Efficient light harvesting in perovskite layer via three-dimensional TiO₂ nanobranched nanorod scaffold

High photovoltaic efficiency, low production cost, and easy fabrication by solution processes make inorganic halide perovskite solar cells (PSCs) the most promising candidate among various types of solar cells [1–8]. A typical PSC with a p-i-n junction configuration includes five layers. A light absorbent perovskite layer places between an n-type electron transfer layer (ETL) and a hole transport layer (HTL). ETL and HTL layers are generally TiO₂ and p-type organic semiconductor (spiro-OMeTAD), respectively. At i–n and p–i interfaces, the separation of the photogenerated charges occurs through in situ electrical field. Photogenerated electrons and holes collect by the FTO and cathode (e.g., Ag) layers, respectively [9–13]. Inadequate light absorption, short diffusion lengths of charge carriers, and charge recombination are the most significant reasons for limited power conversion efficiency of solar cells [14]. High specific surface area and fast charge transport are preferable features for solar cells [15].

Mesoporous nanoparticles based scaffold provides large surface area but suffers from poor electron mobility. On the other hand, one-dimensional nanomaterials such as nanorods and nanowires represent rapid electron conducting. However, their low surface area compensates their favorable advantage. Due to their remarkable specific surface area along with high-speed pathways for charge carriers, three-dimensional (3D) hierarchical nanostructures are outstanding candidates for photoelectrochemical application [16–19]. Daneshvar e Asl and Sadrnezhaad [20] successfully synthesized TiO₂ nanoleafed nanorods thin film, and evaluated it as a novel-designed photocatalyst for the photodegradation of anionic and cationic dyes. The improved photocatalytic performance of TiO₂ is ascribed to large surface area and desirable light harvesting. Chen et al [21] reported a novel facile and one-step synthesis of hierarchical TiO₂ nanotubes, and used this nanostructure in fabrication of the photoelectrode of the dye-sensitized solar cell. Considerable obtained efficiency of 9.89% is attributed to directed charge transport, excellent light harvesting, and effective charge collection. Li et al [22] successfully prepared stainless steel mesh-supported 3D hierarchical SnO₂/Zn₂SnO₄ composite for gas sensor, solar cell, and photocatalysis applications. Xu et al [23] synthesized sheet-like CdS sensitized ZnO film by hydrothermal method. The photoelectrochemical property enhancement is ascribed to the improved light absorption and enhanced charge separation and transportation.

In this paper, TiO₂ mesoporous nanoparticles, nanorods, and nanobranched nanorods thin films were synthesized and utilized as the scaffold for perovskite coating for further use in PSCs. The performance of the thin films were compared in terms of charge carrier extraction, charge transport, and solar energy harvesting through photoluminescence and UV–vis spectroscopies. Perovskite-coated TiO₂ nanobranched nanorods thin film showed the best photovoltaic performance due to desirable characteristics of nanobranched nanorods thin film.

2.1. Materials

2.2. Synthesis

A patterned FTO substrate was prepared via chemical etching with hydrochloric acid and zinc powder. TTIP was mixed with ethanol to make the solution precursor of compact TiO₂ (C-TiO₂) film. C-TiO₂ film were coated on the patterned FTO by spin coating under 2000 rpm for 60 s followed by heating at 80 °C and 500 °C for 10 min and 30 min, respectively. Then, three types of TiO₂ scaffold were deposited on the C-TiO₂ film. Mesoporous nanoparticles (Meso) film was prepared by spin coating TiO₂ paste diluted in ethanol with weight ratio of 1:7 under 4000 rpm for 30 s. Nanorods (NR) and nanobranched nanorods (NRNB) thin films were synthesized on top of C-TiO₂ film via the hydrothermal and aqueous chemistry methods according to our previous work [24]. The coated TiO₂ thin films were annealed at 450 °C for 30 min. For perovskite layer coating, a solution of lead iodide in DMF (1 M) with 70 °C temperature was spin coated on the TiO₂ scaffold at 6500 rpm for 5 s, and dried at 100 °C for 30 min. Then, the layers were exposed to MAI powder in the oven with 150 °C temperature for 2 h. After cooling in ambient temperature, the layers were cleaned by iso-propanol, and then heated at 140 °C for 30 min. The synthesis procedure are summarized in Scheme 1.

Zoom In Zoom Out Reset image size

2.3. Instrumentations and characterizations

FESEM analysis was employed to study the effect of TiO₂ scaffold morphology on the quality of upper perovskite layer. It was thus possible to understand the correlation between crystal growth behavior and grain size with perovskite layer quality. FESEM images of three types of TiO₂ thin film, i.e., Meso, NR, and NRNB, as well as perovskite layer coated on them are shown in figure 1. As it is clear in figures 1(a)–(c), TiO₂ thin films are crack-free and monotonously covered the substrate surface. The trunk of NRNB structure could provide a fast charge pathway to improve electron collection and probably decrease hysteresis effect compared with Meso structure. In analogy with NR arrays, the 3D morphology of NRNB make much higher volume density available for the nucleation of perovskite layer. Moreover, this structure could increment the optical pathways through light scattering. Also, NBs present in the space between NRs could extract photogenerated electrons from MAPbI₃ to overcome the hindrance of short electron diffusion length in the MAPbI₃. Figures 1(d)–(f) shows surface images of perovskite layers prepared onto different TiO₂ scaffolds. Perovskite layer formed on TiO₂ NRNB is more compact and has fewer voids than those obtained by TiO₂ Meso and NR scaffolds. This observation may be attributed to the slower rate of inter diffusion and reaction between the MAI vapor and the PbI₂ film in presence of TiO₂ NRNB. In both TiO₂ NRNB and Meso, nucleation and growth of perovskite is greatly affected by the scaffold. The coating behavior of the perovskite layer is improved by using TiO₂ NRNB. A perovskite capping layer with high coverage helps enhance light absorption and improve the performance of the solar cell. A large proportion of voids can be seen in perovskite layer that was processed by TiO₂ NR. In this sample, there is a high probability of contact between the ETL and HTL. As a result, charge carrier species, i.e., electrons and holes, may recombine within the perovskite layer, and it decreases efficiency of the solar cell.

Zoom In Zoom Out Reset image size

For a complete study of the nature of perovskite layers coated on different TiO₂ scaffolds, XRD analysis was employed (figure 2). Similarity in the XRD patterns between perovskite samples prepared on various TiO₂ scaffolds indicates that all scaffolds produce the same perovskite. The 2θ values of 31.8°, 28.4°, and 14.1° are the main peaks for perovskite crystals. These values correspond to (310), (220), and (110) lattice planes of tetragonal perovskite crystalline structure, respectively [25–28]. Also, as pointed out in figure 2, a peak is observed at 12.7°, corresponding to the (001) lattice plane. This peak relates to the not-reacted PbI₂. In other words, the conversion to perovskite is incomplete. The strongest peak for residual PbI₂ is observed in perovskite layer with NR scaffold. The sample prepared in the presence of NRNB scaffold exhibits the weakest peak of PbI₂, indicating a near complete transformation to perovskite. A large fraction of residual PbI₂ in perovskite layer is expected to be harmful to the solar cell efficiency. This is mainly results from the electrical insulating nature of PbI₂ [29, 30].

Zoom In Zoom Out Reset image size

To determine the optical properties of perovskite layers deposited on various TiO₂ thin films, UV–vis absorption spectra were acquired. The absorption range from 400 nm up to 750 nm is shown for all prepared layers in figure 3, which indicates the formation of perovskite on the TiO₂ scaffolds. The UV–vis absorption is initially improved in presence of TiO₂ NRNB scaffold. Generally, full coverage and compact perovskite layer can increase the UV–vis absorption [31, 32], which can approve the highest UV–vis absorption for perovskite layer deposited on top of TiO₂ NRNB scaffold. A decline in the light absorption is seen for the TiO₂ NR-based perovskite layer, which derives from the partially coverage of perovskite layer with higher voids.

Zoom In Zoom Out Reset image size

To comprehensive study of the optical properties of perovskite layers, Pl spectrums of the perovskite-TiO₂ films were measured. In general, the charge carrier extraction in the perovskite layer is mainly dependent on the perovskite layer morphology. Figure 4(a) displays the steady-state Pl spectrums of perovskite layers deposited on different TiO₂ films. The Pl intensity decrease indicates effective charge carrier extraction happens within the perovskite-TiO₂ films [33–35]. The higher Pl intensity of perovskite-TiO₂ NR is resulted from an inhomogeneous perovskite layer with a high number of voids. A low Pl intensity can be seen for perovskite-TiO₂ NRNB, that is attributed to high coverage and compactness of perovskite layer. Thus, the steady-state Pl analysis indicates that the high charge carrier extraction efficiency occurs in the uniform and smooth perovskite layer. To study the charge carrier extraction/transfer behavior, the time-resolved Pl decay spectra of different perovskite-TiO₂ films were measured, as presented in figure 4(b). The shortest Pl decay lifetime can be seen for the perovskite-TiO₂ NRNB scaffold, which shows the faster charge carrier transfer into ETL and HTL. An efficient and fast charge carrier extraction and transfer process into charge carrier layers results from the compact and unharmed perovskite layer-TiO₂ scaffold interface. An inhomogeneous and partially coverage perovskite layer with voids can provide recombination sites, which leads to an insufficient charge carrier extraction/transfer. The Pl decay lifetimes for the perovskite layers prepared onto different TiO₂ scaffolds were 14.11, 17.21 and 19.99 for TiO₂-NRNB, TiO₂-Meso, and TiO₂-NR, respectively. It is well-known that the charge carriers generated in the perovskite domain can be injected into ETL, and the PL of perovskite layers occurs due to the recombination of localized charge carriers. The spectrum of perovskite layer prepared with TiO₂-NRNB substrate shows a significant depressed PL (14.11 ns) compared with that of the layers with TiO₂-Meso and TiO₂-NR (17.21 ns and 19.99 ns, respectively), indicating that the charge carriers generated can be effectively generated and extracted. In fact, the diffusion length of charge carriers is increased by 21% and 40% in presence of TiO₂-NRNB substrate compare with TiO₂-Meso and TiO₂-NR, respectively.

Zoom In Zoom Out Reset image size

Three types of TiO₂ thin film, i.e., mesoporous nanoparticles, nanorods, and nanobranched nanorods, were synthesized and used as the scaffold for perovskite coating for further use in the perovskite solar cell. Hierarchical nanobranched nanorods thin film presented better optical properties compared with thin films of mesoporous nanoparticles and nanorod arrays. In fact, the large surface area and charge transport property of nanobranched nanorods led to uniform and compact perovskite loading, excellent electron collection, and desired light harvesting performance, which are preferable features for promoting the efficiency of the perovskite solar cells.

This work was partially supported by the Iran National Science Foundation.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

The data that support the findings of this study are available upon request from the authors.