Abstract

Organometal halides are promising materials for photovoltaic applications, offering tunable electronic levels, excellent charge transport, and simplicity of thin-film device fabrication. Two-dimensional (2D) perovskites have emerged as promising candidates over three-dimensional (3D) ones due to their interesting optical and electrical properties. However, maximizing the power conversion efficiency is a critical issue to improve the performance of these solar cells. In this work, we studied the photophysics of a two-dimensional (2D) perovskite (CH3NH3)2Pb(SCN)2I2 thin film using steady-state and time-resolved absorption and emission spectroscopy and compared it with the three-dimensional (3D) counterpart CH3NH3PbI3. We observed a higher bandgap and faster charge recombination in (CH3NH3)2Pb(SCN)2I2 compared to CH3NH3PbI3. This work provides an improved understanding of fundamental photophysical processes in perovskite structures and provides the guideline for the design, synthesis, and fabrication of solar cells.

1. Introduction

Solution-processable earth-abundant three-dimensional (3D) organometal trihalide perovskites have been intensively studied due to their unique electronic and optical properties [1, 2]. Their properties—including high charge carrier mobility, long exciton diffusion length, and low concentration of trap states, open broad prospects for their applications in a variety of industrial and technological areas including solar cells, light-emitting diodes, and field-field transistors [18]. Despite the high-power conversion efficiency of perovskite solar cells and high electroluminescence quantum yield of perovskite light-emitting diodes, poor long-term stability of these materials to moisture, light, and heat remains major obstacles barring widespread commercialization [9]. For this reason, identification and synthesis of alternative materials with similar optoelectronic properties are considerable.

Two-dimensional (2D) hybrid perovskites offer superior ambient stability along with local confinement, anisotropic dimensionality, and versatility of organic chemistry synthesis and flexibility for the modulation of optoelectronic properties [1012]. Therefore, these systems are very promising alternatives to 3D perovskites for photovoltaic and light-emitting diode applications [1319]. Most excitations in these 2D structures are excitons in contrast to free carriers in 3D perovskites and show contrasting properties compared to 3D structures [10]. Recently, 2D layered perovskite (CH3NH3)2Pb(SCN)2I2 has been synthesized, and power conversion efficiency (PCE), triplet generation, and device stability were examined [12, 15, 20, 21]. The PCE of (CH3NH3)2Pb(SCN)2I2 solar cells is far below that of the CH3NH3PbI3 perovskite solar cells [20], and the fundamental understanding of exciton and charge generation in these 2D structures is still lacking. Although lead thiocyanate (Pb(SCN)2I2) was used as an additive to improve the efficiency of 3D perovskites [22], its use as a part of a precursor to make 2D perovskites is very limited.

In this work, we studied the photophysics of a two-dimensional (2D) perovskite (CH3NH3)2Pb(SCN)2I2 thin film using steady-state and transient spectroscopy techniques and compared it with the three-dimensional (3D) counterpart CH3NH3PbI3 to understand the photophysical processes in these perovskites. This work is focused on understanding the fundamental properties of 2D perovskites and identifying the factors that limit the solar cell efficiency. We observed a higher bandgap (2.01 eV) and faster charge recombination in (CH3NH3)2Pb(SCN)2I2 compared to CH3NH3PbI3. Although the optimum bandgap photovoltaic device is 1.40 eV, there are several factors that limit the efficiency of solar cells including morphology of the active layer, charge generation and recombination, charge transfer to the interface layer, and charge transport [2325]. In this paper, charge generation and recombination in the active layer of (CH3NH3)2Pb(SCN)2I2 and CH3NH3PbI3 perovskites are discussed.

2. Materials and Methods

2.1. Synthesis

Methylammonium iodide (CH3NH3I), lead iodide (PbI2), lead thiocyanate Pb(SCN)2, N,N-dimethylformamide (DMF), dimethyl sulfoxide (DMSO), and chlorobenzene (CB) were purchased from Sigma-Aldrich. The precursor solution of (CH3NH3)2Pb(SCN)2I2 was prepared by dissolving 318 mg of CH3NH3I and 323 mg of Pb(SCN)2 in anhydrous DMF. 461 mg of PbI2 and 159 mg of CH3NH3I were dissolved in 1 mL DMF, and 78 μL DMSO was added. The precursor solutions were stirred overnight at room temperature. The glass substrates were cleaned ultrasonically using deionized water, acetone, and isopropanol for 15 min per cleaning solvent and subsequently air dried before spin casting. Thin films were prepared by spin casting the solution on glass substrates at 5000 rpm for 60 seconds in a glovebox with controlled nitrogen environment (<0.5 ppm).

2.2. Characterization
2.2.1. X-Ray Diffraction (XRD)

The XRD measurements were conducted on a Rigaku MiniFlex 600 X-ray diffractometer, and the scan range was 5-60°.

2.2.2. UV-Vis and Photoluminescence

Absorption measurements were carried out using a Shimadzu UV-Vis spectrometer, and steady-state photoluminescence was made using an Edinburgh Instruments FS920 fluorimeter. Time-resolved photoluminescence (TRPL) measurements were performed using a time-correlated single photon counting (TCSPC) spectrometer from Edinburgh Instruments (LifeSpec II) with a 4 MHz variable excitation laser source. The excitation fluence was kept below 1 μJ/cm2 to avoid any higher-order nonlinear processes. Morphological data were collected using a JEOL JSM-6510LV scanning electron microscope (SEM).

2.2.3. Transient Absorption

Transient absorption data were collected using transient absorption spectroscopy setup. This setup consists of the spectrometer (Ultrafast Helios System) and amplified Ti:saphhire laser. The output of amplified Ti:saphhire laser provides 800 nm fundamental pulses at 1 kHz repetition rate which were split into two optical beams to generate pump and probe pulses. One fundamental beam was used to generate a pump beam using an optical parametric amplifier (OPA) system (Coherent Opera Solo). A white light probe was generated by focusing another fundamental beam into a sapphire plate. Pump and probe beams were focused on a sample, and probe light was collected by a charge-coupled device (CCD). The spectral detection region is 450 nm to 800 nm. The thin-film samples were encapsulated using UV curable clue before measurement. The instrument response function (IRF) was ~100 fs FWHM. The samples were excited with the excitation energy 3.1 eV (400 nm), and the fractional change in transmission was detected in the probe range 450 nm to 800 nm at several time delays.

3. Results and Discussion

The crystal structure of both perovskites was analyzed by XRD. Figure 1 shows the XRD patterns of (CH3NH3)2Pb(SCN)2I2 and CH3NH3PbI3 perovskite thin films in the scan range 5-60°. The typical peaks of both (CH3NH3)2Pb(SCN)2I2 and CH3NH3PbI3 samples are within this range. Evenly spaced characteristic peaks at 9.4°, 19.0°, and 28.8° are observed in (CH3NH3)2Pb(SCN)2I2 which are assigned as [200], [400], and [600] planes and are attributed to the (CH3NH3)2Pb(SCN)2I2 phase. The peaks in both perovskites are consistent with the previously published results [26, 27]. The phase of as-synthesized (CH3NH3)2Pb(SCN)2I2 perovskite is orthorhombic at room temperature whereas it is tetragonal for CH3NH3PbI3 [28, 29]. Morphology of prepared thin films was characterized using scanning electron microscopy (SEM). CH3NH3PbI3 and (CH3NH3)2Pb(SCN)2I2 thin films show the different morphology. Figure S1 in supplementary file displays the SEM images of these perovskites which are similar to the images reported in the literatures. The CH3NH3PbI3 thin film shows the mesh-like structure whereas the (CH3NH3)2Pb(SCN)2I2 film has micron-size particles [27, 30, 31].

Figure 2 shows the UV-Vis absorption spectra of (CH3NH3)2Pb(SCN)2I2 and CH3NH3PbI3 perovskite thin films. The (CH3NH3)2Pb(SCN)2I2 film shows a sharp absorption peak at 586 nm (2.1 eV) which is attributed to the 1 s exciton absorption. In low-dimensional systems including 2D (CH3NH3)2Pb(SCN)2I2 perovskites, there is both quantum and dielectric confinement due to the difference in permittivity between the ionic perovskite layer and the bulky organic cation [32]. These effects enhance the electron hole correlations which are manifested in absorption spectrum of these systems as a sharp excitonic peak (Figure 2). In Figure 2(a) (inset), we show the plot of versus energy to obtain the correct value of bandgap of this perovskite. From extrapolation of the linear part of the Tauc plot [33, 34], the bandgap of (CH3NH3)2Pb(SCN)2I2 is estimated as 2.04 eV. This bandgap assignment of (CH3NH3)2Pb(SCN)2I2 is consistent with the reported values in the literature [21, 26, 30, 33, 35]. This bandgap is smaller than that of other typical single-layered Pb-I perovskites which is likely due to the reduced confinement of inorganic sheets that are induced by the shorter spacing and octahedral tilting in the inorganic sheets [36]. The Tauc plot of the CH3NH3PbI3 film (Figure 2(b), inset) indicates that the bandgap of this perovskite is 1.58 eV and is in close agreement with the previous reports [18, 37]. It has been reported that incorporation of SCN- ions in CH3NH3PbI3 induces more hydrogen bonds, more distorted Pb-I-Pb bond, and shorter Pb-S bond lengths [38] which can directly influence the band structure near the bandgap and result in higher bandgap in (CH3NH3)2Pb(SCN)2I2.

PL spectra measured using an Edinburgh Instruments FS920 fluorimeter are shown in Figure 3. We observed an emission main peak at 757 nm and additional low wavelength peak at 600 nm in (CH3NH3)2Pb(SCN)2I2. The higher energy PL peak at 600 nm is much weaker in intensity and is attributed to the reduced dimensionality of the layered perovskite structure. On the other hand, there is only one main emission peak at 773 nm in CH3NH3PbI3. The emission peaks at 757 nm and 600 nm are attributed to triplet and singlet excitons, respectively, in (CH3NH3)2Pb(SCN)2I2 whereas emission peak at 773 nm in CH3NH3PbI3 is due to the free carrier recombination [39]. The low-wavelength emission peak at 600 nm (2.06 eV) in (CH3NH3)2Pb(SCN)2I2 suggests that its bandgap is close to 2.06 eV and is consistent with the bandgap estimated from the Tauc plot (Figure 1(a)). We fitted the main PL peak using the Gaussian distribution and measured the full width at half maximum (FWHM) on both films to check the color purity. The FWHM of the main peak of (CH3NH3)2Pb(SCN)2I2 perovskite is 39 nm whereas it is 36 nm for CH3NH3PbI3 indicating slightly better color purity in CH3NH3PbI3 [40, 41]. We also measured the Stokes shift in these perovskites to probe the energy loss through molecular vibrations. The Stokes shift is very similar in both perovskites (~50 meV) indicating similar vibrational energy loss.

Time-correlated single photon counting (TCSPC) was performed to measure charge/exciton dynamics. Figure 4 shows the PL dynamics of (CH3NH3)2Pb(SCN)2I2 and CH3NH3PbI3 perovskite thin films. It is interesting that the main emission peak of two perovskites exhibits contrasting PL decay dynamics. The characteristic PL lifetimes for the (CH3NH3)2Pb(SCN)2I2 main emission peak at 757 nm, extracted by a double exponential function, are 7.5 ns (54%) and 29.8 ns (46%) whereas the PL lifetimes at 600 nm is 220 ps. This result also confirms that the emission at 600 nm is due to singlet exciton while the emission at the main peak (757 nm) is due to triplet excitons. The observation of very short PL emission lifetime of 600 nm peak and long lifetime PL emission lifetime of 757 nm peak indicates that the recombination in (CH3NH3)2Pb(SCN)2I2 is dominated by excitonic recombination, and the 757 nm peak is associated with triplet excitons [19]. It has also been reported that this peak may be related to triplet excitons or defective states [19]. The characteristic PL lifetimes of CH3NH3PbI3 at 773 nm are 1.9 ns (56%) and 7.8 ns (44%). These results indicate that emissive peaks of two perovskites have contrasting lifetimes. It has been reported that radiative recombination of free electrons and holes is dominant for PL processes in CH3NH3PbI3 and the exciton model is not appropriate for this material at room temperature [4244]. Therefore, CH3NH3PbI3 structure is different from (CH3NH3)2Pb(SCN)2I2 whose recombination dynamics is dominated by excitons [19, 45].

In order to monitor the exciton and charge generation dynamics, we performed femtosecond transient absorption spectroscopy (TAS) on these perovskite structures. While PL is mostly sensitive to emissive species, TAS can provide information about charges and excitons [39, 46]. Figure 5 shows the transient absorption spectra at different time delays after the samples are excited using pump pulses tuned to 4.1 eV (400 nm). There are three different features contributing to the transient spectra. We assign the positive band to ground state bleaching (GSB) and the negative band to a photoinduced absorption (PA) [47, 48]. We observed a sharp positive band due to GSB at 580 nm for (CH3NH3)2Pb(SCN)2I2 whereas this positive band for CH3NH3PbI3 is at 750 nm. We did not observe the GSB feature of ~750 nm in (CH3NH3)2Pb(SCN)2I2. These results clearly indicate that the bandgap of (CH3NH3)2Pb(SCN)2I2 is significantly higher than that of CH3NH3PbI3 and further supports the conclusion from UV-Vis and PL measurements. We also observed a broad weak band in the 600-800 nm range and is possibly due to the defects/impurities present in (CH3NH3)2Pb(SCN)2I2.

Figure 6 shows the ground state bleaching dynamics of (CH3NH3)2Pb(SCN)2I2 and CH3NH3PbI3 perovskite thin films. From biexponential data fitting, we computed the time constant of GSB dynamics. The time constants are 32 ps (48%) and 460 ps (52%) for (CH3NH3)2Pb(SCN)2I2 and 33 ps (42%) and 719 ps (58%) for CH3NH3PbI3. This suggests that early dynamics for both perovskites are the same whereas the long-time components are different. Interestingly, the long-time component of CH3NH3PbI3 is higher than that of (CH3NH3)2Pb(SCN)2I2 in contrast to the PL dynamics. This suggests that the decay of radiative species monitored through PL is different from the decay photoinduced charged carriers monitored through GSB decay. The faster dynamics of GSB in (CH3NH3)2Pb(SCN)2I2 is possibly due to the defects/impurities present in (CH3NH3)2Pb(SCN)2I2.

4. Conclusion

In this work, we prepared CH3(NH3)2Pb(SCN)2I2 2D perovskite and CH3NH3PbI3 3D perovskites. Using steady-state and transient absorption and emission spectroscopies, we studied the exciton and charge generation in these structures. We observed a higher bandgap and faster charge recombination in (CH3NH3)2Pb(SCN)2I2 compared to the CH3NH3PbI3. In addition, excitonic recombination is dominant in PL decay (decay of radiative species) of (CH3NH3)2Pb(SCN)2I2 whereas it is free carrier recombination in CH3NH3PbI3. We also observed the presence of defects/impurities in (CH3NH3)2Pb(SCN)2I2, as evidenced by transient absorption spectra. The low power conversion efficiency of (CH3NH3)2Pb(SCN)2I2 solar cells compared to that of CH3NH3PbI3 is possibly due to the faster charge recombination as observed in TAS. This study will greatly facilitate fundamental understanding of structure-property relations in the hybrid perovskite structures.

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work is supported by NSF PREM in collaboration with the Center for High Resolution Neutron Scattering (CHRNS) through award DMR 1827731. Dr. Zhiping Luo and Dr. Gibin George from Fayetteville State University are acknowledged for providing the supplies through IMREL facility, and Dr. Rachel Wells is acknowledged for helping in SEM image measurements. Dr. David Hoogerheide, Dr. Julie Borchers, and Dr. Dan Neumann at the National Institute of Standards and Technology (NIST) Center for Neutron Research are acknowledged for the fruitful discussion in the project.

Supplementary Materials

Figure S1. SEM images of (a) (CH3NH3)2Pb(SCN)2I2 and (b) CH3NH3PbI3 thin films. (Supplementary Materials)