Mahsa
Shekarnoush
ab,
Francisco S.
Aguirre-Tostado
ac and
Manuel Quevedo
López
*a
aDepartment of Material Science & Engineering, The University of Texas at Dallas, Richardson, Texas 75080, USA. E-mail: mquevedo@utdallas.edu
bDepartment of Chemistry & Biochemistry, The University of Texas at Dallas, Richardson, Texas 75080, USA
cCentro de Investigación en Materiales Avanzados, S. C. (CIMAV), Unidad Monterrey, Apodaca, Nuevo León 66628, Mexico
First published on 18th March 2024
Cubic CsPbI3 is a promising perovskite material for optoelectronic applications. This material possesses an energy band gap of 1.7 eV and an optical absorption coefficient of 105 cm−1, and can be synthesized at high temperatures using various methods. However, cubic CsPbI3 faces a significant challenge with degradation reported under ambient conditions. This degradation results in the formation of a delta phase, which comprises a non-perovskite structure with poor optical and electrical properties at room temperature. This paper proposes the partial substitution of lead in CsPbI3 with divalent cations, such as Sn2+, Mn2+, Ni2+, and Ca2+, to improve the overall stability. These cations were selected because their ionic radii meet the Goldschmidt factor and favor the formation of cubic halide perovskites. For the synthesis of the materials, a solid-state mechanochemical method is used that allows for the incorporation of the above-mentioned cations into the CsPbI3 matrix in a single step at room temperature. Replacing Pb2+ with Sn2+ produced the most stable material and resulted in a cubic perovskite structure with similar optical properties to CsPbI3. In particular, a composition of CsPb0.6Sn0.4I3 resulted in cubic perovskite and did not show degradation when exposed to room temperature, ambient humidity, and atmospheric pressure for up to 25 days. On the other hand, even though Ca2+ possesses a smaller ionic radius than Pb2+, replacing Pb2+ with Ca2+ was not effective in stabilizing CsPbI3 due to the hygroscopic nature of CaI2. Replacing Pb2+ with Mn2+ and Ni2+ produces stable alloys in a controlled environment (glovebox) at room temperature but quickly decomposes to non-perovskite δ-CsPbI3, iodine, and metal oxides when exposed to air. The degradation mechanism of these materials was studied in detail using XPS techniques, revealing potential alternatives to produce stable Sn2+ containing perovskites with properties similar to those of cubic CsPbI3 at room temperature without solvents and increased stability under ambient conditions when Pb2+ was partially replaced with Sn2+.
Preserving the cubic phase structure of CsPbI3 requires a high processing temperature, but as the temperature decreases to room temperature, CsPbI3 tends to degrade.14 Despite numerous efforts, maintaining a pure cubic phase of CsPbI3 at room temperature remains challenging.15 Hence, there is a growing interest in developing stable materials with properties and structures similar to those of cubic CsPbI3.16 Several strategies have been explored to increase the stability and performance of CsPbI3-based devices such as chemical passivation techniques,10 interface engineering,17 and lattice modification approaches.18 Chemical passivation involves introduction of moisture-resistant molecules or additives to protect CsPbI3 from moisture-induced degradation.10 Other alternatives include interface engineering to optimize device architecture and interfaces to prevent moisture penetration and enhance stability.17 Lattice modification techniques, such as partial substitution of Pb with smaller elements, reduce lattice strain and defects and improve stability.19 For example, substituting smaller elements on the Pb site can lead to a decrease in lattice strain.20 Additionally, this process can minimize the formation of defects, ultimately resulting in better resistance against external degradation.21 This structural stabilization mechanism can effectively suppress phase transitions, lattice distortions, and the formation of non-perovskite phases, thus increasing material stability.
To achieve these effects, the substituting atoms must meet specific criteria. Firstly, a balanced electronic neutrality (or zero charge) must exist between the cations and anions.22 Secondly, the octahedral factor (μ) = RB/RX should fall within 0.44 ≤ μ ≤ 0.9.23 This octahedral factor range enables a favorable geometric configuration, leading to the structurally stable perovskite crystal lattice. Lastly, the Goldschmidt tolerance factor (t), defined as t = (RA + RX) √2 (RB + RX), has to be within 0.8 ≤ t ≤ 1.24 In this context, RA, RB, and RX represent the ionic radii of the A, B, and X sites, respectively.24 The established acceptable range for the tolerance factor (t) in cubic perovskites is 0.9 ≤ t ≤ 1.25 Therefore, selecting elements with appropriate ionic radii and chemical compatibility with the perovskite lattice is crucial for maintaining structural integrity and avoiding phase segregation.
However, transitioning from the theoretical framework to the practical incorporation of alternative elements into the CsPbI3 structures presents its challenges. To incorporate alternative elements into CsPbI3 structures, the standard liquid-based synthesis methods often introduce complex variables like solubility, choice of solvent, stabilizing agents, and temperature control.26 These factors can significantly affect the final structure and composition of the product.27 To avoid these issues, alternative synthesis methods to achieve materials that are structurally close to cubic CsPbI3 are currently being investigated. In this regard, ball milling, which is a solid-state technique typically conducted at room temperature, has emerged as a promising alternative.28 Ball-milling offers a more straightforward and controlled synthesis process, circumventing the complexities associated with liquid-based methods.29 Although there have been a few studies on synthesizing CsPb1−xSnxI3 using ball milling,30 this method has rarely been explored for stabilizing CsPbI3 through B-site modification without the use of additives and stabilizing agents under ambient conditions.
In this paper CsPbI3 perovskite materials with divalent cations replacing Pb2+ that have optical and structural characteristics like the cubic phase of CsPbI3 at room temperature have been studied. The synthesis is based on a mechanochemical method (ball milling) focusing on substituting lead with elements that have a smaller ionic radius that should produce a perovskite structure with enhanced stability. The proposed method differs from other conventional approaches because it excludes ligands, solvents, or stabilizing agents, without using controlled environment enabling a better evaluation of the chemical reactions involved. Our results identified Sn2+ as the most effective element to stabilize CsPbI3 in its solid state and helped understand the degradation mechanisms when Ni2+, Mn2+, and Ca2+ are used. The results described a new approach to improving the stability of CsPbI3, specifically replacing “B” site elements using simple mechanochemical methods.
Fig. 1 Calculated tolerance and octahedral factors for CsMI3 perovskites (M = Pb2+, Cd2+, Ca2+, Sn2+, Ni2+, and Mn2+). |
Fig. 2a shows that the XRD pattern of CsPb1−xSnxI3 compounds with a Sn2+/Pb2+ molar ratio ranges from 0.05 to 0.5 after a 12-hour ball-milling synthesis. The XRD pattern for the 0.05 molar ratio reveals a dominant peak at 13.02°, corresponding to the (012) plane of δ-CsPbI3. As the molar Sn2+/Pb2+ ratio increases to 0.1, an additional peak at 14.46° appears, indicating the presence of the (100) planes for cubic CsPbI3, suggesting the transformation towards the cubic phase. Further increasing the Sn2+/Pb2+ molar ratio from 0.1 to 0.4 leads to a considerable reduction in the intensity of δ-CsPbI3 peaks and an increase in the cubic CsPbI3 phase. Remarkably, for a Sn2+/Pb2+ molar ratio, the XRD results indicate a pure cubic phase. A slight shift towards higher 2θ degrees is evident in all Sn-containing samples compared to pure CsPbI3, confirming lattice contraction in CsPb1−xSnxI3. This is attributed to the incorporation of the smaller Sn2+ in the lattice structure.
Fig. 2 XRD patterns of 12-h ball-milled CsPb1−xMxI3 for several M2+ atoms, including Sn2+ (a), Ca2+ (b), Ni2+ (c), and Mn2+ (d), at various M2+/Pb2+ molar ratios. |
For a 0.5 Sn2+/Pb2+ molar ratio, the orthorhombic CsSnI3 and cubic CsPbI3 phases coexist within the structure. XRD peaks at 14.25°, 25°, and 28.7° confirm orthorhombic CsSnI3, while the peak at 14.46° confirms cubic CsPbI3. This indicates that a Sn2+ molar ratio exceeding 0.4 introduces orthorhombic CsSnI3 into the initially formed cubic crystalline structure. This is a consequence of the excess of Sn2+ in the reaction, enabling the formation of the orthorhombic CsSnI3 phase, which has a formation energy lower than cubic CsPbI3 (−1.270 eV per atom for cubic CsPbI3 and −1.203 eV per atom for orthorhombic CsSnI331). These results imply that maintaining a Sn2+ ratio below 0.5 is ideal to promote a cubic structure.
The XRD patterns of ball-milled CsPb1−xCaxI3 are shown in Fig. 2b. Interestingly, none of the samples showed a cubic phase; instead, all samples exhibited XRD patterns consistent with δ-CsPbI3. This observation is consistent with Fig. 1 where theoretical aspects of cubic perovskites were discussed. In addition to theoretical considerations, the absence of the cubic phase can be attributed to the highly hygroscopic nature of CaI2 within the reaction environment. The presence of the δ-CsPbI3 phase is a consequence of the presence of water in the ambient during the synthesis. The cubic phase cannot be achieved when using Ca2+ without using a tightly controlled environment to minimize water during the reaction. These samples were not further analyzed, and the paper focuses on Sn2+, Ni2+, and Mn2+.
Fig. 2c and d show the XRD patterns of CsPb1−xMxI3, ball-milled for 12 hours, with M representing Ni and Mn, respectively. All XRD patterns are consistent with δ-CsPbI3, with no evidence of the cubic CsPbI3 phase. Upon opening the reaction container, the initial black powder quickly transitioned to yellow within a few minutes. This indicates degradation upon exposure to air. This was explored through XPS, and a detailed discussion is presented in the following sections.
Fig. 3 Comparison of XPS spectra for CsPb0.6Sn0.4I3 samples at 0 days (a) and after 25 days (b) exposed to ambient conditions. CPSI refers to the CsPb0.6Sn0.4I3 composition. |
For both as prepared and 25-day exposed CsPb0.6Sn0.4I3 samples, the C 1s band shows two distinct chemical components. The first band at 284.8 eV indicates C–C structures commonly associated with surface-adsorbed carbon. The second band, observed at 286.67 eV, is characteristic of carbon in a C–O chemical environment. This latter band is notably enhanced in the samples that have been exposed to ambient conditions for 25 days, suggesting significant interaction with the environment.
Similarly, the O 1s region shows a single chemical environment in both samples, corresponding to oxygen in an OH environment. This feature is likely a result of the samples and precursors being exposed to moisture in the ambient air. The increased OH intensity in the exposed samples compared to the as-prepared samples further supports this conclusion, indicating an interaction with moisture and oxygen in the air.
The Cs 3d region exhibits a doublet at 723.96 eV and 737.90 eV, corresponding to the Cs 3d5/2 and Cs 3d3/2 states, respectively. These bands confirm the presence of Cs atoms in the CsPb0.6Sn0.4I3 structure. These bands remain unchanged in both the as-prepared and 25-day exposed samples, indicating the stability of the Cs atoms in the CsPb0.6Sn0.4I3 perovskites without any significant chemical change.
Similarly, for the Pb 4f region, a doublet is observed at 137.41 eV and 142.30 eV, corresponding to Pb 4f7/2 and Pb 4f5/2, respectively. This doublet is consistent with the incorporation of Pb atoms in the CsPb0.6Sn0.4I3 structure and, as in the Cs bands, shows no shift or additional bands in either the as-prepared or the exposed samples, suggesting a stable Pb environment in the CsPb0.6Sn0.4I3 perovskites.
The Sn 3d region shows a doublet at 485.60 eV and 494.1 eV, associated with Sn 3d5/2 and Sn 3d3/2, indicative of Sn2+ in the CsPb0.6Sn0.4I3 structure. The stable position of these bands in both the as-prepared and the 25-day exposed samples further supports the chemical stability of Sn2+ in the CsPb0.6Sn0.4I3 perovskites. This observation effectively eliminates concerns regarding the potential oxidation of Sn2+ to Sn4+ during the synthesis or after ambient exposure.
Lastly, the I 3d region shows a doublet at 618.63 eV and 630.12 eV, corresponding to I 3d5/2 and I 3d3/2, confirming the presence of iodine. The absence of significant shifts or new bands indicates a stable iodine environment in the CsPb0.6Sn0.4I3 perovskites.
Fig. 4 illustrates the Kubelka–Munk function derived from UV-vis diffuse reflectance spectra32 and photoluminescence (PL) spectra of as-prepared CsPb0.6Sn0.4I3 and 25-day exposed samples.
The Kubelka–Munk function results for both the as-prepared and exposed samples are quite similar, with a bandgap of 1.58 eV. Similarly, the photoluminescence (PL) spectra for both samples show an emission peak at a wavelength (λmax) of 794 nm with no significant shift in the emission peak for the sample after 25 days of exposure, suggesting that the optical properties remain stable over time. This consistent optical behavior confirms that the CsPb0.6Sn0.4I3 sample retains its inherent characteristics, highlighting its robustness and stability when exposed to ambient conditions for 25 days.
XPS was used to thoroughly investigate the synthesized materials and to study the composition of the as-prepared CsPb0.9Ni0.1I3 samples and monitor compositional changes as a function of exposure time. Two batches of samples were prepared: one in a controlled environment (glove box) and another with exposure to ambient conditions. XPS analyses were conducted in an ultra-high vacuum.
Fig. 5 illustrates the XPS spectra for the C 1s, O 1s, Cs 3d, Pb 4f, Ni 2p, and I 3d regions of the CsPb0.9Ni0.1I3 samples for the controlled and ambient-exposed environments. In the controlled environment, the C 1s region showed distinct binding energies for C in various chemical states - C–C, C–O, and CO at 284.8 eV, 286.2 eV, and 288.4 eV, respectively. This carbon presence is likely due to residual carbon and oxygen trapped in the precursor cavities during sample preparation. In the ambient-exposed samples, these carbon states were similarly observed.
In the O 1s region, samples from the controlled environment showed a single band indicative of CO or C–OH bonds, consistent with the C 1s findings. However, in contrast, the samples exposed to air showed additional bands at 529.75 eV and 531.35 eV. These new bands correspond to the formation of nickel oxides, specifically NiO and Ni2O3, which indicates significant compositional changes following exposure to air.
The Cs 3d region exhibited a distinctive doublet band with binding energies at 724.5 eV and 738.45 eV, indicative of Cs in the original CsPb0.9Ni0.1I3 perovskite. Also, a new doublet band is observed at 726 eV and 739.93 eV for the exposed sample, indicating Cs–I formation and confirming a partial decomposition of the perovskites. In contrast, the Pb 4f region showed minimal changes in its binding energies after exposure, implying a limited impact on the chemical environment for Pb2+.
The Ni 2p region showed two initial doublet bands at 855.14 eV and 858.91 eV, associated with Ni2+. Post-exposure, an additional doublet band at 858.80 eV was observed, indicating the oxidation of Ni2+ to Ni3+. This finding is consistent with the O 1s region data.
For iodine, both samples exhibited a doublet band at 620 eV and 631.46 eV, corresponding to the I 3d5/2 and I 3d3/2 states, signifying iodine integration into the perovskite structure. In air-exposed samples, a new doublet band at 618.75 eV in the I 3d region suggested the presence of an I–I bond.
As explained in the XRD section, exposure of CsPb0.9Ni0.1I3 samples to ambient conditions predominantly led to the formation of the non-perovskite phase δ-CsPbI3. XPS analysis further confirmed the presence of Cs–I, NiO, and Ni2O3 in the samples. The XPS and XRD data collectively suggest that CsPb0.9Ni0.1I3 perovskite samples undergo a transformation into non-perovskite forms such as δ-CsPbI3, CsI, NiO, Ni2O3, and I2 when exposed to ambient conditions.
The O 1s region shows a similar behavior to that in Ni-containing samples. Samples in a controlled environment exhibit a C–OH band, while exposure to air results in two new bands at 530.1 eV and 531.22 eV. These new bands are attributed to the formation of Mn–OH and Mn–O bonds, respectively. This change suggests that the sample undergoes decomposition and reacts with H2O and O2 molecules in the air, leading to the formation of these new chemical bonds. This evidence of alteration further underscores the sensitivity of CsPb0.8Mn0.2I3 to ambient conditions and the consequent chemical transformations.
The Cs 3d, Pb 4f, and I 3d regions for CsPb0.8Mn0.2I3 samples show the same behavior as in CsPb0.9Ni0.1I3. This suggests a consistent trend in the chemical states of cesium, lead, and iodine across different perovskite compositions when a portion of lead is substituted by either nickel or manganese.
For the Mn 2p region, the samples in a controlled environment exhibit a doublet band at 642.75 eV, which is characteristic of the Mn2+ oxidation state. However, after exposure to air, in addition to the Mn2+ bands, a broad band appears at 644.25 eV. This additional band is attributed to MnO, indicating the oxidation of Mn2+. This finding is consistent with the O 1s region, where the appearance of new bands (such as those for Mn–OH and Mn–O bonds) suggests the oxidation of manganese. The correlation between the Mn 2p and O 1s regions in CPMI samples highlights the oxidation of manganese upon exposure to air, underscoring the material's sensitivity to environmental changes.
Ultimately, this degradation leads to the formation of the more stable non-perovskite δ-CsPbI3 phase.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ma00034j |
This journal is © The Royal Society of Chemistry 2024 |