Ultrasonication-assisted synthesis of CsPbBr3 and Cs4PbBr6 perovskite nanocrystals and their reversible transformation

  1. Longshi Rao1,2,
  2. Xinrui Ding1,
  3. Xuewei Du1,
  4. Guanwei Liang1,
  5. Yong Tang1,
  6. Kairui Tang3 and
  7. Jin Z. Zhang2ORCID Logo

1Engineering Research Centre of Green Manufacturing for Energy-Saving and New-Energy Technology, School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou 510640, China
2Department of Chemistry and Biochemistry, University of California, Santa Cruz, CA 95064, USA
3Mechanical Engineering, Pennsylvania State University, Harrisburg, PA 17057, USA

  1. Corresponding author email

This article is part of the thematic issue "Low-dimensional materials and systems".

Guest Editor: S. Walia
Beilstein J. Nanotechnol. 2019, 10, 666–676. https://doi.org/10.3762/bjnano.10.66
Received 31 Oct 2018, Accepted 13 Feb 2019, Published 06 Mar 2019


We demonstrate an ultrasonication-assisted synthesis without polar solvent of CsPbBr3 and Cs4PbBr6 perovskite nanocrystals (PNCs) and their reversible transformation. The as-prepared CsPbBr3 PNCs and Cs4PbBr6 PNCs exhibit different optical properties that depend on their morphology, size, and structure. The photoluminescence (PL) emission and quantum yield (QY) of the CsPbBr3 PNCs can be tuned by changing the ultrasound power, radiation time, and the height of the vibrating spear. The optimized CsPbBr3 PNCs show a good stability and high PL QY of up to 85%. In addition, the phase transformation between CsPbBr3 PNCs and Cs4PbBr6 PNCs can be obtained through varying the amount of oleylamine (OAm) and water. The mechanism of this transformation between the CsPbBr3 PNCs and Cs4PbBr6 PNCs and their morphology change are studied, involving ions equilibrium, anisotropic growth kinetics, and CsBr-stripping process.

Keywords: CsPbBr3 PNCs; Cs4PbBr6 PNCs; polar-solvent-free; reversible transformation; ultrasonication


Metal halide perovskite nanocrystals (PNCs) are promising candidates for application in the fields of light-emitting diodes (LEDs) [1,2], high-efficiency solar cells [3], low-threshold lasers [4], and photodetectors [5]. Compared to traditional semiconductors, colloidal PNCs demonstrate excellent properites, such as tunable photoluminescence (PL) throughout the visible spectrum, super high PL quantum yield (QY), low trap-state density, and narrow emission linewidth [6-8]. The crystal structure of CsPbX3 (X = Cl, Br, I) PNCs consists of a 12-fold coordinated network created by [PbX6]4− octahedra in which the Cs+ ions reside in the periphery of this network [9,10]. These PNCs are prone to structural instabilities and phase transformations involving ion migration and interface hydration [11]. However, this phase and structure versatility has become the great advantage of PNCs in their technical applications, especially in optoelectronics. Although the focus has been on the CsPbX3 structure, researchers start to turn their attention on synthesizing new perovskite materials, such as Cs4PbX6 PNCs. Under Cs-rich or Pb-poor synthesis conditions, zero-dimensional (0D) structures of Cs4PbX6 NCs can be achieved, demonstrating a crystalline structure with well-separated octahedra [PbBr6]4− isolated by Cs+ ions [12,13]. This specific structure is expected to result in strong quantum confinement and electron–phonon interactions. This inspires researchers to further explore this structure. The key to this exploration is the development of various approaches to the synthesis of high-quality PNCs.

Since, in 2015, Kovalenko and co-workers synthesized CsPbX3 PNCs by using a hot-injection method, great successes in the controlled synthesis and application of the CsPbX3 PNCs have been achieved in a very short time [14]. To date, the most commonly adopted approach for synthesizing highly efficient PNCs are solution-based procedures, including hot injection, solvothermal synthesis, microreactor synthesis, and room-temperature (RT) ligand-mediated reprecipitation, in which shape and size are tuned through the control of temperature, reaction time, and composition of the precursors [15-17]. Chen et al. demonstrated a facile solvothermal method for preparing CsPbX3 PNCs with adjustable optical properties [18]. Additionally, Li's group reported a poly(lactic acid)-assisted anion-exchange method using a microreactor for tuning the emission spectra of CsPbX3 PNCs from green to near-ultraviolet, which might be applicable for mass production [19]. Besides, great efforts have been made to prepare PNCs through the chemical transformation of pre-synthesized PNCs [20-22]. For example, Wu et al. reported a CsX-stripping method that enabled the transformation of nonluminescent Cs4PbX6 PNCs to highly luminescent CsPbX3 PNCs through an interfacial reaction [20]. They focus on investigating the water-triggered transformation process between Cs4PbX6 PNCs and CsPbX3 PNCs in a different phase. Similar methods were applied to explore new perovskite materials such as Cs4PbBr6. Zhai et al. showed a simple solvothermal approach for synthesizing CsPbBr3 nanoplatelets and their phase transformation to Cs4PbBr6 PNCs [23]. Liu and co-workers also demonstrated that CsPbBr3 PNCs were successfully converted to Cs4PbBr6 PNCs through a “ligand-mediated transformation” method with the addition of oleylamine (OAm) [24]. Udayabhaskararao’s group showed the reversible transformation from CsPbX3 to Cs4PbX6 through the ratio of oleic acid (OA) to OAm in a Brønsted acid–base equilibrium [25]. Despite the progress made in obtaining PNCs, in general, inert conditions, high temperature and pre-synthesized precursors are required for hot injection. In addition, RT methods were mostly carried out by mixing a polar solvent with a large amount of nonpolar solvent. Since PNCs are reported to be very sensitive to polar solvents, these methods result in the inevitable degradation of PNCs, especially for iodine-based PNCs [26-28]. Therefore, in order to obtain PNCs with high PL QY and stability, it is crucial to develop synthesis methods free of polar-solvents.

To date, some attempts have been made to synthesize PNCs without the use of polar solvents. Tong’s group demonstrated the single-step and polar-solvent-free synthesis of CsPbX3 PNCs with tunable halide ion composition and thickness through the direct ultrasonication of precursors [29]. Whereas this method has been reported for synthesizing PNCs without using polar solvents, it does not allow for a control over dimensionality and phase transformation. We recently reported a fast, low-cost, environmentally friendly, and polar-solvent-free strategy for synthesizing all-inorganic CsPbBr3 NCs with tunable shape and size [30]. During this process, we found that a great excess of OAm results in the formation of a derivative of CsPbBr3 NCs, i.e., Cs4PbBr6. However, the underlying transformation mechanism has not been fully understood. Following this, we set out here to expand this study to control the phase transformation. CsPbBr3 PNCs as precursor were obtained by modifying the approach initially presented by Tong, which was recently elaborated by our group [29,30]. We demonstrated in detail how, by tuning the ultrasound power and time, the PL emission of CsPbBr3 PNCs can be precisely controlled. Benefitting from this knowledge, here we attained CsPbBr3 PNCs with a high PL QY (ca. 85%) by optimizing the immersion height of the vibrating spear in the liquid. In addition, the phase transformation of CsPbBr3 PNCs to Cs4PbBr6 PNCs was achieved in this study by direct ultrasonication of solid powders or by adding OAm in the solution of pre-synthesized CsPbBr3 PNCs. Finally, inspired by the method proposed by Wu et al. [20], a successful structure conversion from Cs4PbBr6 PNCs to CsPbBr3 PNCs was obtained here by simply adding different amounts of water into pre-synthesized Cs4PbBr6 PNCs. The mechanism behind phase transformation and morphology change were investigated by using a combination of spectroscopy and microscopy techniques.

Results and Discussion

Characterization of CsPbBr3 PNCs

The typical procedure for synthesizing CsPbBr3 and Cs4PbBr6 PNCs and for reversibly transforming them is illustrated in Scheme 1. Cs2CO3 and PbBr2 were loaded into the liquid paraffin/OAm/OA solution. Then, the precursors were processed by tip-sonication and purified via centrifuging in the presence of methyl acetate as precipitation agent. Subsequently, the sediment was redispersed in toluene for further characterization. The reversible transformation between Cs4PbBr6 PNCs and CsPbBr3 PNCs was achieved by changing the amounts of OAm and water. Detailed synthesis conditions are given in the Experimental section.


Scheme 1: Illustration of synthesizing CsPbBr3 and Cs4PbBr6 PNCs without polar solvent using ultrasonication assistance, and their reversible transformation by adding OAm and water.

The crystal structure and morphology of the as-prepared samples were determined by XRD and TEM. As shown in Figure 1a, the diffraction pattern clearly indicates that orthorhombic CsPbBr3 PNCs (PDF card #18-0364) were formed. No other phases were observed, suggesting the high purity of the samples. The TEM image shown in Figure 1b demonstrates that the CsPbBr3 PNCs have a regular square morphology. HRTEM was further carried out to measure the lattice spacing of the product. Figure 1c shows a lattice spacing distance of ca. 0.41 nm for the CsPbBr3 PNCs. The size distribution shown in Figure 1d indicates that the well-dispersed CsPbBr3 PNCs have an average diameter of ca. 11.7 nm. To explore the optical properties of colloidal CsPbBr3 PNCs, UV–vis absorption spectra and PL emission spectra were recorded. As shown in Figure 1e, the first excitonic absorption peak was located at 510 nm and the strong PL emission band centered at 516 nm was observed with a narrow full width at half maximum (FWHM) of 18 nm, indicating a narrow polydispersity of the PNCs obtained by this method. The PL QY of the as-prepared CsPbBr3 PNCs measured to be ca. 85% (Rhodamin 101 as reference, PL QY is 100%) following a previously published report [31]. In addition, Supporting Information File 1, Figure S1 clearly demonstrates the improved photostability and chemical stability of CsPbBr3 PNCs.


Figure 1: Characterization of the CsPbBr3 PNCs prepared using ultrasonication assistance. (a) XRD patterns. Black line and blue line represent experimental data and standard reference, respectively. (b) TEM image. (c) HRTEM image. (d) Size distribution. (e) UV–vis absorption spectrum (red line) and PL emission spectrum (blue line). Inset is a photograph under 365 nm UV irradiation.

Effect of synthesis conditions

Our previous study has shown that ultrasound power and radiation time have a great influence on the optical properties of the CsPbBr3 PNCs [30]. In this study, we found that the immersion height of the vibrating spear in the solvent influences the product properties (the effect will be discussed later). We divided the height of liquid into five equal parts, i.e., from the bottom to the surface of the liquid, 1/5, 2/5, 3/5, 4/5, and 5/5.

We first investigated the effect of ultrasound power on the optical properties of CsPbBr3 PNCs. To avoid breaking the bottle, the immersion height of the vibrating spear and radiation time are 4/5 and 30 min, respectively. Figure 2a shows the change of UV–vis absorption spectra and PL spectra of the CsPbBr3 PNCs that were synthesized at 90, 120, 150, 180, and 210 W of ultrasound power, while keeping other synthesis conditions unchanged. If the ultrasound power is less than 90 W, there is no UV–vis absorption peak and a very weak PL intensity, implying that almost no PNCs formed. However, the first characteristic absorption peak changes to red slowly with an increase of ultrasound power, corresponding to the red-shift of the PL emission peak, which is similar to the findings we recently reported [30]. While higher ultrasound power supports faster dissolution, it has also a strong impact on the homogeneity of the PNCs. For example, when the ultrasound power is 210 W, the UV–vis absorption at long wavelengths is very high, indicating large crystals were formed with strong scattering. Therefore, it is necessary to choose the appropriate ultrasound power. The normalized PL emission peaks in Figure 2b shift from 505 to 523 nm, indicating that our approach can precisely modulate PL emission.


Figure 2: (a) UV–vis absorption spectra and PL emission spectra, and (b) normalized PL emission spectra of the CsPbBr3 PNCs that were synthesized at 90, 120, 150, 180, and 210 W of ultrasound power, respectively. (c) UV–vis absorption spectra and PL spectra, and (d) normalized PL emission spectra of CsPbBr3 PNCs that were synthesized at 10, 20, 30, 40, 50, and 60 min of radiation time.

Furthermore, we studied the influence of radiation time on the CsPbBr3 PNCs. The immersion height of the vibrating spear and ultrasound power are 4/5 and 120 W, respectively. As shown in Figure 2c and Figure 2d, when the radiation time is increased, both the UV–vis absorption and PL spectra are red-shifted, we suggest that it is the size effect that is dominant over ionic bond strength in causing the spectral shift, which is different from the effects of radiation time that we observed recently [30]. This phenomenon indicates that the immersion height of the vibrating spear would affect PNCs properties.

The effects of the immersion height of the vibrating spear in the liquid were also investigated. The total liquid height was divided into five equal parts as shown in the inset of Figure 3a. Figure 3a shows the PL intensity (UV–vis absorbance at 400 nm) of four samples that were synthesized by setting the immersion height of the vibrating spear to 1/5, 2/5, 3/5, and 4/5 of the total liquid height. As the immersion height increases, the corresponding PL intensity obviously decreases. In addition, the PL QY in Figure 3b further confirmed that the CsPbBr3 PNCs exhibit the best performance when the immersion height of the vibrating spear is set at 1/5 of the total liquid height. Ultrasonication results in a combination of thermal, vibrational, and acoustic cavitation, i.e., the formation, growth, and implosive collapse of bubbles in liquids [32-34]. In the center of these bubbles, extremely high temperatures of about 5000 K and high pressures of about 20 MPa were achieved by high-intensity ultrasound [32], enabling a quick decomposition of the particles. The lower the immersion height of vibrating spear, the higher temperature and pressure is achieved, which, as a result, benefits the formation of PNCs.


Figure 3: (a) PL emission spectra and (b) PL QY of the CsPbBr3 PNCs that were synthesized at 1/5, 2/5, 3/5, and 4/5 of total liquid height, respectively. 1/5, 2/5, 3/5, 4/5, and 5/5 in (a) inset is one of the five equal parts from liquid bottom to liquid surface. R-101 in (b) is rhodamine 101 (PL QY, 100%) as the standard sample to test PL QY of PNCs.

Characterization of Cs4PbBr6 PNCs

The current approach can be further used for controlling phase and structure transformations in the PNCs. The method introduced in this work enables the successful synthesis of rhombohedral Cs4PbBr6 PNCs via changing the amount of OAm. The amount of OAm was increased to 3.0 mL, while all other conditions were kept the same. The phase of the obtained product was characterized by XRD, as shown in Figure 4a. The XRD pattern with peaks at 2θ = 12.9, 20.1, 22.4, 25.6, 28.6 30.3, 30.9, 34.1, 39.3, and 45.7° correspond to diffractions from (110), (113), (300), (024), (214), (223), (006), (134), (330), and (600) crystal planes of rhombohedral Cs4PbBr6 (PDF card #73-2478) [13]. Meanwhile, weak peaks of CsPbBr3 were observed, indicating both CsPbBr3 and Cs4PbBr6 PNCs were formed during the process.


Figure 4: Characterization of the Cs4PbBr6 PNCs synthesized using ultrasonication. (a) XRD patterns. Black line represents experimental data; blue line (below) and pink line (upper) represent standard reference bulk CsPbBr3 and Cs4PbBr6, respectively. (b) TEM image, red squares represent CsPbBr3 PNCs. (c),(d) HRTEM images, green arrows in (c) represent metallic lead nanoparticles and inset in (d) is SEM image of Cs4PbBr6 PNCs, respectively. (e) UV–vis absorption spectra (red line) and PL emission spectra (blue line). Insets are high resolution absorption spectrum and representative digital photograph.

TEM was further performed to characterize the morphology of as-prepared PNCs. Figure 4b shows the formation of Cs4PbBr6 PNCs with hexagonal crystal structure and confirms the existence of square-shaped CsPbBr3 PNCs. Additionally, Figure 4c shows that the small black spots (green arrows) existing on the surface of the Cs4PbBr6 PNCs are metallic lead nanoparticles that have been reported before [23,35,36]. The HRTEM image shown in Figure 4d demonstrates an interplanar spacing of 0.39 nm, corresponding to the (300) crystal plane of bulk Cs4PbBr6, which is also consistent with the PDF card #73-2478. The size of the Cs4PbBr6 PNCs is defined here as the longest distance between hexagonal corners, which is ca. 65 nm for the example shown. Besides, the SEM image illustrates that Cs4PbBr6 particles are hexagonal prisms with a thickness of ca. 15 nm, as presented in Figure 4d inset and Supporting Information File 1, Figure S2.

The absorption spectrum of the Cs4PbBr6 PNCs is very different from that of the CsPbBr3 PNCs, as shown in Figure 4e. The first excitonic absorption has been shifted from 510 nm for CsPbBr3 PNCs to 315 nm for Cs4PbBr6 PNCs. This absorption feature is consistent with that of bulk Cs4PbBr6, which was proven to be the localized 6s1/2–6p1/2 transition within the isolated [PbBr6]4− octahedra separated by Cs+ ions [24]. A weak characteristic UV–vis absorption and a PL emission peak for CsPbBr3 PNCs at 505 nm and 520 nm, respectively, further confirmed the existence CsPbBr3 PNCs.

In Cs4PbBr6 PNCs, typically a green emission arises either from defects or from impurities or from a combination of both [37,38]. Herein, the purified Cs4PbBr6 PNCs did not demonstrate PL emission over the whole visible spectrum due to their wide bandgap (Eg(Cs4PbBr6) = 3.94 eV), while the observed weak PL emission results from a small portion of CsPbBr3 impurities in the Cs4PbBr6 PNCs (see Figure 3a–e). Since CsPbBr3 PNCs exhibit a high PL QY, the green PL emission is ascribed to minor CsPbBr3 impurities in the samples. This result coincides with previous works on Cs4PbBr6 PNCs that show a strong green emission at about 500 nm and confirmed that these green PL emissions originate from CsPbBr3 PNC impurities [10,39,40].

Furthermore, the effect of the amount of OAm on the phase transformation was investigated. As shown in Figure 5a, when the amount of OAm ranged from 0.5 to 3.0 mL, the first characteristic absorption peak (ca. 510 nm) and the PL emission intensity of CsPbBr3 PNCs slowly decrease, while new strong absorption features in the UV region (ca. 315 nm) emerge, which have been confirmed to result from the formation of Cs4PbBr6 PNCs [25]. When adding equal amounts of OAm and OA, there are no other peaks in the UV–vis absorption spectra except for that at ca. 510 nm. With increasing amount of OAm, the absorption intensity at ca. 510 nm decreases. Simultaneously, the absorption intensity at ca. 315 nm increases, while the PL intensity decreases (Figure 5b), and is blue-shifted followed by an increase in the FWHM of the PL peak. All these effects suggest the decomposition of the CsPbBr3 PNCs. Based on this process it can be concluded that the excess amount of OAm triggers the transformation between CsPbBr3 PNCs and Cs4PbBr6 PNCs.


Figure 5: The change of (a) UV–vis absorption spectra, (b) PL spectra, and (c)–(h) TEM images of CsPbBr3 PNCs synthesized at 0.5, 1.0, 1.5, 2.0, 2.5, and 3.0 mL of OAm, respectively, while keeping OA unchanged. (i) Schematic illustration of the morphology change from CsPbBr3 PNCs to Cs4PbBr6 PNCs.

The morphology change from CsPbBr3 PNCs to Cs4PbBr6 PNCs was further confirmed by using TEM. When the amount of OAm is between 0.5 and 1.5 mL, the morphology of PNCs gradually becomes irregular and some hexagonal shapes emerge (Figure 5c–e). As the amount of OAm increase to 3.0 mL, Cs4PbBr6 PNCs with homogeneous hexagonal shape can be achieved (Figure 5h). These results suggest that the growth kinetics of this process can be controlled by adding OAm, and the PNCs are prone to crystallize in the Cs-rich Cs4PbBr6 phase when OAm is present in excess. The transformation from CsPbBr3 to Cs4PbBr6 leads to a remarkable change in crystal structure and atomic composition. Udayabhaskararao et al. demonstrated that this transformation is driven by recrystallization induced by micelle formation or soft-ligand templating [25]. This mechanism, however, cannot explain the phase transformation from CsPbBr3 to Cs4PbBr6 in this work because there are no intermediate stages observed by TEM. Therefore, we suppose the transformation between the two phases involves ion equilibria. A large amount of OAm can form oleylammonium and dissolve PbBr2, resulting in the formation of lead oleate and oleylammonium bromide, thus driving the transformation [41]. An even larger amount of OAm can also dissolve CsPbBr3 PNCs and accelerate the transformation into Cs4PbBr6 PNCs. This process is related to Ostwald ripening that was found during the nucleation and growth of PNCs [42]. Therefore, the formation of Cs4PbBr6 PNCs is promoted by the capacity of the organic ligands to dissolve PbBr2 and by the dissociation of CsPbBr3 PNCs.

A series of TEM images (Figure 4c–h) clearly confirm that the morphology of these PNCs can be tuned easily by changing the amount of OAm in the precursor solution, while keeping the amount of OA unchanged. We attribute this morphology change to the crystal anisotropy induced by the growth kinetics. The capping ligands are preferentially attached to the PNCs facets, resulting in different growth rates on different crystal facets [43]. A schematic illustration of the morphology change between CsPbBr3 PNCs and Cs4PbBr6 PNCs is shown in Figure 5i. When equal amounts of OA and OAm are added, the reaction favors isotropic growth, since OA and OAm play a cooperative role (Figure 5i(1)). When more OAm is added, the long-chain OAm are more easily bound to the surface of the PNCs and restrict the perpendicular growth (001) [44]. Additionally, the growth rates for the side planes are different due to excess amount of OAm easily aggregated at the boundary of two adjacent planes [45], which possibly leads to the formation of (100) planes and the appearance of diamond-like product, as demonstrated in Figure 5i(2,3). This inhibiting effect is distinct when OAm is added in large excess, yielding a hexagonal structure with sharp edges, as shown in Figure 5i(4).

Reversible transformation between Cs4PbBr6 PNCs and CsPbBr3 PNCs

After a few weeks, the prepared Cs4PbBr6 PNCs solution became milky white, indicating that untransformed CsPbBr3 PNCs decomposed completely. We further explored a possible reversible transformation by introducing different amounts of water. When little water was added, the color of Cs4PbBr6 solution changed from colorless to light-green rapidly (Supporting Information File 2), implying a possible structural transformation. In order to monitor the transformation process, different amounts of water were gradually dropped into a Cs4PbBr6 PNCs solution (Figure 6a). When more water was added, the solution became green-yellow.


Figure 6: (a) The transformation process from Cs4PbBr6 PNCs to CsPbBr3 PNCs. The products were illuminated under a 365 nm UV light. Both (b) UV–vis absorption spectra and (c) PL spectra of products were recorded during the transformation process. (d) XRD diffraction patterns of typical products prepared by changing the amount of added water, demonstrating the transformation from rhombohedral Cs4PbBr6 PNCs to orthorhombic CsPbBr3 PNCs. (e) Stability of as-prepared CsPbBr3 PNCs in ambient environment.

Furthermore, the transformation process was studied by using UV–vis absorption and PL emission spectroscopy. As shown in Figure 6b, the colorless Cs4PbBr6 PNCs solution exhibits a strong first excitonic absorption peak at 315 nm. When a small amount of water was slowly added, the intensity of the first excitonic absorption peak declined gradually, indicating the decomposition of Cs4PbBr6 PNCs. Correspondingly, a weak absorption peak at 510 nm emerged. Moreover, as the amount of water was increased, the intensity of the absorption peak at about 510 nm increased steadily (inset in Figure 6b). Compared with the excitonic absorption peak of Cs4PbBr6 PNCs, the product displayed only weak absorbance. Figure 6c demonstrates the PL emission spectra of the samples during the transformation process. After the addition of small amounts of water, a PL emission peak at 518 nm appeared and gradually increased in intensity, suggesting a luminescent product was formed. XRD measurements were carried out to determine the phase of the obtained product. As shown in Figure 6d, the XRD diffraction pattern of final product is consistent with bulk orthorhombic CsPbBr3 (PDF card #18-0364), suggesting the formation of CsPbBr3 PNCs. Moreover, the PL QY of as-prepared CsPbBr3 PNCs was calculated to be ca. 70%. Interestingly, the CsPbBr3 PNCs show a high stability in ambient environment, as shown in Figure 6e. Upon the addition of a large amount of OAm and upon ultrasonication, the conversion from CsPbBr3 PNCs to Cs4PbBr6 PNCs was achieved (Supporting Information File 1, Figure S3) and can be repeated more than two times, similar to previous reports [25].

The addition of led to the decomposition of Cs4PbBr6 and the formation of CsPbBr3 PNCs, triggered by the stripping of water-soluble CsBr. During this process, the rhombohedral Cs4PbBr6 PNCs are slowly converted to orthorhombic CsPbBr3 PNCs (Figure 6d) and the rate of this conversion depends on the amount of water. The CsBr-stripping can be proven by the reduction of crystal size from 65 to 11.7 nm (Figure 1 and Figure 4). This is consistent with the findings reported by Wu and co-workers [20]. As the solubility of liquid paraffin or capping ligands in water is very low, further dissolution of CsPbBr3 PNCs is inhibited, which is similar to the effect demonstrated by Wu’s group who took advantage of the very low (only 9.5 mg/L) solubility of hexane in water [20]. The above result indicates that the CsPbBr3 PNCs have a higher stability than Cs4PbBr6 PNCs against water.


In summary, we demonstrate the effect of small changes in the environment of capping ligands and water on the crystal structure and stoichiometry of PNCs. This study expanded our recent work of synthesizing differently shaped CsPbBr3 PNCs [30]. Similarly as demonstrated in our recent study, by changing the ultrasound power and radiation time, the PL emission of CsPbBr3 PNCs could be easily tuned. More importantly, with lower the immersion heights of the vibrating spear higher PL QY of CsPbBr3 PNCs were achieved. The as-prepared CsPbBr3 PNCs show a high PL QY of up to 85% and a considerable photostability and chemical stability. The Cs4PbBr6 PNCs are obtained via direct ultrasonication of precursors or after adding OAm in the pre-synthesized CsPbBr3 PNCs solution. The phase transformation of orthorhombic CsPbBr3 NCs to rhombohedral Cs4PbBr6 NCs is promoted by the capacity of organic ligands to dissolve PbBr2, and by the formation of lead oleate and the dissociation of CsPbBr3 PNCs. Morphology changes are mainly ascribed to the anisotropic growth of the crystals. In addition, a reverse transformation from Cs4PbBr6 PNCs to CsPbBr3 PNCs can be achieved by adding water to pre-synthesized Cs4PbBr6 PNCs. The developed ultrasonication assistance results in the successful control over the phase transformation of PNCs, which can find widespread application in photoelectronic devices. We anticipate that this work can be extended to prepare other halide perovskites.



Cesium carbonate (Cs2CO3, 99%), lead bromide (PbBr2, 98%), liquid paraffin (90%), oleic acid (OA, 90%), oleylamine (OAm, 70%), and anhydrous toluene (99.8%) were purchased from Shanghai Aladdin Biochemical Technology Co. The chemicals used in the present work were of analytical grade and used without further purifications.

Synthesis of CsPbBr3 PNCs and Cs4PbBr6 PNCs

CsPbBr3 PNCs: The PNCs were prepared via modifying the procedures reported by Tong and Rao and co-workers [29,30]. In a typical process, Cs2CO3 (0.15 mmol) and PbBr2 (0.30 mmol) powders were added to a mixture of 10 mL liquid paraffin (LP), 0.50 mL OA and 0.50 mL OAm. Then the reaction medium was processed by tip-sonication at a power of 120 W for 40 min. During the sonication, the colorless reaction medium gradually transformed into a yellow and then an orange-yellow solution, which suggests the formation of PNCs and demonstrates strong fluorescence emission under 365 nm UV light excitation. After completion of the reaction, unreacted precursors and excess ligands were removed by centrifugation at a speed of 3000 rpm for 10 min and then the precipitate was redispersed in 5.0 mL of toluene. Then, the colloidal solution was centrifuged at a speed of 12000 rpm for 5 min and the sediment was redispersed in toluene for further characterization.

Cs4PbBr6 PNCs: Cs2CO3 (0.15 mmol) and PbBr2 (0.30 mmol) powders were added to a mixture of 10 mL liquid paraffin, 0.50 mL OA and 3.0 mL OAm, while keeping other synthesis conditions as the same as that of CsPbBr3 PNCs.

Reversible transformation from Cs4PbBr6 PNCs to CsPbBr3 PNCs

50–250 μL of water was added to 5.0 mL of the pre-synthesized Cs4PbBr6 PNCs solution and shaken slightly, which is a modification of the work reported by Wu and co-workers [20].


The crystal surface morphology of the PNCs was characterized by transmission electron microscopy (TEM, JEM-2100F, JEOL, Japan) with an accelerating voltage of 100 kV. High-resolution TEM (HRTEM) was carried out on a JEOL JEM-2100F instrument operating at 200 kV. The crystal phases of the products were measured using an X-ray diffractometer (XRD, D8-Advance, Bruker, Germany) with a Cu Kα radiation source (λ = 0.15418 nm) at a counting rate of 2° per minute in the scanning angle (2θ) range from 5° to 50°. The surface morphology of Cs4PbBr6 PNCs was characterized by using a field-emission scanning electron microscope (SEM, Merlin). The UV–vis absorption spectra of the samples were measured using a UV–vis spectrometer (Shimadzu, Japan) over the wavelength range from 300 to 700 nm, at 1 nm intervals. The PL spectra of the PNCs were recorded using a fluorescence spectrophotometer (RF-6000, Shimadzu, Japan) using a Xe lamp as an excitation source.

Supporting Information

Supporting Information File 1: Additional PL spectra, SEM image, and UV–vis absorption spectra.
Format: PDF Size: 316.7 KB Download
Supporting Information File 2: Video showing the transformation from Cs4PbBr6 to CsPbBr3 PNCs after addition of water.
Format: MP4 Size: 10.7 MB Download


This work is financially supported by National Natural Science Foundation of China (51735004, 51805173, and 51775199), Natural Science Foundation of Guangdong Province (2014A030312017), Science & Technology Program of Guangdong Province (2017B010115001), and the SCUT Fundamental Research Funds for the Central Universities.


  1. Pan, J.; Quan, L. N.; Zhao, Y.; Peng, W.; Murali, B.; Sarmah, S. P.; Yuan, M.; Sinatra, L.; Alyami, N. M.; Liu, J.; Yassitepe, E.; Yang, Z.; Voznyy, O.; Comin, R.; Hedhili, M. N.; Mohammed, O. F.; Lu, Z. H.; Kim, D. H.; Sargent, E. H.; Bakr, O. M. Adv. Mater. (Weinheim, Ger.) 2016, 28, 8718–8725. doi:10.1002/adma.201600784
    Return to citation in text: [1]
  2. Li, J.; Xu, L.; Wang, T.; Song, J.; Chen, J.; Xue, J.; Dong, Y.; Cai, B.; Shan, Q.; Han, B.; Zeng, H. Adv. Mater. (Weinheim, Ger.) 2017, 29, 1603885. doi:10.1002/adma.201603885
    Return to citation in text: [1]
  3. Jeon, N. J.; Na, H.; Jung, E. H.; Yang, T.-Y.; Lee, Y. G.; Kim, G.; Shin, H.-W.; Il Seok, S.; Lee, J.; Seo, J. Nat. Energy 2018, 3, 682–689. doi:10.1038/s41560-018-0200-6
    Return to citation in text: [1]
  4. Wang, X.; Shoaib, M.; Wang, X.; Zhang, X.; He, M.; Luo, Z.; Zheng, W.; Li, H.; Yang, T.; Zhu, X.; Ma, L.; Pan, A. ACS Nano 2018, 12, 6170–6178. doi:10.1021/acsnano.8b02793
    Return to citation in text: [1]
  5. Yang, T.; Zheng, Y.; Du, Z.; Liu, W.; Yang, Z.; Gao, F.; Wang, L.; Chou, K.-C.; Hou, X.; Yang, W. ACS Nano 2018, 12, 1611–1617. doi:10.1021/acsnano.7b08201
    Return to citation in text: [1]
  6. Vickers, E. T.; Graham, T. A.; Chowdhury, A. H.; Bahrami, B.; Dreskin, B. W.; Lindley, S.; Bonabi Naghadeh, S.; Qiao, Q.; Zhang, J. Z. ACS Energy Lett. 2018, 3, 2931–2939. doi:10.1021/acsenergylett.8b01754
    Return to citation in text: [1]
  7. Koscher, B. A.; Swabeck, J. K.; Bronstein, N. D.; Alivisatos, A. P. J. Am. Chem. Soc. 2017, 139, 6566–6569. doi:10.1021/jacs.7b02817
    Return to citation in text: [1]
  8. Wu, K.; Liang, G.; Shang, Q.; Ren, Y.; Kong, D.; Lian, T. J. Am. Chem. Soc. 2015, 137, 12792–12795. doi:10.1021/jacs.5b08520
    Return to citation in text: [1]
  9. Akkerman, Q. A.; Park, S.; Radicchi, E.; Nunzi, F.; Mosconi, E.; De Angelis, F.; Brescia, R.; Rastogi, P.; Prato, M.; Manna, L. Nano Lett. 2017, 17, 1924–1930. doi:10.1021/acs.nanolett.6b05262
    Return to citation in text: [1]
  10. Yin, J.; Zhang, Y.; Bruno, A.; Soci, C.; Bakr, O. M.; Brédas, J.-L.; Mohammed, O. F. ACS Energy Lett. 2017, 2, 2805–2811. doi:10.1021/acsenergylett.7b01026
    Return to citation in text: [1] [2]
  11. Yin, J.; Maity, P.; De Bastiani, M.; Dursun, I.; Bakr, O. M.; Brédas, J.-L.; Mohammed, O. F. Sci. Adv. 2017, 3, e1701793. doi:10.1126/sciadv.1701793
    Return to citation in text: [1]
  12. Saidaminov, M. I.; Almutlaq, J.; Sarmah, S.; Dursun, I.; Zhumekenov, A. A.; Begum, R.; Pan, J.; Cho, N.; Mohammed, O. F.; Bakr, O. M. ACS Energy Lett. 2016, 1, 840–845. doi:10.1021/acsenergylett.6b00396
    Return to citation in text: [1]
  13. Seth, S.; Samanta, A. J. Phys. Chem. Lett. 2017, 8, 4461–4467. doi:10.1021/acs.jpclett.7b02100
    Return to citation in text: [1] [2]
  14. Protesescu, L.; Yakunin, S.; Bodnarchuk, M. I.; Krieg, F.; Caputo, R.; Hendon, C. H.; Yang, R. X.; Walsh, A.; Kovalenko, M. V. Nano Lett. 2015, 15, 3692–3696. doi:10.1021/nl5048779
    Return to citation in text: [1]
  15. Pan, A.; He, B.; Fan, X.; Liu, Z.; Urban, J. J.; Alivisatos, A. P.; He, L.; Liu, Y. ACS Nano 2016, 10, 7943–7954. doi:10.1021/acsnano.6b03863
    Return to citation in text: [1]
  16. He, X.; Qiu, Y.; Yang, S. Adv. Mater. (Weinheim, Ger.) 2017, 29, 1700775. doi:10.1002/adma.201700775
    Return to citation in text: [1]
  17. Bekenstein, Y.; Koscher, B. A.; Eaton, S. W.; Yang, P.; Alivisatos, A. P. J. Am. Chem. Soc. 2015, 137, 16008–16011. doi:10.1021/jacs.5b11199
    Return to citation in text: [1]
  18. Chen, M.; Zou, Y.; Wu, L.; Pan, Q.; Yang, D.; Hu, H.; Tan, Y.; Zhong, Q.; Xu, Y.; Liu, H.; Sun, B.; Zhang, Q. Adv. Funct. Mater. 2017, 27, 1701121. doi:10.1002/adfm.201701121
    Return to citation in text: [1]
  19. Rao, L.; Tang, Y.; Yan, C.; Li, J.; Zhong, G.; Tang, K.; Yu, B.; Li, Z.; Zhang, J. Z. J. Mater. Chem. C 2018, 6, 5375–5383. doi:10.1039/c8tc00582f
    Return to citation in text: [1]
  20. Wu, L.; Hu, H.; Xu, Y.; Jiang, S.; Chen, M.; Zhong, Q.; Yang, D.; Liu, Q.; Zhao, Y.; Sun, B.; Zhang, Q.; Yin, Y. Nano Lett. 2017, 17, 5799–5804. doi:10.1021/acs.nanolett.7b02896
    Return to citation in text: [1] [2] [3] [4] [5] [6]
  21. Palazon, F.; Urso, C.; De Trizio, L.; Akkerman, Q.; Marras, S.; Locardi, F.; Nelli, I.; Ferretti, M.; Prato, M.; Manna, L. ACS Energy Lett. 2017, 2, 2445–2448. doi:10.1021/acsenergylett.7b00842
    Return to citation in text: [1]
  22. Palazon, F.; Almeida, G.; Akkerman, Q. A.; De Trizio, L.; Dang, Z.; Prato, M.; Manna, L. Chem. Mater. 2017, 29, 4167–4171. doi:10.1021/acs.chemmater.7b00895
    Return to citation in text: [1]
  23. Zhai, W.; Lin, J.; Li, Q.; Zheng, K.; Huang, Y.; Yao, Y.; He, X.; Li, L.; Yu, C.; Liu, C.; Fang, Y.; Liu, Z.; Tang, C. Chem. Mater. 2018, 30, 3714–3721. doi:10.1021/acs.chemmater.8b00612
    Return to citation in text: [1] [2]
  24. Liu, Z.; Bekenstein, Y.; Ye, X.; Nguyen, S. C.; Swabeck, J.; Zhang, D.; Lee, S.-T.; Yang, P.; Ma, W.; Alivisatos, A. P. J. Am. Chem. Soc. 2017, 139, 5309–5312. doi:10.1021/jacs.7b01409
    Return to citation in text: [1] [2]
  25. Udayabhaskararao, T.; Houben, L.; Cohen, H.; Menahem, M.; Pinkas, I.; Avram, L.; Wolf, T.; Teitelboim, A.; Leskes, M.; Yaffe, O.; Oron, D.; Kazes, M. Chem. Mater. 2018, 30, 84–93. doi:10.1021/acs.chemmater.7b02425
    Return to citation in text: [1] [2] [3] [4]
  26. Huang, H.; Xue, Q.; Chen, B.; Xiong, Y.; Schneider, J.; Zhi, C.; Zhong, H.; Rogach, A. L. Angew. Chem. 2017, 129, 9699–9704. doi:10.1002/ange.201705595
    Return to citation in text: [1]
  27. Fang, F.; Chen, W.; Li, Y.; Liu, H.; Mei, M.; Zhang, R.; Hao, J.; Mikita, M.; Cao, W.; Pan, R.; Wang, K.; Sun, X. W. Adv. Funct. Mater. 2018, 28, 1706000. doi:10.1002/adfm.201706000
    Return to citation in text: [1]
  28. Li, X.; Yu, D.; Cao, F.; Gu, Y.; Wei, Y.; Wu, Y.; Song, J.; Zeng, H. Adv. Funct. Mater. 2016, 26, 5903–5912. doi:10.1002/adfm.201601571
    Return to citation in text: [1]
  29. Tong, Y.; Bladt, E.; Aygüler, M. F.; Manzi, A.; Milowska, K. Z.; Hintermayr, V. A.; Docampo, P.; Bals, S.; Urban, A. S.; Polavarapu, L.; Feldmann, J. Angew. Chem., Int. Ed. 2016, 55, 13887–13892. doi:10.1002/anie.201605909
    Return to citation in text: [1] [2] [3]
  30. Rao, L.; Tang, Y.; Song, C.; Xu, K.; Vickers, E. T.; Bonabi Naghadeh, S.; Ding, X.; Li, Z.; Zhang, J. Z. Chem. Mater. 2019, 31, 365–375. doi:10.1021/acs.chemmater.8b03298
    Return to citation in text: [1] [2] [3] [4] [5] [6] [7]
  31. Brouwer, A. M. Pure Appl. Chem. 2011, 83, 2213–2228. doi:10.1351/pac-rep-10-09-31
    Return to citation in text: [1]
  32. Qi, X.; Zhou, T.; Deng, S.; Zong, G.; Yao, X.; Fu, Q. J. Mater. Sci. 2014, 49, 1785–1793. doi:10.1007/s10853-013-7866-8
    Return to citation in text: [1] [2]
  33. Mphuthi, N. G.; Adekunle, A. S.; Fayemi, O. E.; Olasunkanmi, L. O.; Ebenso, E. E. Sci. Rep. 2017, 7, 43181. doi:10.1038/srep43181
    Return to citation in text: [1]
  34. Yuwen, L.; Yu, H.; Yang, X.; Zhou, J.; Zhang, Q.; Zhang, Y.; Luo, Z.; Su, S.; Wang, L. Chem. Commun. 2016, 52, 529–532. doi:10.1039/c5cc07301d
    Return to citation in text: [1]
  35. Akkerman, Q. A.; Motti, S. G.; Srimath Kandada, A. R.; Mosconi, E.; D’Innocenzo, V.; Bertoni, G.; Marras, S.; Kamino, B. A.; Miranda, L.; De Angelis, F.; Petrozza, A.; Prato, M.; Manna, L. J. Am. Chem. Soc. 2016, 138, 1010–1016. doi:10.1021/jacs.5b12124
    Return to citation in text: [1]
  36. Dang, Z.; Shamsi, J.; Palazon, F.; Imran, M.; Akkerman, Q. A.; Park, S.; Bertoni, G.; Prato, M.; Brescia, R.; Manna, L. ACS Nano 2017, 11, 2124–2132. doi:10.1021/acsnano.6b08324
    Return to citation in text: [1]
  37. Seth, S.; Samanta, A. J. Phys. Chem. Lett. 2018, 9, 176–183. doi:10.1021/acs.jpclett.7b02931
    Return to citation in text: [1]
  38. Akkerman, Q. A.; Abdelhady, A. L.; Manna, L. J. Phys. Chem. Lett. 2018, 9, 2326–2337. doi:10.1021/acs.jpclett.8b00572
    Return to citation in text: [1]
  39. Zhang, Y.; Saidaminov, M. I.; Dursun, I.; Yang, H.; Murali, B.; Alarousu, E.; Yengel, E.; Alshankiti, B. A.; Bakr, O. M.; Mohammed, O. F. J. Phys. Chem. Lett. 2017, 8, 961–965. doi:10.1021/acs.jpclett.7b00105
    Return to citation in text: [1]
  40. Chen, D.; Wan, Z.; Chen, X.; Yuan, Y.; Zhong, J. J. Mater. Chem. C 2016, 4, 10646–10653. doi:10.1039/c6tc04036e
    Return to citation in text: [1]
  41. Almeida, G.; Goldoni, L.; Akkerman, Q.; Dang, Z.; Khan, A. H.; Marras, S.; Moreels, I.; Manna, L. ACS Nano 2018, 12, 1704–1711. doi:10.1021/acsnano.7b08357
    Return to citation in text: [1]
  42. Koolyk, M.; Amgar, D.; Aharon, S.; Etgar, L. Nanoscale 2016, 8, 6403–6409. doi:10.1039/c5nr09127f
    Return to citation in text: [1]
  43. Liang, Z.; Zhao, S.; Xu, Z.; Qiao, B.; Song, P.; Gao, D.; Xu, X. ACS Appl. Mater. Interfaces 2016, 8, 28824–28830. doi:10.1021/acsami.6b08528
    Return to citation in text: [1]
  44. Liu, W.; Zheng, J.; Cao, S.; Wang, L.; Gao, F.; Chou, K.-C.; Hou, X.; Yang, W. Inorg. Chem. 2018, 57, 1598–1603. doi:10.1021/acs.inorgchem.7b02941
    Return to citation in text: [1]
  45. Li, G.; Wang, H.; Zhu, Z.; Chang, Y.; Zhang, T.; Song, Z.; Jiang, Y. Chem. Commun. 2016, 52, 11296–11299. doi:10.1039/c6cc05877a
    Return to citation in text: [1]

© 2019 Rao et al.; licensee Beilstein-Institut.
This is an Open Access article under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0). Please note that the reuse, redistribution and reproduction in particular requires that the authors and source are credited.
The license is subject to the Beilstein Journal of Nanotechnology terms and conditions: (https://www.beilstein-journals.org/bjnano)

Back to Article List

Other Beilstein-Institut Open Science Activities

Keep Informed

RSS Feed

Subscribe to our Latest Articles RSS Feed.


Follow the Beilstein-Institut


Twitter: @BeilsteinInst