2. Theoretical scenario

There are reports that prove that Density functional theory (DFT) calculations have already carried out before the first perovskite solar cell was reported experimentally [4, 13]. Various theoretical methods were adopted using exchange‐correlation functionals such as Local density approximation (LDA) [83], Generalized gradient approximation (GGA) [51], hybrid functional methods (HSE), quasiparticle GW methods, spin‐orbit‐coupling (SOC), and van der Waals interactions. LDA underestimates and GGA overestimates the lattice parameters. It is observed that when dispersion interactions are included, the calculated results match well with the experimental results. It is found that adding dispersion corrections increases the binding and corrects the GGA errors.

However, the defects does not affect much as they do not create a detrimental deep level within the bandgap [84, 85] that could be carrier traps and recombination centers for electron‐hole in solar cells. Ringwood [86] has included that the contribution of charges depends on the differences in electronegativity. Since lead is considered as a provider of the charge and size, it holds the perovskite crystals all together.

2.2.1. Ambipolar activities

The ambipolar activities of these materials can be defined by taking effective mass into consideration which is defined by formula:

 m*= ħ2[δ2ε(k)δ2ε(k)]1E3

where ε(k) is the energy dispersion relation functions, described by the band structures. If the band is more dispersive (flat), near the band edges, the effective mass is lighter (heavier). In perovskite materials, the lone‐pair Pb s electrons play a vital role. The electronic structure of CH3NH3PbI3 is inverted. The conduction band matrix is derived from Pb p orbitals, and the valence band matrix is a mixture of Pb s and I p (s‐p semiconductor) orbitals. A cation Pb p orbital has a much higher energy level than anion p orbitals, although the CBM is derived from Pb p orbitals, Therefore, the lower conduction band of CH3NH3PbI3 is more dispersive than the upper valence band, similarly the upper valence band of CH3NH3PbI3 is dispersive due to the strong s‐p coupling around the Valence band maximum (VBM). Due to the balance between the hole effective mass and the electron effective mass, CH3NH3PbI3 leads to higher ambipolar activities. It might be possible that many‐body effect plays a role for small carrier effective mass, as the effective mass calculated by the GW + SOC method [87] is even lower. The effective hole and electron masses are given in Table 4.

Materialsm e */m 0m h */m 0Bandgap (eV)
Silicon0.260.291.11
GaAs0.070.341.43
CsSnI30.190.09 (0.15)1.14
CsSnI3 (SOC)0.160.07
CH3NH3PbI30.35 (0.32)0.31 (0.36)1.5–2.0
CH3NH3PbI3 (SOC)0.18 (0.23)0.21 (0.29)

Table 4.

Calculated effective masses (electron and holes) and bandgap (eV) for different materials. Experimental values are in parenthesis

2.2.2. Optical absorption spectra

The optical absorption spectra of perovskite materials are determined by the energy bandgaps and partial density of states (pdos). The pdos graph for different materials is depicted in Figure 7. The energy bandgap measures the probability of each photoelectric transition and the density of states measures the total number of possible photoelectric transitions. Thus, we can easily conclude that the optical absorption coefficient of a material is closely related to its electronic structure. However, the effect of optical absorption spectra is not considered in the Shockley‐Queisser limit [42]. The theoretical maximum efficiency depends on the thickness of the absorber layer. Recently, a method has been developed by Yu et al. [88], in which they calculated the maximum efficiency based on the absorber thickness by taking absorption coefficient and absorber layer thickness both into consideration. So theoretical calculations were carried out on this basis and it was found that halide perovskites (CH3NH3PbI3 and CsPbI3) exhibit much higher conversion efficiencies for any given thickness. These materials are also capable of achieving high efficiencies with very thin absorber layers. On the basis of experimental calculations, it is proved that CH3NH3PbI3 perovskite has the capability of achieving a high fill factor. Improved interfaces and contact layers also improve the performance of a solar cell, while Pb chalcogenides exhibit abnormal bandgap changes with lattice constant and strain [89].

Figure 7.

(a) The periodic structural model of Σ5 (310) GB for CH3NH3PbI3. (b) Comparison of DOS of bulk CH3NH3PbI3calculated from unit cell. (c–f) pdos of selected atoms highlighted in the above structure. Adapted with permission from reference [137].

2.2.3. Ferroelectricity

One more theoretical aspect is the dipole moment of the noncentrosymmetric organic cation in perovskite materials. It was shown from electric dipole calculations of the organic cation that hybrid perovskites exhibit spontaneous electric polarization, which might be due to the two reasons: the alignment of the dipole moments of organic cations and the intrinsic lattice distortion breaking the crystal centrosymmetry. On the basis of this concept, it was proposed in the studies that the presence of ferroelectric domains will result in internal junctions might support electron‐hole separation and transportation. However, the calculated value of CH3NH3PbI3 bulk polarization is 38 mC/cm2, which is comparable to the value of ferroelectric oxide perovskites such as KNbO3 (30 mC/cm2) [90]. Frost and coworkers [91] suggested that it may be possible that the boundaries of ferroelectric domains may form “ferroelectric highways” that facilitate the transportation of electrons and holes. Furthermore, it was proposed that the favorable highways are energetically chosen in such a way that the holes and electrons avoid any collision with the opposite charges. It is directly seen in the recent experiment by direct observation of ferroelectric domains in the β phase of CH3NH3PbI3. Another important factor is that V OC can be larger than the bandgap, and charge separation and carrier lifetime can be enhanced due to the internal electric field [92].

2.2.4. Interface and surface

The surface and interface between the absorber, carrier transport layers, and electrode contact layers are also important for efficient carrier transportation. However, the two‐step method, vacuum deposition and vapor‐assisted solution processing methods [85], have improved the quality much better by the one‐step method. The vacuum deposition method is used in small molecule‐based devices, which makes the use of insoluble materials more stable than their soluble analogues. There are at least three aspects worth consideration.

2.2.4.1. Band alignment

The bandgaps and band alignments of perovskites can also be tuned by the chemical management of compositional elements, including organic cations [93, 94], Pb [9597], and halogen elements [98, 99]. This is another way to optimize band alignment at interfaces.

2.2.4.2. Interface structure and passivation

The unusual hysteresis of the IV curve of perovskite solar cells, which would reduce the working cell efficiency, was suspected to be related to the interface properties [99, 100].

2.2.4.3. Surface

Abate et al. [79] reported the existence of trap states at the perovskite surface, which generated charge accumulation and consequent recombination losses in working solar cells. They identified under coordinated iodine ions as responsible and used supramolecular halogen bond complexation for passivation.

2.2.5. Point defects

The p‐ or n‐type absorbers were made from materials with intrinsic defects, or using intentional doping intrinsic defects that create deep energy levels in the absorber usually act as Shockley‐Read‐Hall nonradiative recombination centers and carrier traps, reducing the carrier lifetime and thus V oc. A good solar cell absorber must exhibit proper doping and defect properties. There are many types of defects as a donor and acceptor which lies in the semiconductors. The formation energy of a defect depends on the chemical potential and environmental factors such as precursors, partial pressure, and temperature. So we can conclude that these experimental conditions play a vital role to determine the formation energies of all the possible defects and further impact the polar conductivity in these materials. Defect formation energies determine the polar conductivity of a semiconductor, whereas defect transition levels determine the electrical effect of any particular defect [101].

Besides point defects, Kim et al. [102] used DFT‐GGA to calculate the DOS and partial charge densities of two types of neutral defects in β phase CH3NH3PbI3: (a) Schottky defects (equal numbers of positive and negative vacancies) and (b) Frenkel defects (equal numbers of vacancies and interstitials of the same ion). The tunable polar conductivity and shallow defect properties may help to explain why high‐performance perovskite solar cells, with extremely long carrier lifetimes [40, 103] can be produced by a diverse range of growth approaches and a wide variety of solar cell architectures. These point defects would suggest new methods for perovskite solar cell architecture. It was observed that deep point defect levels could exist through large atomic relaxations, which is attributed to the strong covalency of the system [104].

2.2.6. Structural disorder

In a recent investigation, Choi et al. [105] found that most of CH3NH3PbI3 (70%) is highly disordered with a local perovskite structure extending over a range of only 1.4 nm, which is about 2 lattice constants of the α phase [106].

The mesoporous scaffold confined need the perovskite within the pores and reshaped the structures of perovskites. On the other hand, the low‐temperature growth process inevitably leads to polycrystalline perovskites with grain boundaries (GBs). Experimentally, it is very difficult to investigate the structural and electronic properties directly, as it requires a high resolution transmission spectroscopy (HRTEM). So, we have to rely on the theoretical calculations that can give direct insights into the electrical properties of structural disorders and topological defects in hybrid perovskites. Recent combined theoretical and experimental studies [106] have demonstrated that Cl segregated into the GB part of polycrystalline CdTe solar cells effectively taming the detrimental effects at GBs.

Due to the structural complicity of CH3NH3PbI3, the GB structures were constructed based on CsPbI3. It was observed that the DOS of the supercells with GBs are very similar to those of single‐crystal phases. None of these GBs introduce defect states within the bandgap region. The GW band structure diagram is given in Figure 8.

Figure 8.

DOS graph of MASnI3 and MAPbI3materials. Adapted with permission from reference [116].

3. Properties

3.1. Electrical properties

Hybrid perovskites exhibit unprecedented carrier transport properties that enable their stellar performance in photovoltaics. So more attention is needed to develop understanding the material properties and ways to improve these properties in all key directions for research. The electrical properties of perovskite materials are seen in the ambipolar carrier transport behavior and long carrier lifetime. These electrical properties are further investigated on the basis of corresponding device structure.

3.1.1. Intrinsic electrical properties

The electrical characteristics of the materials are determined by the carrier type, concentration, and mobility, which is dependent on the method of preparation. It is necessary to use smooth and uniform films to perform measurements. The carrier type is determined by Hall measurements of the conductivity's response to an applied magnetic field, thin‐film transistor's response to a gating electric field, and thermoelectric measurements of the Seebeck coefficient. For example, CH3NH3PbI3 indicated n‐type conductivity, a carrier concentration of ~109 cm−3, and an electron mobility of 66 cm2/V/s [24]. Carrier concentration can also be adjusted by tuning the stoichiometry of the precursors during solution‐phase synthesis and even switch the carrier type to the p‐type when excess CH3NH3I is used in two‐step synthesis. The electron concentration was measured to be ~1017–1018 cm−3, and it was proposed that the iodide vacancies are responsible for the n‐type conductivity [107]. The electron mobility for n‐type films deposited from stoichiometric precursors was determined to be 3.9 cm2/V/s from the Hall measurements, although CH3NH3SnI3 prepared by a solid‐state reaction in a vacuum‐sealed tube showed an electron mobility of 2320 cm2/V/s [24], while solution processed material measured mobility of 200 cm2/V/s. It was observed that the electron mobility of polycrystalline CH3NH3PbI3 films is larger than the thin‐film mobility of polymers [107, 108] and colloidal quantum dots (10−3–1 cm2/V/s) [109] comparable to CdTe (10 cm2/V/s) [110] CIGS, Cu2ZnSnS4 (CZTS) (10–102 cm2/V/s) [111, 112], and polycrystalline Si (40 cm2/V/s) [101]. Film morphology plays an important role as the dark and light conductivities of CH3NH3PbI3−xCl x deposited on a planar scaffold on mesostructured aluminum oxide are quite different [113]. To further increase the photovoltaic performance and radiative lifetime, solvent annealing has been applied to increase the grain size of the films to ~1 μm [114].

3.1.2. Extrinsic electrical properties:

The techniques used to measure the electrical parameters are given in subsections.

3.1.2.1. Impedance spectroscopy (IS) [115, 116]

This technique is used to identify the frequency dependence of capacitance, to measure charge diffusion lengths and lifetimes and to investigate carrier trapping and recombination. The carrier diffusion length was derived and has been estimated to be about ~1 μm for CH3NH3PbI3−xCl x [83].

3.1.2.2. Electron beam‐induced current (EBIC) [117]

Another method to obtain the electrical parameters is EBIC from which the calculated carrier diffusion length forCH3NH3PbI3−xCl x is 1.5–1.9 μm [40]. The carrier diffusion length is comparable or longer than that of other polycrystalline semiconductors with direct bandgaps used in solar cells [76, 77, 118120]

3.2. Optical properties

It is very important to understand the optical response of the materials, as optical properties are the most important feature of perovskite materials and they provide insights into the electronic and chemical structures. The ability to tune the optoelectronic properties with ease presents a major attraction among researchers. Few important parameters that are used to define these properties are discussed herein:

3.2.1. Optical constants

A lot of research has been conducted on tuning the bandgap of perovskite, but a more detail understanding of these materials awaits further research. The major problem that occurs in perovskite materials is the difficulty of producing continuous films of sufficient smoothness [121] to avoid measurement artifacts from spectroscopic measurements of transmittance, reflectance, and ellipsometry. The absorption coefficients determined from the absorption of CH3NH3PbI3 films on quartz [122] and glass [123] yield values of ~104 cm−1 near the band edge without providing any corrections for the surface's inhomogeneity, so for accurate measurements is important to calculate the absorption coefficients based on the optical constants of CH3NH3PbI3 [124]. It is observed that the absorption spectrum for CH3NH3PbI3 differs, when deposited within a mesoscopic template and planar substrates, which might be due to the changes in the crystallite morphology that affects the optical transitions [125, 126].

3.2.2. Excitons

Exitons play an important role in perovskites. The studies indicate, however, that there is not significant population of excitons in photovoltaics made from CH3NH3PbI3, whose exciton‐binding energy has been reported between 20 and 50 meV, comparable to the thermal energy at room temperature [127, 128]. These values have been obtained by fitting temperature‐dependent absorption spectra using the measured [88] reduced mass of the exciton. Excitonic radius from the binding energy and an appropriate dielectric constant study is still a subject of debate [129]. The excitonic transition significantly enhances the absorption of hybrid perovskites near the band edge [130, 131].

3.2.3. Photoluminescence

The photoluminescence (PL) efficiency depends on the pump fluence. The trapping of photogenerated charges competes effectively with direct radiative recombination of electrons while holes reduce luminescence at low excitation energies. The PL efficiency ofCH3NH3PbI3is ~17–30%. The PL efficiency falls at higher pumping and high charge carrier densities. The PL lifetime measurements reported shorter lifetime (between 3 and 18 ns) at low pump fluencies [127, 132134]. These longer lifetimes have been found in a semiconductor in doped and undoped GaAs films. This might be due to the photon recycling and the PL lifetime dependency on surface recombination than radiative recombination. So we can conclude that photon cycling plays a major role in their excited state dynamics, when nonradiative decay pathways are suppressed. The absorption spectra and photoluminescence for perovskite materials are shown in Figure 9.

Figure 9.

(a) Absorption spectra, (b) photoluminescence spectra of FAPbI xBr3−x (varying I:Br ratio), (c) XRD spectra of the phase transition Br‐rich cubic phase to the I‐rich tetragonal phase. Adapted with permission from reference [37].

3.2.4. Vibrational spectroscopy

IR spectroscopy also plays an important role in determining the chemical composition. If we look into the chemical structure of CH3NH3PbI3, CH3NH3PbBr3, and CH3NH3PbCl3, the first one is tetragonal, while the other two are cubic. Raman‐active modes are precluded in the symmetry of the lattice for cubic structures [135], though a weak broadband at 66 cm−1 is observed in CH3NH3PbCl3. For CH3NH3PbI3, the resonant Raman spectrum (DFT calculations) has been observed with nodes below 100 cm−1 (approximately) related to the inorganic octahedron. The higher energy modes indicate the disorder of CH3NH3 + cations. A lot of work in this field is still required to investigate how the modes shift occurs with the structural changes. Raman nodes can provide better tool in understanding the in homogeneity of perovskite films with submicron spatial resolution.

4. Pitfalls

4.1. Hysteresis

Perovskite solar cells exhibit an anomalous hysteresis in the current‐voltage and resistivity‐temperature dependence curves [136]. Though it was predicted that the hysteresis on resistivity verses temperature curves is associated with the structural phase transition while the reason for current‐voltage curves are still unknown. In an extensive [E‐CE6] studies carried out by Prof. Erik Christian Garnett et al. [136], several explanations have been proposed as ion migration, filling of interface, or surface trap states, accumulation of charges at grain boundaries and ferroelectricity, yet no convinced conclusion has been drawn. In structural perception, the cubic phases of the chloride and bromide perovskites do not allow a polar ferroelectric distortion. Various hypotheses have been suggested and it was further predicted that hysteresis should depend on the magnitude of the dipole moment of the organic cationic species and the connecting halide cage. Though the origin of this phenomenon is not yet understood properly, a number of possible causes have been proposed in which the noted causes are ferroelectricity or the presence of mobile ionic species [136]. The illustration for the hysteresis in the electrical transport in hybrid perovskites is given in Figure 10.

Figure 10.

Hysteresis representation in hybrid perovskites. (a) I‐V graph of CH3NH3PbI3 (single crystal) at room temperature, (b) schematic I‐V curve, (c) proposed phenomena for its origin. Adapted with permission from reference [104].

Here, it is necessary to mention that reporting results from single JV sweeps, even in the absence of hysteresis, or choosing scan rates to report the highest efficiencies, will lead to misleading results. As it might be possible that the certified efficiencies for perovskite solar cells are deemed “not stabilized” though they were measured with negligible hysteresis.

4.2. Thermal and operational stability

There are so many reports that claim that perovskite solar cells have been shown to be stable for many hundreds of hours without any encapsulation. However, the solar cells were stored in the dark and only measured occasionally. So we can conclude that the sealing from environmental ageing is necessary because of operation at elevated temperature and humidity. Stability has become a bigger problem for tin (II) perovskites due to the decrease in stability of the oxidation state of tin (II) compare to lead (II).

4.3. Toxicity

Due to the toxic nature of lead, concerns have been raised on the possible environment and legalization problems from perovskite solar cells based on water soluble lead compounds. So efforts have been made to replace lead with other metal ions without degrading the photophysical properties with quantum mechanical calculations. As lead halogen perovskites are water soluble, the most pessimistic view is the consequences of damaged solar cells and panels with potential exposure to water followed by dissolution and distribution of lead ions into buildings, soil, air, and water.

Lead is known to damage the nervous system and cause brain disorders. In this direction, a theoretical study carried out by De Angelis and group [137] has replaced Pb by Sn (Figure 11) with effective development of the GW method with spin‐orbit coupling to accurately model the properties of CH3NH3SnI3 and then compared it to the CH3NH3PbI3. They predicted that MASnI3 is a better electron transporter than MAPbI3 by the SR‐DFT method. Another study carried out by Jesper Jacobsson and group [138] has provided deep physical insights into the photophysical nature of a metal‐halogen perovskite by removing lead with strontium, which is relatively nontoxic and inexpensive. CCSD calculations and DFT study were performed on the two basic structures of CH3NH3SrI3 and CH3NH3PbI3 to extract and compare the electronic structures and the optical properties. This is based on the fact that the ionic radii of Sr2+ and Pb2+ are almost identical, so the exchange could be made as it will not affect the crystal structure. CH3NH3SrI3 gives a bandgap of 1.6 eV, which is fairly close to the experimental value reported to be around 1.55 eV [5, 42]. The second effect that was caused by shifting Sr for Pb is that the shape of the pdos graphs for both the halogen and the organic ion is shifted and slightly distorted. The lower electronegativity of Sr compared to Pb shifts the electronic cloud closer to the iodine atoms in the lattice, which perturb the local dipole moment as well as the bonding angles between the iodine octahedra and consequently their columbic interaction with the methylammonium dipoles. The charge distribution is similar to the two structures, with higher charge density around lead compared to strontium due to the higher atomic number of lead.

Figure 11.

Pictorial representation of replacement of lead by strontium in perovskite solar cells [138].

5. Roadmap

The Perovskite solar cell (PSC) field has now become an emerging field and reports on further improvement in performance are expected in the near future, achieving PCE of more than 30% efficiency has now become a realistic goal. Furthermore, PSC can be used as top cells in two‐level tandem configurations using crystalline silicon or copper indium gallium selenide‐based photovoltaic devices as bottom cells. It is expected that by using silicon‐based tandems, PCEs of 28–30% can be achieved. Yet there are issues related to the stability and toxicity, hysteresis in perovskite solar cells, which has to be solved. Experimental and theoretical investigations have demonstrated that that halide perovskites exhibit a series of superior electronic and optical properties for solar cell applications, such as proper bandgap and band alignment, high optical absorption, bipolar carrier conductivity, tunable doping ability, and benign defect properties. A lot of studies are required to optimize the material properties and to find new perovskite candidates for high‐efficiency, stable solar cells. Band structure engineering of CH3NH3PbI3 needs to be extensively investigated by replacing organic cations, Pb or I, with other choices. Furthermore, the mechanisms of performance degradations have to be resolved in a more prominent manner. Water‐corroded perovskites as rapid degradation occur in moist environments. So the reaction mechanism between H2O and the perovskite surface could be carefully studied, leading to the development of new methods for stabilizing perovskites. Although some groups have fabricated the long‐term stable perovskites in the laboratory through chemical composition engineering [32, 88], the fundamental reason for alloy stabilization of the structures requires more study. However, it is predicted that the study should converge to the p‐i‐n planar heterojunction perovskite solar cell to understand the device structure and properties from single crystal.

6. Conclusions

The intense appeal of hybrid organic‐inorganic perovskite materials such as solar cells is exceptionally promising. Their enhance optoelectronic properties, deposition techniques, and device structure have led to the higher power conversion efficiencies. Due to the high absorption coefficients and panchromatic absorptions of perovskite, they have become ideal materials for thin film solar cells. However, some complexities as the poor stability in humid air and the toxicity of lead used are a matter of concern. In some perovskite materials, the hysteresis is also pronounced due to the strong dependence of photocurrent to the voltage scan conditions. Still the exceptional performance of hybrid perovskite materials has created revolution in the field of renewable energy with cheap solar cells. Highly efficient solar cells with record performance are still an important milestone to be achieved. The highly innovative and new elegant designs, deep insights into the photophysics and mechanisms of cell operation should now be the main focus of future research.

Finally, we can conclude that the recent advances with perovskite materials will motivate the researchers to expand their horizons to other inorganic or organic pigments, for which the power of mesoscopic solar‐cell architectures will emerge to offers more promising opportunities.

Acknowledgments

The author acknowledges the financial assistance by the DST WOS‐A (CS‐1005/2014). The author is also thankful to her mentors Dr. G. Narahari Sastry, Head, Center for Molecular Modeling and Dr. K. Bhanuprakash, Chief Scientist, I& PC division, CSIR‐Indian Institute of Chemical Technology for the useful discussions and suggestions.

How to cite and reference

Link to this chapter Copy to clipboard

Cite this chapter Copy to clipboard

Ruby Srivastava (February 22nd 2017). Perovskite as Light Harvester: Prospects, Efficiency, Pitfalls and Roadmap, Nanostructured Solar Cells, Narottam Das, IntechOpen, DOI: 10.5772/65052. Available from:

chapter statistics

1099total chapter downloads

More statistics for editors and authors

Login to your personal dashboard for more detailed statistics on your publications.

Access personal reporting

Related Content

This Book

Next chapter

Recent Progresses in Perovskite Solar Cells

By Serafettin Demic, Ahmet Nuri Ozcivan, Mustafa Can, Cebrail Ozbek and Merve Karakaya

Related Book

First chapter

New Quantum Cipher Optical Communication: Y-00

By K. Harasawa

We are IntechOpen, the world's leading publisher of Open Access books. Built by scientists, for scientists. Our readership spans scientists, professors, researchers, librarians, and students, as well as business professionals. We share our knowledge and peer-reveiwed research papers with libraries, scientific and engineering societies, and also work with corporate R&D departments and government entities.

More About Us