2.3. Halide Perovskite Based Solar Cells
2.3.1. Pb-Based Perovskites Active Layers
Since Miyasaka and coworkers integrated HPs as promising light harvesters to provide a step forward in a new generation of solar cells in 2009
[127], these materials have been consolidated as the most prominent semiconductors for photovoltaics. Typically, HPs show a 3D structure based on ABX
3 formula (A = Cs
+, cesium; MA
+, methylammonium; FA
+, formamidinium; B = Pb
2+, Sn
2+; X = Cl
−, Br
−, I
−, or combinations), where the composition and particle size engineering can promote the preparation of HPs with diverse photophysical, chemical, and electronic properties
[128][129][130]. Due to the fact that a solar device should collect as much incident illumination as possible in a broad range of the energy spectrum to reach an efficient light-to-current transformation, low-band gap HPs are ideal for this process. Therefore, HPs with high iodide content are in the focus of many research groups, obtaining different strategies to reach a suitable active layer, compatible and stable with the selective contacts of the solar cell, to facilitate the carrier mobility through the device
[131][132]. Although an acceleration in the number of publications have been reported to show current alternatives to prepare adequate HPs active layers (open the door to an extensive number of reviews), here, the researchers will highlight some of these contributions where the modification of the perovskite layer has been vital to increasing the performance of the solar devices.
Similar to the generations of solar devices shown above, PSCs are composed by an electron transporter layer (ETL) and hole transporter layer (HTL), which favor the carrier transport through the device
[133]. Moreover, a transparent conducting glass as fluorine-doped tin oxide (FTO) is used to promote the carrier mobility and allow the direct illumination of the active layer, while on top of the PSC architecture, metals such as silver or gold are deposited
[134]. As the researchers previously mentioned, iodide HPs are the most relevant photomaterials to achieve a high photoconversion efficiency (PCE) in PSCs, especially perovskites based on FAPbI
3. This material presents a band gap around 1.4 eV, indicating a broad absorption capability of solar energy in the UV-vis-NIR spectrum
[135][136][137]. This band gap value also corresponds to a theoretical PCE ~32.3%, near to the Shockley–Queisser limit of efficiency
[138]. Unfortunately, this feature is achieved for the photoactive phase of the FAPbI
3, denominated as black α-phase, which is stable at high temperatures, but is eventually transformed into the yellow δ-phase at room conditions
[139]. To suppress the fast α-to-δ phase conversion into this kind of perovskite, Meng and coworkers
[140] have prepared bulk films based on FA
0.94MA
0.06Pb(I
0.94,Br
0.06)
3, which have been coupled with carbon electrodes on top, in a n-i-p device configuration, to accelerate the carrier separation and transport into the fabricated solar cell. Here, bromide anions were incorporated into the perovskite to provide better stability to the material since iodide species are more labile, being rapidly diffused from the perovskite lattice to create structural defects. However, the presence of Br also offers the enlargement of the band gap, hampering the sunlight absorption capability of the perovskite. In this context, a high FA content ~94% was introduced into the material, to guarantee a narrow band gap. Attending to the final device, the PCE was estimated to be 20.04% at 1 sun illumination. At this point, the 94% of the initial performance of the solar cell was maintained for 1000 h.
On the other hand, Kim and coworkers
[141] have deeply investigated the impact of the addition of MACl as an additive for the stabilization of FAPbI
3 to obtain highly-performance solar cells. Here, they demonstrated that the α-to-δ phase transition is promoted by the weak interaction between FA
+ cations and the iodide species from the [PbI
6]
4− octahedra building blocks, which is one of the key factors to trigger the octahedra tilting
[142][143][144]. Thus, the presence of Cl provides a better p orbital localization from I, improving the FA-I interactions beyond compensating halide vacancies into the perovskite structure. Then, MA
+ cations also facilitate the perovskite stabilization, causing a contraction of the cubo-octahedra sites in the material. MA
+ shows a higher dipolar moment than that of FA cations (~10 times), strengthening the volume shrinkage, making closer a direct interaction between FA and I.
To date, the octahedral tilting is induced by the loss of halide species such as iodide anions during the perovskite film preparation
[139][145][146], generating a high density of halide vacancies (thereby, high density of carrier traps is produced), leading to the degradation of the photophysical and electronic features of the material. More specifically, the I
− anions released from the perovskite structure are easily oxidized to obtain I
0, which is the main species to trigger chemical chain reactions and accelerate the quenching of the intrinsic properties of the active layer
[146]. For this purpose, different organic agents have been introduced to avoid the formation of halide vacancies and strengthen the Pb-I bonds in the octahedra building blocks, as in the case of caffeine
[147], theobromine
[148], alkylammonium halides
[148], phosphonopropionic acid
[149] and some amino-based organic ligands
[150]. Although some of these organic agents show a high binding capability to the [PbI
6]
4− octahedra units, avoiding the halide deficiency, they also exhibit fast decomposition under heating or can acts as barrier layers (due to their long carbon chains). This fact hinders the carrier extraction from the perovskite, hampering the net performance of the solar device. Therefore, with the purpose of reducing the iodide migration and thereby inhibit the formation of halide defect sites in the perovskite layer, Zhao and coworkers
[151] have added 3-amidinopyridine (3AP) during the material preparation, inducing the coordination between amidino moieties and the Pb-I frameworks. In this scenario, a maximum PCE of 25.3% was obtained in optimized PSCs under 1 sun irradiation, preserving ~92% of the initial efficiency for 5000 h in ambient air.
Other alternatives to stabilize the perovskite active layer have been focused on the direct incorporation of large alkylammonium species to react with the 3D HPs to form multidimensional 2D/3D systems with longer durability
[150][152][153]. Considering that big cations do not match with the Goldschmidt tolerance factor, breaking the 3D preferential crystalline structure of the perovskite, the [PbX
6]
4− octahedra in form of stacking layers are separated by these of organic agents
[154][155]. The 2D perovskites are environmentally stable and promote Vander Waals interactions with the Pb-X frameworks. However, these materials exhibit a large band gap and poor carrier transport capability, decreasing the PCE in modified PSCs. In this way, it is favorable the mixture of 2D/3D systems to facilitate the separation/mobility of electrons through the device. Grancini and coworkers
[156] have proposed the incorporation of 2-thiophenemethylammonium halides, 2-TMAX (X = Cl, Br and I) during the preparation of triple-cation [(FAPbI
3)
0.87(MAPbBr
3)
0.13]
0.92(CsPbI
3)
0.08 perovskite, where the thiophene organic species react with the photomaterial to create the 2D layer. By adding 2-TMABr and 2-TMAI agents, a suitable band structure (valence band shift) in between the 2D-3D perovskites is mediated, favoring the hole extraction into the heterojunction, contrary to 2-TMACl, where the carrier diffusion was hampered. Accordingly, PSCs with a maximum PCE of 20.8% (under 1 sun) were fabricated, keeping 74% of their initial performance for 1000 h of operation without any encapsulation. Another example is mentioned by Zhou and coworkers
[155], where the 2-thiophenemethylammonium (ThMA) spacer was also added during the preparation of FA/MA based 3D perovskite. Here perovskite crystals were orientally grown, inducing the passivation of a high density of carrier traps to avoid the non-radiative recombination losses. Here, devices provided high PCE up to 21.5%; preserving 99% of its initial performance after 1680 h ambient air.
Then, a mixture of organic spacers such as iso-butylammonium iodide (i-BAI) and FA-iodide (FAI) reported by Cho and coworkers
[157] suppressed the interfacial carrier traps and thereby reduced the interfacial energy barrier to avoid carrier diffusion. A PCE as high as 21.7% was obtained for the optimized solar device, keeping 87% of the original efficiency after 912 h with a relative humidity of 75%.
To maximize the absorption ability and electronic features of the bulk HPs, these materials have also been integrated with other types of solar devices such as Si, copper indium gallium selenide, different PSCs, DSSCs and QDs-based solar cells to fabricate tandem devices, as shown above
[158]. Some works have reached near to 30%, as the case of the contribution reported by Albrecht and coworkers
[159], where a monolithic perovskite tandem solar cell (PTSC) was fabricated. Here, a wide-band gap Cs
0.05(FA
0.77MA
0.23)
0.95Pb(I
0.77Br
0.23)
3 bulk film was integrated with a Si-based solar device, also incorporating diverse hole transporter layers such as C
60, 2PACz and Me-4PACz, to accelerate the carrier extraction. A maximum of PCE ~29% was achieved for the optimized PTSC, retaining the 95% of the initial performance after 300 h under operation. Interestingly, De wolf and coworkers
[160] have incorporated MgF
2 as an interlayer between the perovskite based on triple cation Cs
0.05FA
0.8MA
0.15Pb(I
0.755Br
0.255)
3 and C
60 to improve the electron extraction (better adjustment of the relative energy positions in the perovskite) and separate the C
60 from the active layer surface to restrain the non-radiative recombination dynamics. Accordingly, a stabilized PCE of 29.3% was achieved, where the optimized PTSC device retained ~95% of its initial performance after 1000 h of operation, even carrying out damp-heat testing at 85 °C with 85% relative humidity. Recently, this research group provided a new record of the PCE ~33.2% using a similar perovskite layer on top of the cell, while a textured Si solar device is in the bottom of the configuration.
Even though the surface passivation with interesting alkylammonium ligands and the incorporation of 2D/3D heterostructures enhance the operational performance of the solar devices, it is reasonable that some differences in the PCE values are obtained. So, the highest PCE value is reached by filling/replacing structural defects in the active layer, improving its optical and electronic properties, suppressing carrier traps, and facilitating carrier transport. Although 2D/3D assembly can also achieve defect passivation in the perovskite, the appearance of a heterostructure can promote the emergence of interfacial energy levels where the carriers can be trapped, hampering the development of a higher device efficiency.
2.3.2. Partially Pb-Substituted/Pb-Free Perovskite Active Layers
As is highlighted above, a series of approaches are prominent to reach long-term stability and suitable electronic properties in perovskites to favor carrier separation/extraction and mobility through the fabricated device. However, it is also known that one of the main drawbacks of the PSCs is the inherent toxicity of Pb, which hampers the future commercialization of this technology
[161][162]. Therefore, the fact of substituting Pb partially or completely with less toxic metals, preserve/increase the final properties of the active layer, and thereby, reaching high performance in the solar device, is still a huge challenge. One of the most attractive metals to replace Pb is Sn, which also generate [SnX
6]
4− octahedra units to generate multidimensional HPs structures as the case of Pb and provide HPs lower band gap than that of Pb-analogous
[163]. Nevertheless, Sn-HPs exhibit a poor stability consequence of the fast Sn
2+-to-Sn
4+ oxidation under ambient conditions, degrading the net efficiency in solar devices
[164][165]. Hence, the researchers will focus on the fabrication of PSCs based on a mixture of Sn/Pb- or Sn-HPs introducing novel strategies to avoid Sn oxidation in the active layer and thereby increase the performance of Sn-PSCs.
Firstly, for Sn/Pb-HPs systems, Wang and coworkers
[166] have studied the addition of antioxidants such as tea polyphenol (TP) during the preparation of optimized CsPb
0.5Sn
0.5I
2Br bulk film to avoid the rapid Sn oxidation. Through the coordination bonds between high Lewis acidity Sn sites and hydroxybenzene groups, the Sn
4+/Sn
2+ ratio into the HP was decreased from 68.1% to 5.85%, deducing that TP is a good reducing agent. In addition, this antioxidant mediated an improved HP crystallization, making that at 0.5 wt% TP, a homogeneous film with a low content of pinholes is obtained. Therefore, low density of grain boundaries appears to hinder the carrier extraction. A maximum PCE of 8.1% was reached under 1 sun illumination, retaining 95% of original performance after ~1440 h into a N
2-filled glove box. Meanwhile, Guo and coworkers
[167] have used the mixture of 2D/3D HPs in order to prepare suitable heterostructures for PSCs. Here, authors have introduced bulky alkylammonium based on 2-(4-fluorophenyl)ethylammonium iodide (FPEAI) during the preparation of (MAPbI
3)
0.75(FASnI
3)
0.25 films. In presence of the organic cation, highly oriented crystals are grown, and phase segregation occurring as the result of the Sn oxidation (emerging two distinguished signals from MAPbI
3 and FASnI
3 HPs), is suppressed. This fact makes that the recombination lifetime is longer than that of the absence of the organic molecules, associated with the reduction of high density of trap states. In context, a high PCE of 17.51% was reached, keeping 90% of the initial performance after 1200 h into a glove box, and 70% of initial efficiency close to 400 h under ambient air.
Then, Xu and coworkers
[168] employed the surface passivation concept to prolong the stability of the FA
0.7MA
0.3Pb
0.7Sn
0.3I
3 HPs and remove structural defects from the material using guanidinium bromide (GABr). Considering that GA
+ cations provide more charge density than FA
+ and MA
+ cations, this could be one of the main explanations to restrain the fast Sn
2+-to-Sn
4+ transformation. In addition, iodide vacancies are formed by the depletion of Sn
2+, creating high content of halide vacancies into the perovskite. Thus, Br
− domains fill/replace these empty states to enhance the final properties of the prepared active layer. The optimized device has reached a maximum PCE of 20.63% without any encapsulation, conserving over 85% of the original efficiency after 1000 h under ambient air, and 80% of the original efficiency after 24 h of operation at 80 °C. Lastly, Huang and coworkers
[160] have demonstrated that the addition of an alkylammonium pseudo-halide additive based on octylammonium tetrafluoroborate (OA
+BF
4−) for the preparation of MA-free Sn/Pb HPs. While OA
+ cations can fill some A-site cations into the HPs, BF
4− species compensate the iodide vacancies generated by the Sn oxidation, also reducing the likelihood of emerging interstitial iodide and iodine. The suppression of defect density, main factor to promote carrier trapping, facilitates the fabrication of PSCs with PCE up to 23.7% (a record for Sn/Pb-PSCs), keeping over 88% of the initial performance after 1000 h of continuous operation at 50 °C in air and tracking the device under MPP.
For the case of the stabilization of Sn-based PSCs, some examples can be addressed. Mora-Seró and coworkers
[169] have made use of the chemical engineering strategy, where Dipropylammonium iodide (DipI) and sodium borohydride (NaBH
4) (known as a good reducing agent) have been added to the preparation of FASnI
3 films and prevent the Sn oxidation. While DipI can block the loss of iodine in form of I
2, NaBH
4 reduce the released I
2 to produce iodide into the HPs, avoiding the halide migration. Therefore, the Sn
2+/Sn
4+ and I
−/I
2 ratios are increased, inhibiting the formation of interstitial iodine/iodide vacancies, which are the main energy states where the hole/electrons can be trapped. In this alternative for stabilizing Sn, optimized devices boost a high PCE of 10.61%, providing up to 96% of the initial performance after 1300 h of continuous operation at MPP in N
2 condition. Another approach to stabilizing the black phase of the FASnI
3, Mi and coworkers
[170] have introduced trimethylthiourea (3T) during the spin-coating process for HPs preparation. The presence of this bifunctional ligand induces coordination bonds with Sn
2+ cations through the formation of Sn-S species. Simultaneously, N-H moiety from the organic agent produces hydrogen bond iodide species from the octahedra building blocks to avoid their diffusion out from the HPs lattice. Due to these effects, the carrier recombination is prolonged in comparison to the pristine HPs, hindering the emergence of carrier traps. In this context, PSCs were able to develop a maximum PCE up to 14%, maintaining 100% of the original efficiency for more than 700 h, under N
2 atmosphere. On the other hand, He and coworkers
[171] have included 4-fluoro-phenethylammonium bromide (FPEABr) into the precursor solution to prepare 2D/3D Sn-HPs with low density of structural defects. FPEA cations are incorporated into the perovskite surface, while F- and Br-anions fill/replace halide vacancies. In terms of the device architecture, the 2D Sn-perovskite acts as an interface layer between the active layer and PEDOT:PSS hole transporter layer, restraining the iodide migration into the device. In presence of 10 wt% FPEABr, encapsulated 2D/3D PSCs generate a PCE up to 14.81%, exhibiting over 80% initial performance after 432 h under operation at room temperature, and 60% of initial efficiency after 1 h at 80 °C, both kinds of tests under ambient air.