Synthesis of Asymmetrical CsPbBr₃/TiO₂ Nanocrystals with Enhanced Stability and Photocatalytic Properties

Abstract

Practical applications of CsPbX₃ nanocrystals (NCs) are limited by their poor stability. The formation of a heterojunction between CsPbX₃ NCs and oxides is an effective means to protect perovskite from polar solvents and other external factors. Significantly improving the stability and photocatalytic properties of the core/shell perovskite is very important for its application in photoelectric and photocatalytic technology. Here, we report the synthesis of asymmetrical CsPbBr₃/TiO₂ core–shell heterostructure NCs at the single-particle level by hot-injection liquid-phase synthesis and sol–gel method, where each CsPbBr₃ NCs is partially covered by titanium dioxide. We tested not only the optical properties of the material but also the electrochemical impedance and photocurrent density of the material in sodium sulfate solution. It is shown that the type II arrangement is generated at the heterogeneous interface, which greatly facilitates the separation of electron–hole pairs and increases the carrier transport efficiency. Compared with CsPbBr₃ NCs, CsPbBr₃/TiO₂ has higher photocatalytic efficiency. More crucially, due to the protection of the titanium dioxide shell, the product has higher long-term stability in humid air compared with bare CsPbBr₃ NCs. The asymmetrical core–shell heterostructure prepared in this study not only improves the stability of CsPbX₃ NCs but also provides some ideas for optoelectronic device applications and TiO₂-based photocatalysts.

1. Introduction

CsPbBr₃ is a typical metal halide perovskite material that has become one of the most promising materials for optoelectronic applications, including solar cells [1,2,3,4], light emitting diodes (LED) [5,6,7,8,9,10], photodetectors [11], and lasers [12,13,14,15,16], due to its excellent photoelectric and catalytic properties such as narrow optical emission range, high photoluminescent quantum yield (PLQY), high photogenerated charge mobility, and tunable band gap. In recent years, the rapid development of research in this field has led to significant advances in the study of optoelectronic properties and materials. Inspired by the results of photovoltaic applications, CsPbBr₃ nanocrystals (NCs) is a potential candidate for conducting efficient photocatalysis [17,18]. Since the pioneering work reported by Kovalenko et al. [19] in 2015, significant progress has been made in the preparation methods and photocatalytic applications of CsPbX₃ NCs (X = Cl, Br, I).

With the growth in population and economic activities, the increasing demand for water resources and the aggravation of water pollution have further promoted water shortage on a global scale. In the case of abundant seawater resources, the use of catalysts to purify seawater can not only alleviate the shortage of freshwater resources but also greatly reduce the cost of the reaction system process. Solar-powered photocatalysis is one of the effective means to achieve green sustainable fuel preparation and high value-added chemical production in the future [20]. Halide perovskites have become a new type of photocatalyst material due to their advantages, such as high absorption efficiency, long carrier migration distance, and easy adjustment of component structure and morphology. Wang and coworkers reported the preparation of CsPbBr₃ NCs, and the obtained CsPbBr₃ NCs showed superior photocatalytic activity and excellent stability in a mixed medium of water/methanol vapor [21].

However, perovskite is often described as a crystalline liquid due to the soft and dynamic properties of the crystal lattice, so light-induced carriers are less likely to be trapped and scattered. At room temperature, polarons are formed by the coupling of electrons and holes caused by the ionic properties and strong structural dynamics of the PbX lattice, thus shielding the Coulomb potential and reducing the charge capture and scattering between polarons and between polarons and charged defects or optical phonons. Although there is a high defect tolerance in the lead halide perovskite NCs materials, its defects will still reduce its luminous performance. The high photoluminescence quantum yield of perovskite can only achieve 80–90% in the green–red fluorescence range of the spectrum. In addition, as an ionic crystal, perovskite NCs is not very stable in the process of purification and storage under harsh environmental conditions (such as air, polar solvents, heat, and light). They usually decompose or dissolve quickly, which greatly limits their practical application in photoelectric devices, catalysis, and other fields. The interaction between ligands and atoms on the perovskite NCs surface is highly dynamic. During purification, surface ligand shedding caused by polar solvents leads to an increase in nanocrystal defects, resulting in decomposition and loss of fluorescence properties [22]. In recent years, researchers have proposed many methods to solve the worldwide problem of poor stability of metal halide perovskite, such as surface modification, polymer encapsulation, silicon coating, atomic layer deposition passivation, etc. Imran et al. [23] reported the synthesis of asymmetric CsPbX₃–Pb₄S₃Br₂ NCs heterojunction by two-step direct synthesis using preformed subnanometer CsPbBr₃ clusters. The existence of an epitaxial interface of Pb₄S₃Br₂ material increases the structural rigidity of the perovskite lattice and improves the stability of perovskite material. Although surface passivation by organic ligands or polymers can significantly improve the stability of perovskites, these insulating shells severely limit the charge transport of metal halide perovskite themselves for optoelectronic, catalytic, and other applications.

In addition to polymers [24,25,26,27,28,29,30,31] and inert shell layers [16,32,33,34,35,36], semiconductors were also introduced to improve the stability and photophysical properties of perovskite. More importantly, by using another semiconductor with different conduction and valence bands, one can easily tune the bandgap or charge transfer processes and thus improve photophysical properties. Prior to shell growth of perovskite NCs, shell materials were carefully selected to design the core–shell structure to allow for favorable band arrangement. In type I core–shell structures, the shell material has a wider energy band gap than the core, so the edges of the core material of the conduction and valence bands are located in the energy gap of the shell, having a crossing band arrangement. The band gap of the core material determines the band gap energy of this heterogeneous structure, while the shell provides passivation of surface defects and limits electrons and holes within the core. This can improve the overall optical properties and stability of NCs [37]. Type II structures have a staggered arrangement in which the edges of the valence and conduction bands of the core material have lower or higher energies than the edges of the shell material. The band gap of type II structures is determined by the energy separation between the conduction band edge of one semiconductor and the valence band edge of another semiconductor. Carriers are spatially separated in these structures, thus offering potential benefits for applications, such as optoelectronics and photocatalysis. TiO₂, as a chemically inert and nontoxic photocatalyst, is the most important material used in seawater desalination and wastewater treatment [38,39,40]. Due to its rapid electron–hole recombination, forming a type II heterojunction between CsPbBr₃ and titanium dioxide can not only solve the problem of stability but also improve the charge transport efficiency of perovskite, laying the foundation for its photoelectric and photocatalytic applications.

The sol–gel method has been widely used for the modification of perovskite nanoparticles with various oxide materials, such as silica and titanium dioxide. However, since CsPbBr₃ NCs are very sensitive to water or temperature and fluorescence of perovskites quenching occurs before the formation of silica or titanium oxide shell layers, it is challenging to modify their surfaces using conventional sol–gel methods. In the available reports, Li et al. [41] demonstrated that TiO₂ shell coating improved the stability of perovskite but resulted in partial decomposition of CsPbBr₃ NCs or the dissolution of corners and edges during hydrolysis and calcination. Ji et al. [42] studied the synthesis of highly stable CsPbBr₃/TiO₂ nanocomposites by a top-assisted low-temperature solvothermal method. Although the material improves the stability of the perovskite NCs and preserves the edge of the perovskite core from being destroyed, the agglomeration is severe and the size is large, which poses obstacles for the photoelectric and photocatalytic applications. For example, in perovskite light-emitting diode (PeLED) applications, small-size particles facilitate the formation of thin films with high uniformity. Therefore, we are eager to develop a method that can achieve heterogeneous coating of CsPbBr₃ with TiO₂ at the single-particle level to improve stability.

Herein, we certify the synthesis of an asymmetric CsPbBr₃/TiO₂ core–shell nanocrystals heterojunction at the single-particle level in anhydrous solvent and at low temperature, aiming to improve the stability of perovskite, thereby improving its photoelectric and photocatalytic performance. Based on the band alignment, it is confirmed that the obtained CsPbBr₃/TiO₂ heterojunction NCs has a type II heterostructure with enhanced carrier transport efficiency. It is demonstrated that the CsPbBr₃/TiO₂ NCs show good optoelectronic properties and enhanced water stability under water test conditions, which provides a reference value for the preparation of high-efficiency, long-life photovoltaic and luminescent devices.

2. Results and Discussion

Firstly, cesium precursor (cesium oleate) was prepared by using cesium carbonate as cesium source, adding oleic acid and trin-octylphosphine ligand as raw materials. Then, lead bromide precursor was prepared by adding octadecene, oleic acid, and oleylamine to dissolve lead bromide powder. Cesium oleate was injected into the lead bromide precursor by hot-injection method. After a few seconds of reaction, the reaction solution was cooled to room temperature in an ice bath and centrifugally washed to obtain CsPbBr₃ NCs. Next, we chose titanium butoxide as the titanium precursor, slowly dropped the diluted titanium precursor into the toluene solution of CsPbBr₃ NCs at a certain speed, and stirred the solution at room temperature for 3 h. The hydrolysis reaction occurred, so that TiO₂ was deposited on the surface of CsPbBr₃ NCs, and then the reaction solution was placed in a Teflon-lined stainless-steel autoclave for 10 h. Finally, the CsPbBr₃/TiO₂ light yellow powder was obtained by centrifugation and drying (see Section 3 for details).

The morphologies of CsPbBr₃ NCs and asymmetrical CsPbBr₃/TiO₂ NCs were analyzed by transmission electron microscopy (TEM). As shown in Figure 1A, the CsPbBr₃ NCs materials exhibit highly uniform monodispersion with clear edges and uniform size, with an edge length of about 17 nm, indicating the preparation of high-quality perovskite nanocrystals. Figure 1B is the TEM image of asymmetrical CsPbBr₃/TiO₂ core–shell NCs. It can be obviously observed that each nanocrystalline particle has two regions with different image contrast, indicating that the material is composite structure. CsPbBr₃/TiO₂ NCs are composed of two different structural components, CsPbBr₃ and TiO₂, with an obvious heterogeneous interface, in which the dark part is CsPbBr₃ NCs, and the light part is TiO₂, which further indicates the formation of core–shell heterojunction. Importantly, after the titanium oxide coating, the edges of the perovskite are not dissolved, achieving precise coating at the single-particle level, as well as the distribution of these elements.

The crystal structures of CsPbBr₃ NCs and CsPbBr₃/TiO₂ NCs were characterized using an X-ray diffractometer. As shown in Figure 2 for the X-ray diffraction analysis (XRD) of CsPbBr₃ NCs and CsPbBr₃/TiO₂ NCs, the XRD peaks of CsPbBr₃ NCs appeared at 2θ = 15.2°, 21.6°, 24.1°, 26.3°, 30.5°, 30.8°, 34.5°, 38.0°, 43.9°, 46.7°, and 54.4°, corresponding to (001), (010), (110), ($\overline{2}$02), (210), (002), ($\overline{1}$12), (102), (020), (021), and (401) diffraction planes, respectively. The diffraction peaks of CsPbBr₃ NCs match well with the standard monoclinic perovskite phase (PDF#00-054-0751) of CsPbBr₃ NCs, with splitting diffraction peaks at 2θ ≈ 30°. For CsPbBr₃/TiO₂ NCs, in addition to the XRD peak observed for the monoclinic phase of CsPbBr₃ NCs, an additional weak diffraction peak (marked by blue circles in the Figure 2) can be observed at 2θ ≈ 25°, which corresponds to the (101) diffraction plane of anatase TiO₂ (PDF#00-021-1272). The above analysis indicates the formation of anatase type TiO₂ on the surface of CsPbBr₃ NCs. The lower intensity of the XRD peak of titanium dioxide relative to the CsPbBr₃ NCs diffraction peak indicates that the titanium dioxide is partially crystallized during the hydrothermal treatment and has weak crystallinity.

Since most applications of perovskite are related to its photoelectric and catalytic properties, it is desirable that materials have both environmental stability and efficient charge transport. To gain more insight into the material charge transfer kinetics, we recorded the photoluminescence (PL) emission spectra and time-resolved photoluminescence spectra (TRPL) of CsPbBr₃ NCs and asymmetrical CsPbBr₃/TiO₂ NCs (Figure 3). As shown in Figure 3A, the photoluminescence emission peaks of CsPbBr₃ NCs and CsPbBr₃/TiO₂ NCs at about 522 nm and 528 nm, respectively, can be observed. Compared with CsPbBr₃ NCs, the PL peak of CsPbBr₃/TiO₂ NCs is slightly red-shifted by 6 nm. After TiO₂ shell coating, the PL intensity decreases. Figure 3B shows the TRPL of CsPbBr₃ NCs and asymmetrical CsPbBr₃/TiO₂ NCs. The fitted decay curves reveal that the CsPbBr₃/TiO₂ NCs (τ = 1.73 ns) have shorter carrier lifetime compared with CsPbBr₃ (τ = 6.08 ns), and the decreased PL lifetime indicates improved charge diffusion and separation, i.e., the rapid charge transfer from CsPbBr₃ NCs to the TiO₂ shell, thus leading to enhanced photovoltaic performance. The reduced fluorescence lifetime further manifests that after TiO₂ shell coating, the CsPbBr₃/TiO₂ core/shell NCs generate a new nonradiative path, probably an electron transfer from the conduction band of CsPbBr₃ NCs to the conduction band of titanium dioxide, indicating the formation of type II heterojunctions, which greatly facilitates the separation of electron–hole pairs and CsPbBr₃/TiO₂ NCs with higher carrier mobility [43].

To confirm our hypothesis, we investigated the electronic energy band structures of CsPbBr₃ NCs and asymmetrical CsPbBr₃/TiO₂ NCs using ultraviolet–visible absorption spectroscopy (UV–Vis). Compared with the normal hydrogen electrode (NHE), the conduction band and valence band edge of CsPbBr₃ NCs are ≈ −1.07 eV and ≈1.35 eV, respectively. The conduction and valence bands of CsPbBr₃ are higher than those of titanium dioxide. Therefore, a type II band arrangement can be formed in the CsPbBr₃/TiO₂ heterostructure NCs. In this case, the lowest energy hole should be confined to the CsPbBr₃ nucleus, and electrons can delocalize on the CsPbBr₃ nucleus and the titanium dioxide shell, thus promoting charge transfer and reducing PLQY. From the UV–Vis absorption spectra (Figure 4A), it is clear that CsPbBr₃ NCs and CsPbBr₃/TiO₂ NCs showed significant absorption peaks in ultraviolet range. Based on Einstein’s photoelectric effect and the Kubelka–Munk equation [44], the surface optical band gaps of CsPbBr₃ NCs and CsPbBr₃/TiO₂ NCs were estimated to be 2.42 eV and 2.40 eV(Figure 4B), respectively, by extrapolating the linear region of the absorption edge in the UV–Vis spectrum to the energy axis intercept. The reduction in the electron transition band gap is consistent with the fluorescence lifetime results, which further proves that the charge transport characteristics of perovskite nanocrystals are optimized after titanium oxide coating. The work function (Φ) is another important parameter in the study of electron transfer in duplicate semiconductor heterojunctions. It can be calculated from the energy difference of the surface valence band maximum and the Fermi level of the material. According to the above data, the work functions of anatase TiO₂ and CsPbBr₃ NCs are calculated to be 6.0 eV and 4.58 eV, respectively, indicating that the Fermi level of anatase TiO₂ is lower than that of CsPbBr₃ NCs. When they come into contact, electrons flow from CsPbBr₃ to anatase titanium dioxide, making the same Fermi level consistent and producing internal electric field at the CsPbBr₃/TiO₂ NCs interface. These results are consistent with the results of the fluorescence lifetime test. The coating of titanium oxide is beneficial to charge separation in perovskite and improves the catalytic activity.

Based on the energy levels of CsPbBr₃ NCs and TiO₂ NCs, we can obtain the band structure of the sample. Obviously, it was found that a type II nanocomposite was formed, in which both the conduction band and valence band levels of the semiconductor CsPbBr₃ were higher than those of the semiconductor titanium dioxide. Once irradiated by light (λ = 546.5 nm), electrons in the valence band of CsPbBr₃ leap to the conduction band and form holes in their original positions. Under the action of the internal electric field, the carriers generated and leapt to the depletion region will be separated rapidly in the heterojunction, where the electrons will move toward the TiO₂ conduction band and the holes will transport to CsPbBr₃, and the photocurrent output due to the photoresponse can be obtained through the electrochemical workstation. The holes from the TiO₂ valence band will be transferred to the CsPbBr₃ valence band under the action of the external electric field, and the electrons will migrate in the opposite direction, thus producing a stable photoresponse, and the optical signal will be converted into an electrical signal.

The electrochemical impedance spectra (ESI) and the photoresponsive current densities of CsPbBr₃ NCs and CsPbBr₃/TiO₂ NCs materials were tested as shown Figure 5. The charge transport properties of CsPbBr₃ NCs and CsPbBr₃/TiO₂ NCs were further investigated by photoelectrochemical studies. The Nyquist plot is generally used in EIS and the Nyquist plot of the most common equivalent circuit model is semicircular in shape. In the equivalent circuit model, Cpe represents the double-layer capacitance, Rs represents the solution resistance from the reference electrode to the working electrode, and Rc is the charge transfer resistance of the electrode. The diameter of the semicircle is equal to the carrier migration resistance, and the smaller the diameter of the semicircle, the larger the carrier migration rate. As shown in Figure 5A, the curvature of the impedance spectrum of CsPbBr₃/TiO₂ NCs is significantly greater than that of CsPbBr₃ NCs. Additionally, because the curvature of the circle is inversely proportional to the radius, the CsPbBr₃/TiO₂ NCs has a smaller semicircle arc compared with CsPbBr₃ NCs. In other words, CsPbBr₃/TiO₂ NCs have a smaller carrier migration resistance than CsPbBr₃ NCs. To sum up, with the formation of TiO₂ shells, the charge transfer resistance of the CsPbBr₃/TiO₂ NCs is significantly reduced. In order to record the photocurrent generated by the photoresponse and exclude the effect of other factors, the light source is usually switched on at certain intervals to record the photocurrent profile. As shown in Figure 5B, when the light source was turned off, the recorded photocurrent density was almost zero; when the light source was turned on after 30 s, the current instantaneously increased; and when the light source was turned off at an interval of 30 s, the current instantaneously decreased. Compared with CsPbBr₃ NCs, CsPbBr₃/TiO₂ NCs have higher current density. This result further confirms that the formation of heterostructure between CsPbBr₃ and TiO₂ improves the carrier migration efficiency of the perovskite material and has better charge transport properties.

3. Materials and Methods

3.1. Materials

The cesium carbonate (Cs₂CO₃, 99.9% trace metals basis, Aladdin), oleic acid (OA, 80–90%, Aladdin), oleylamine (OAm, 80–90%, Aladdin), titanium butoxide (C₁₆H₃₆O₄Ti, TBOT, >99.0%, Aladdin), trinoctylphosphine (TOP, 90%, Aladdin), 1-octadecene (ODE, >90% (GC), Aladdin), lead (II) bromide (PbBr₂, 99.0%, Aladdin), and toluene were purchased from Sinopharm Chemical Reagent Co. All reagents labeled Aladdin are from Aladdin Reagent (Shanghai) Co., Ltd., Shanghai, China.

3.2. Methods

Preparation of Cesium Oleate: In a typical synthesis, 400 mg of Cs₂ CO₃ (1.23 mmol), 15 mL ODE, 2 mL TOP, and 1.25 mL oleic acid were added into a 25 mL three-neck round bottom flask connected with a double row of tubes and dried under N₂ at 120 °C for 1 h. Cesium carbonate powder completely dissolved, and the mixed solution became transparent, indicating that cesium carbonate and oleic acid completely reacted to produce cesium oleate (Cs-OA) solution. According to the purity of cesium carbonate, the purity of cesium oleate can be calculated to be 7.55%. The Cs-OA solution was stored at room temperature and preheated to 140 °C before synthesis of CsPbBr₃ nanocrystals.

Synthesis and Purification of CsPbBr₃ NCs: Briefly, 138 mg of PbBr₂ (0.374 mmol) was mixed with 10.0 mL ODE in another 25 mL three-neck round bottom flask connected with a double row of tubes and was dried under vacuum at 120 °C for 1 h. Then, 1.0 mL OAm and 1.0 mL OA were slowly added into the flask under the N₂ atmosphere. After the solution became clear, the temperature was raised to 170 °C, followed rapidly injection of 0.8 mL of Cs-OA. After 5 s, the solution turned yellow, and the reaction was quenched by immersing the flask into an ice-water bath. The quenched solution was centrifuged at 10,000× g rpm for 5 min. To remove supernatant, the centrifuged CsPbBr₃ NCs were dispersed in 5 mL dry toluene, were washed twice with dry toluene, and dispersed into 25 mL of dry toluene for further use.

Preparation of asymmetrical CsPbBr₃/TiO₂ heterojunction: Typically, when the stirring rate of the reaction system is 1500 rpm, titanium butoxide solution (80 µL titanium butoxide/2 mL dry toluene) is added to a 10 mL CsPbBr₃ NCs toluene solution. After stirring for 3 h, the reaction solution was transferred into a Teflon-lined stainless-steel autoclave, and then the solvothermal reaction process was carried out at 160 °C for 10 h. After cooling down to room temperature, the obtained product was centrifuged at 10,000× g rpm for 5 min to collect precipitates, washed twice with dry toluene, and dried at 40 °C for 6 h; then, it was ground to obtain an asymmetrical CsPbBr₃/TiO₂ powder sample.

3.3. Characterization

3.3.1. Materials Characterization

Transmission Electron Microscopy (TEM) characterization was used. TEM images were acquired on a JEOL JEM-1400 Plus electron microscope equipped with a thermionic gun at an accelerating voltage of 120 kV. The samples of CsPbBr₃ NCs and CsPbBr₃/TiO₂ NCs were prepared by depositing a diluted nanocrystal suspension in toluene onto carbon-coated copper grids.

3.3.2. Optical Characterization

UV–Vis absorption spectra of CsPbBr₃ NCs and CsPbBr₃/TiO₂ NCs samples was collected by Shimadzu UV-3600. The steady-state photoluminescence (PL) spectra and steady-state and time-resolved photoluminescence (TRPL) spectra were measured on a Shimadzu RF-6000 spectrophotometer with LabSolutions RF software using an excitation wavelength (λex) of 420 nm for CsPbBr₃ NCs and the asymmetrical CsPbBr₃/TiO₂ NCs. Nanocrystals samples were prepared by diluting NC solutions in toluene in quartz cuvettes with a path length of 10 mm.

3.3.3. (Photo)electrochemistry Characterization

For (photo)electrochemistry measurement, a Zennium electrochemical workstation (Germany, Zahner Company) was used, and the measurement was performed in a three-electrode setup with the working electrode of the sample electrode, counter electrode of platinum disk, and reference electrode of Ag/AgCl (saturated KCl). (EAg/AgCl = +0.1989V vs. NHE) in the (photo)electrochemical electrolyte (0.5 M Na₂SO₄). For photocurrent measurement, the light source was a 405 LED, the light intensity was tested with a Newport photometer. Electrochemical impedance spectra were measured in 0.5 M Na₂SO₄ at −0.1 V vs. NHE had an amplitude of 10 mV (Frequency: 100 mHz–20 kHz) [33].

4. Conclusions

In summary, CsPbBr₃ NCs were firstly prepared by thermal injection liquid-phase synthesis, and then asymmetrical CsPbBr₃/TiO₂ core–shell NCs were synthesized by sol-gel method. The prepared CsPbBr₃ NCs were uniformly distributed, homogeneous in size, and highly monodisperse, with a size of around 17 nm. The TEM images of asymmetrical CsPbBr₃/TiO₂ core–shell NC materials clearly show that the CsPbBr₃ and TiO₂ materials are distributed in different regions with different image contrast, indicating the formation of heterogeneous structures. The red shift of 6 nm in the photoluminescence emission peak and the decrease in the photoluminescence emission peak and fluorescence lifetime are due to the difference in grain size and charge migration between CsPbBr₃ and TiO₂. The work function calculation shows that the Fermi level of anatase TiO₂ is lower than that of CsPbBr₃ NCs. When they touch each other, electrons flow from CsPbBr₃ to anatase titanium dioxide, making the same Fermi level consistent. The energy band structures of the materials were obtained by UV–Vis analysis, demonstrating that asymmetrical CsPbBr₃/TiO₂ NCs form a type II heterostructure with narrower forbidden band widths and shorter carrier migration paths, as well as higher carrier mobilities. By measuring the electrochemical impedance and photocurrent density of CsPbBr₃ NCs and asymmetrical CsPbBr₃/TiO₂ NCs, it is further verified that asymmetrical CsPbBr₃/TiO₂ NCs have better carrier transport properties compared with CsPbBr₃ NCs. This work provides a low-cost and easy-to-handle method to achieve good water stability and excellent optoelectronic properties of perovskite, which shows promising applications in photocatalysts and high-performance optoelectronic devices.