Lead nitrate (99.0%, Sinopharm Chemical Reagent Co., Ltd.), potassium iodide (99.0%, Tianjin Sailboat Chemical Reagent Technology Co., Ltd.), polyvinylpyrrolidone (99.0%, Shanghai Zhanyun Chemical Co., Ltd.), sodium citrate (99.0%, Beijing Chemical Works), EDTA (Beijing Chemical Works) and lead iodide as the standard sample (99.95%, Sinopharm Chemical Reagent Co., Ltd.) were acquired commercially and used without further purification. Experimental pH meter (PHSJ3F, Shanghai Jingke Technology Co., Ltd.), electronic balance (AUY220, Shimadzu Corporation, Japan), X-ray diffractometer(ZEISS, Germany), field emission scanning (ZEISS SUPRA55, Germany), energy dispersive spectrometer (OXFORD INCA X-ACT, America) and FT-Raman spectroscopy (8400S, Shimadzu Corporation, Japan) were used in this study.

The preparation of lead iodide in the present study is carried out through a pH-constantly controlled double-jet precipitation method. A schematic diagram of the synthesis procedure is shown in Fig. (1). In this typical synthesis procedure, 1.0 g (3 mmol) of Pb (NO₃)₂ was dissolved in 100 ml of deionized water to prepare 0.03 mol/L Pb (NO₃)₂ solution (solution A), and adjusted to pH=4 with HNO₃. At the same time, 1.1 g (6.6 mmol) of KI was dissolved in 100 ml of deionized water to prepare a 0.066 mol/L KI solution (solution B), which was adjusted to pH=4 with HNO₃. The next step was to measure 200 ml of deionized water (solution C), and adjust it to pH=4 with HNO₃. Then, solution A and solution Bwere added dropwise to solution C at a rate of 1 ml/min with the help of two peristaltic pumps, respectively. During the whole feeding process, the solution mixture was kept in a state of continuous stirring. When the feeding was completed with continuous stirring for 10 min, the supernatant liquid was filtered off. The lead iodide precipitate was then washed 3 times with deionized water of pH=4, and then the precipitate was collected and dried at 60 °C for 24 hours.

In order to study the synthesized lead iodide powder, a series of analyses were conducted as follows. An X-ray diffractometer (ZEISS, Germany) with monochromatic Cukα radiation (λ= 1.5406 Å) at 40 kV and 50 mA was used to record the X-ray diffraction pattern of the lead iodide powder. The samples were scanned at a scanning rate of 5^◦ min⁻¹ in the 2θ range of 10-90^◦. The appearance of the lead iodide powder was observed by using a scanning electron microscope, and the elemental composition characteristics of the powder were identified by an energy dispersive spectrometer (OXFORD INCA X-ACT, America). An FT-Raman spectroscopy instrument (Shimadzu Corporation, UV2600, Japan) was used to identify the functional groups of the lead iodide powder in the wavenumber range of 4000-500 nm.

Sixteen samples were obtained under different conditions, and XRD was used to characterize these precipitated powders. Fig. (2) shows the XRD patterns of the lead iodide samples obtained at different pH values. It can be seen from Fig. (2) that the diffraction peaks of lead iodide powder prepared at pH=2 and 4 are in good agreement with the diffraction peak of standard PbI₂, which indicates that the purity of the sample was quite high. On the other hand, the diffraction peaks of lead iodide powder obtained at pH=6 also showed PbIOH as an impurity of lead iodide. Obviously, pH is a key factor for the chemical composition of the precipitated particles in the double jet precipitation process.

Fig. (1) The schematic diagram of the double-jet precipitation setup used for the preparation of lead iodide.

Fig. (2) XRD patterns of the PbI2 powders synthesized at different pH-values under the following experimental conditions: [Pb(II)]=0.03mol/L, t=70min,V=1.0ml/min.

Table 1
Thermodynamic data for the calculation of each species in the Pb²⁺-I^--H₂O aqueous solution.

In order to learn more details about the aqueous solutions, the concentrations of lead and iodide species were calculated based on the mass balance of coordination between chemical species and thermodynamic constants of the Pb²⁺-I^--H₂O system as shown in Table 1. As demonstrated in Fig. (3), under acidic conditions (pH 1~5), Pb²⁺, PbI⁺ and PbI₂ are the dominant species, while at pH above 5.3, PbIOH would dominate as the only stable solid phase, which could be regarded as the mixing product of PbI and Pb(OH)₂. Both the experimental and calculated results were found to be identical in that they verify the important role of pH on the formation of pure lead iodide in the precipitation process. In the present study, the double jet precipitation process has a significant advantage of controlling the reaction pH very conveniently, which can avoid any pH fluctuation with an optimal range leading to the formation of PbIOH. On the basis of the above study, it could be concluded that the acidic pH condition determines the formation of pure PbI₂ in the synthesis process that should be carefully controlled in the complete precipitation process.

Fig. (3) Distribution species of Pb2+ and I- in the Pb2+-I--H2O aqueous solution.

Fig. (4) XRD patterns of the PbI2 powders synthesized at different lead ion concentrations under the following experimental conditions: pH=4, t=70min, V=1.0ml/min.

Fig. (4) shows the XRD results of lead iodide samples prepared under different lead ion concentrations. It can be seen that the solid prepared at lead ion concentrations of 0.01 M and 0.1 M is pure PbI₂. Interestingly, in the case of 0.01M Pb(II) concentration, the main peaks of PbI₂ at 25.91°, 34.27°, 39.51° did not increase sufficiently in size as was the case when using a 0.1M Pb(II) solution, while the peaks at 12.67°, 38.67°, and 52.39° became much stronger, indicating that some crystalline structures did not get enough precursor materials, which may be limited by the diffusion of lead and iodide ions. On the basis of the crystal growth theory, it can be deduced that in the case of very low Pb(II)-concentrations, e.g. 0.01M lead ions, crystalline PbI₂ particles have a strong ability to grow naturally according to their inherent crystalline habit in which the special crystal facets grow preferentially, such as (001), (003) and (004), as shown in Fig. (4).

Fig. (5) shows the effect of feeding rates on the chemical compositions of the precipitated powder. The sample grows better as a crystal at a feed rate of 2.0 ml/min, indicating that at a higher feeding rate, the transportation rate of the precursors of PbI₂ crystalline nuclei growth is much larger, leading to a more sufficient growth of the crystals, which is quite identical to the variation behavior found in the case of low concentrations of lead ions. As shown in Fig. (4), the peaks at 12.67°, 38.67° and 52.39° dominate at low concentrations, while the peaks at 25.91°, 34.27° and 39.51° become the preferential ones in the case of high concentrations of lead ions.

Fig. (5) XRD patterns of the PbI2 powders synthesized at different feeding rates under the following experimental conditions: pH=4, [Pb(II)]= 0.03mol/L, t=70min.

Fig. (6) XRD patterns of the PbI2 powders synthesized in the presence of different surfactants at all the concentrations of 1g/L under the following experimental conditions: pH=4, [Pb(II)]=0.03mol/L, t=70min,V=1.0ml/min.

Fig. (6) shows the diffraction peaks of the lead iodide powder prepared by adding different surfactants. It can be seen that the purity of the lead iodide sample produced with an addition of PVP (polyvinylpyrrolidone) as the surfactant is the lowest, while the purity of the samples with sodium citrate and EDTA as surfactants is similar, which may be due to the similar coordinative role of sodium citrate and EDTA. The slow release of lead ions makes the growth of crystalline nuclei possible according to the crystalline behavior of PbI₂, not leading to the random aggregation of tiny nuclei particles in the presence of PVP.

In order to investigate the effect of sodium citrate and EDTA on the formation of lead iodide, it is necessary to assess the concentration of the different lead species in the solution to explain the experimental results. Table 2 shows the thermodynamically calculated concentration of the species in the Pb²⁺-I^--Cit-H₂O and Pb²⁺-I^--EDTA-H₂O aqueous solutions. Fig. (7) shows the lead species in the solutions in the presence of citrate and EDTA, and it can be found that the introduction of chelating reagents will decrease the yields of the precipitated PbI₂ solid, due to the competition for lead ions between the chelating reaction and the precipitation process. The stronger the chelating affinity, the less the precipitation. This means that the yield of precipitated PbI₂ in the presence of EDTA will be much smaller than that in the presence of citrate. However, considering the strict demands of solar cell fabrication, residual surfactants, such as citrate, EDTA and PVP may worsen the photoelectronic properties of the cells. Thus, in the practical preparation process, additional surfactants are not recommended to be introduced in the precipitation system. For comparison, the effect of EDTA, PVP and citrate on the formation of PbI₂ is worth investigating to determine the influence of the additive on the shape of precipitated lead iodide.

Table 2
Thermodynamic data for calculated concentrations in the Pb²⁺-I^--Cit-H₂O and Pb²⁺-I^--EDTA-H₂O aqueous solutions.

Fig. (7) Distribution species concentration of Pb2+ and I- in the Pb2+-I--Cit-H2O and Pb2+-I--EDTA-H2O aqueous solutions.

Fig. (8) shows the morphology of lead iodide produced at different values of pH. It can be seen that there is not much difference between the samples obtained at pH=2 and pH=4, which are approximately hexagonal flakes with a size of about 10 μm. The particles have good crystallinity. The energy spectrum results show that the sample is lead iodide. However, when the pH value rises to 6, the chemical composition of the precipitated sample can be found to change greatly. There is also a significant change in the morphology. Lead iodide particles do not precipitate in sheet-like hexagons, which seem like flocculated particles with larger blocks inside. Obviously, the chemical composition of the precipitates determines the morphology, where the chemical is affected by the solution pH, according to the preceding discussion.

As stated in the Weimarn rule [41Weimarn, V.P.P. The precipitation laws. Chem. Rev., 1925, 2(2), 217-242.
[http://dx.doi.org/10.1021/cr60006a002] -43Ping, Y. Study of influence of cobalt oxalate precipitation process on cobalt powder size. Cemented Carbide, 2001, 18(1), 12-15.], when the concentration of the lead ions is high, e.g. 0.1mol/L, the nucleation will dominate the precipitation process, leading to the formation of very fine crystals, while in the case of dilute lead ions of a concentration of 0.01 mol/L, the growth of nuclei will become preferential to nucleation, producing larger particles. Fig. (9) illustrates the morphology of lead iodide prepared at different lead ion concentrations. We can observe that when the concentration is 0.01M, the shape of lead iodide is flake-like with a size larger than 10μm; when the concentration is 0.1M, the particles having a size of about 5 μm are closely packed with a sheet having a size of about 10 μm. Fig. (10) shows the morphology of lead iodide at different feed rates. At a feeding rate of 0.5 mL/min., the surface of the sample is rough and the overall size is large. When the feeding rate is 2.0 mL/min., the surface of the sample is smooth, but the particle size varies from 5μm to 10 μm. Fig. (11) shows the morphology of lead iodide samples after the addition of PVP, sodium citrate and EDTA. The sample with PVP had a rough surface and no regular shape, while the sample with sodium citrate had a small particle size, and the particle morphology of the sample with EDTA was a much fine flake-like shape.

Fig. (8) SEM and EDS images of the PbI2 powders synthesized at different pH: pH=2 (a), pH=4 (b), pH=6 (c) under the following experimental conditions: [Pb(II)]=0.03mol/L, t=70min,V=1.0ml/min.

Fig. (9) SEM images of the PbI2 powders synthesized at different lead ion concentrations: [Pb(II)]=0.01mol/L (a), [Pb(II)]=0.03mol/L (b), [Pb(II)]=0.1mol/L (c).

Fig. (10) SEM images of the PbI2 powders synthesized at different feeding rates: V=0.5ml/min (a), V=1.0ml/min (b), V=2.0ml/min (c) under the following experimental conditions: pH=4, [Pb(II)]=0.03mol/L, t=70min.

Fig. (11) SEM images of the PbI2 powders synthesized using different surfactants: 1g/L PVP (a), 1g/L sodium citrate (b), 1g/L EDTA (c) under the following experimental conditions: pH=4, [Pb(II)]=0.03mol/L, t=70min,V=1.0ml/min.

Fig. (12) FTIR absorption spectra of the PbI2 powders synthesized at different pH under the following experimental conditions: [Pb(II)]=0.03mol/L, t=70min,V=1.0ml/min.

Fig. (13) FTIR absorption spectra of the PbI2 powders synthesized at different lead ion concentrations under the following experimental conditions: pH=4, t=70min, V=1.0ml/min.

Fig. (14) FTIR absorption spectra of the PbI2 powders synthesized at different reaction times under the following experimental conditions: pH=4, [Pb(II)]=0.03mol/L, V=1.0ml/min.

Fig. (15) FTIR absorption spectra of the PbI2 powders synthesized at different feeding rates under the following experimental conditions: pH=4, [Pb(II)]=0.03mol/L, t=70min.

Fig. (16) FTIR absorption spectra of the PbI2 powders synthesized in the presence of different surfactants under the following experimental conditions: pH=4, [Pb(II)]=0.03mol/L, t=70min,V=1.0ml/min.

Fig. (17) Pictures to demonstrate the dissolving state of the PbI2 powders in DMF reagent synthesized under the following conditions: [Pb(II)]=0.03mol/L, t=70min,V=1.0ml/min, pH=2 (No.1), pH=4 (No.2), and pH=6 (No.3).

As mentioned in the preceding section, from the X-ray diffraction results, we learned that all the samples prepared under acidic conditions were high-purity lead iodide. For further observation, the effect of synthesis parameters on the changes in the purity of lead iodide was analyzed by FT-IR spectroscopy. It can be seen from Fig. (12) that the change in pH has a great influence on the synthesis of lead iodide. When the pH is 6, the synthesized product is not lead iodide. The key to synthesizing high-purity lead iodide is to control the pH. It can be seen from Fig. (13) that the absorption peak decreases when the concentration of lead ions is increased. It can be seen from Fig. (14) that during the titration process, as the reaction time increases, the absorption peak near 670 cm^-1 of lead iodide gradually increases, and after the end of the mixed titration, the absorption peak gradually decreases. It can be seen from Fig. (15) that there is no obvious change in the respective absorption peaks of lead iodide, therefore, the purity has no significant relationship with the feeding rate. It can be seen in Fig. (16) that the absorption peaks appear at 1380.93 cm^-1 and 829.33 cm^-1 after the addition of PVP, indicating that PVP has an effect on the purity of lead iodide. There is no significant difference in the absorption peak between lead iodide added with sodium citrate and lead iodide added with EDTA, indicating that sodium citrate and EDTA will not affect the phase purity of lead iodide.

To evaluate the solubility of the precipitated powder, which is an essential parameter that can preliminarily identify the possibility of the precipitated particles to be used as a raw material for the fabrication of perovskite solar cell film, an attempt was made to dissolve the prepared PbI₂ solids in DMF to form homogeneous clear solutions. This can be used to check the quality of PbI₂, as a conventional method.

0.5 gram of the sample, obtained at different pH (2, 4 and 6), was taken and mixed in 1 ml DMF solvent for 15 min at 80°C, and the mixtures were observed in regard to their color and turbidity. It can be seen in Fig. (17) that the powder samples precipitated at pH 2 and 4 were completely dissolved and formed transparent yellow solutions, while the sample formed at pH 6 gave a grey-white turbid suspension because of the insolubility in DMF of PbIOH that was obtained at pH 6. Evidently, the pH determines the phase and chemical compositions of the precipitated particles, and pure lead iodide can dissolve in DMF easily to form a homogeneous solution, which is good for further application in the fabrication of film coating for perovskite solar cells.