1. Research and analysis of metal oxidation states of perovskite materials
Perovskite semiconductors represent an emerging material platform for a variety of optoelectronic applications, including solar cells, LEDs, photodetectors, lasers, and thermoelectric devices. Although the unprecedented rate of performance improvements in this class of materials has thrust it into the spotlight, key challenges such as instability and toxicity remain. A key factor that is often overlooked in this field is device instability, or performance changes over time, which are closely related to changes in the oxidation state of the metal in the perovskite material. A deeper understanding of this issue could shed light on addressing key challenges of device instability and environmental effects. However, analysis of this issue is insufficient.
2. Introduction to results
Through comprehensive analysis, the study emphasized that B-site metal cations are the core of the redox process of perovskite semiconductors. Based on a selection of representative literature, the authors critically analyze the role of oxidation states in determining material and device properties, including the redox reactions leading to these states and the corresponding characterization methods. Finally, this study proposes ways to improve device efficiency and stability from the perspective of oxidation state control in halide perovskites, highlighting the potential of oxidation states as a versatile toolbox for realizing novel device functions.
3. Results and Discussion
Key Point 1: Metal Oxidation States in Perovskites: Redox Chemistry and Characterization Tools
The degradation and decomposition of halide perovskites is related to changes in the oxidation state of metal cations, known as redox reactions (Figure 1A). The reduction or oxidation tendency of different metal cations is determined by their reduction potential, as shown in a Frost diagram (also called an oxidation state diagram) (Figure 1B). From a thermodynamic point of view, the more negative the reduction potential (E0), the easier it is for the reaction to consume energy and make it less likely to occur. If E° is less negative or becomes positive, the reaction can occur more easily or spontaneously. The E0 of Ag+→Ag0, Bi3+→Bi0, Eu3+→Eu2+ are 0.80 V, 0.31 V and -0.35 V respectively (Figure 1B), indicating that Ag+ and Bi3+ have a tendency to reduce, and Eu2+ has a tendency to oxidize. For Pb2+, only a small amount of energy is needed to drive the reduction reaction due to its negative E0 (-0.126 V). However, other external factors may trigger redox reactions and even change the course of the reactions. For example, the E0 values of Sn2+ to Sn0 (-0.13 V) and Sn4+ to Sn2+ (+0.15 V) are not high, indicating that Sn2+ is easily oxidized or reduced. The presence of oxygen in the air makes oxidation more likely to occur. If more than one metal cation exists in the same perovskite material (such as Cs2AgBiBr6), the Gibbs free energy of the respective reactions can be used to predict the reduction (or oxidation) order of the metal cations. To accurately identify oxidation states, the most commonly used characterization tool is X-ray photoemission spectroscopy (XPS), which uses spectral fingerprints to determine oxidation states from shifts and splitting of core energy levels. There are two powerful techniques based on XPS: X-ray absorption near-edge structure (XANES) measurements and Auger electron spectroscopy (AES).
XANES measurements provide information on energy changes that occur with increasing oxidation state at the absorption site, and AES provides more sensitive oxidation state analysis. When combined with improved Auger parameters, AES can provide more reliable identification of oxidation states in molecules and solids that are less susceptible to energy drift caused by sample charging. The basic working principle of these spectroscopic methods is shown in Figure 2. As shown in Figures 2C and 2D, the incident electron beam interacts with the sample and excites the inner shell electrons. The transition of an electron from the outer shell to the inner shell of an atom releases an X-ray photon, which may subsequently excite another electron (called an Auger electron) to the vacuum energy level. For example, the oxidation state of tin (Sn) changes from Sn0 (Sn metal) to Sn2+ (CsSnI3), and then to Sn4+ (Cs2SnI6), with a 2.4 eV shift in its 3d nuclear level (Figure 2E). However, since the binding energy difference between Sn2+ (486 eV) and Sn4+ (486.6 eV) is small, distinguishing them can be difficult. AES technology can more directly characterize the chemical environment and oxidation state of the sample surface.
Point 2: Reduced metal cations in halide perovskites
External factors (light, electric field and thermal radiation) can induce reduction reactions of metal cations (Pb2+, Ag+ and Bi3+) and convert them into neutral metal states (Pb0, Ag0 and Bi0). Different oxidation states lead to different structures and physical properties of halide perovskites, affecting their device performance. These neutral metallic states often act as nonradiative recombination centers, reducing device efficiency. They can also cause material degradation, limiting the operational life of the device. The following authors summarize the origin of reduction-induced Pb0, Ag0, and Bi0 states, their impact on material properties, and possible solutions to mitigate these detrimental effects.
Lead halide perovskites: Metallic lead (Pb0) is formed during the degradation of lead halide perovskites (LHP) under light, which is called photoinduced decomposition (photolysis). Halogen vacancies can promote the transformation of Pb2+ in the perovskite structure into metal Pb0. Electrons can be captured by halogen vacancies before being recombined with holes or extracted. These electrons are then captured by Pb2+ close to the electron trap, resulting in the formation of Pb0. For example, due to the limited solubility of MAPbBr3, the introduced Br vacancies are located at the corners of the PbBr6 octahedron, resulting in insufficient coordination of Pb2+ ions at the perovskite grain boundaries and easy reduction to metal Pb0 (Figure 3A).
Inhibit the formation of Pb0 and improve device performance: Metal cations can serve as passivators for Pb0 defects. Studies have shown that the ion pair formed by europium ions Eu3+↔Eu2+ can serve as a "redox shuttle pair" and can simultaneously oxidize Pb0 and reduce I0 defects during the cycle transition, thereby significantly improving stability and high power conversion efficiencies (PCEs). During this redox process, the Eu3+-Eu2+ ion pair did not undergo significant depletion and was considered to be a redox transition cycle (Fig. 3C). Fe3+ additive has the same ability to oxidize metal Pb0, but the difference is that I0 is not reduced at the same time, thus preventing this cycle from occurring. In addition, it has been recently reported that doping RbCl can convert PbI2 into a relatively inactive compound (PbI2)2RbCl (PIRC), significantly reducing the formation of Pb0.
In addition to metal cations, organic additives are also used to prevent the formation of Pb0 in LHP. They are divided into different types according to their role in the redox process: the first type of additives act as oxidants, directly converting Pb0 into Pb2+, thereby inhibiting the formation of metallic Pb0. It has been reported that the introduction of benzoquinone (BQ) (Figure 4A) into the CH3NH3PbI3 film can significantly improve the operating life of the device, and hydrogen peroxide (H2O2) (Figure 4B) is used as a post-treatment to oxidize Pb0 on the perovskite surface. Improved photoluminescence quantum efficiencies (PLQEs). Reaction with spiro-OMeTAD was achieved by using 1-ethyl-3-methylimidazole bis(trifluoromethylsulfonyl)imide (EIm-TFSI) (Figure 4C). It penetrates into the HTM and forms a complex with excess PbI2 (EIm·xPbI2) in the perovskite layer, At the same time, metal Pb0 is oxidized to Pb2+ and I0 is reduced to I−. The second type of organic additives mainly inhibits the formation of Pb0 by reducing halogen vacancies. For example, the addition of hypophosphorous acid (Fig. 4D) can reduce oxidized I2 to I−, producing lower density metallic lead Pb0. Oleic acid, n-octylamine or n-octyl ammonium bromide (Figure 4E) can fix the Br− anion at the VBr− position of the octahedron on the surface of MAPbBr3, reducing the density of Br vacancies. At the same time, the fixed Br− compensates for the charge imbalance of the [PbBr6]4− octahedron and prevents Pb2+ from being reduced to Pb0. In addition, a novel phosphine as a point defect (Pb0) stabilizer was reported to improve the performance of PSCs. Partially oxidized tributylphosphine (TBUP) (Figure 4F) combined with additional TBUP oxide was found to be effective in reducing Pb0 and I0 concentrations during device preparation and operation.
Silver-based double perovskite
The source of Ag0 in lead-free double perovskite: Dark spots in Cs2AgInCl6 nanocrystals were observed by transmission electron microscopy (TEM) and found to be metallic Ag0. Initially, these Ag0 were attributed to electron beam irradiation; however, time-based reaction experiments revealed that metallic Ag nanocrystals were formed during the growth process. It was found that Ag0 in Cs2AgInCl6 could be detected by high-resolution X-ray photoelectron spectroscopy (XPS) before performing TEM measurements. Similarly, in Cs2AgBiBr6, even in a very weak reducing environment, there is a tendency to be reduced to metallic Ag0 due to the low reduction potential of Ag+.
Effect of Ag0 on lead-free double perovskite: As shown in Figure 5A, Ag0 nanoparticles (NPs) formed during the synthesis process are widely used for the growth of Cs2AgInCl6 nanocrystals. Silver species formed as amine ligands reduce Ag+ prior to halide implantation. The halide precursor is then injected into the reaction mixture, allowing the perovskite to rapidly form NCs on the silver seed crystal. The change in the dielectric medium around the silver NPs causes the local surface plasmon resonance peak to red-shift from 450 nm to 517.6 nm. UV-induced electrochemical Ostwald ripening demonstrated charge transfer from perovskite to metallic Ag NPs, leading to enhanced plasmon absorption and PL quenching (Figure 5B). Experiments demonstrate that the size of metallic Ag0 NPs in Cs2AgInCl6NCs becomes smaller and more uniform with increasing synthesis temperature (Fig. 5C), thereby improving PLQEs.
Point 3: Oxidized metal cations in halide perovskites
Oxidation can lead to changes in the oxidation state and physical properties of lead-free halide perovskites. B-site cations usually exhibit heterovalent oxidation states, which have a significant impact on their optical properties. For example, tin-based halide perovskites (THP) are promising candidates to replace toxic lead-based perovskites due to their similar electronic configuration. The band gap is ∼1.32 eV, and the theoretical PCE limit of CsSnI3 is about 33%. Compared with LHP, Sn2+ is easily oxidized to Sn4+ in THPs, accompanied by the formation of traps.
Tin-based halide perovskite
Source of Sn4+ in tin-based halide perovskites: Compared with traditional semiconductor materials (such as Si and GaAs) in which atoms are covalently bonded to each other, the chemical bonds in metal halide perovskites show predominantly ionic properties. It has been widely reported that Sn's lack of lanthanide shrinkage results in higher atomic energy levels, lower electronegativity, and smaller s and p orbital splitting compared to Pb, indicating that Sn in Sn-based perovskites is more susceptible to losing all valence electrons to form a higher valence state.
Oxidation of Sn2+ is widely observed in THP after exposure to air. The entire redox process can be divided into five steps. Initially, oxygen in the air reacts with THP to produce SnI4. Initially, oxygen in the air reacts with THP to produce SnI4, SnO2 and AI. Then, by reacting with O2 and H2O, the by-product SnI4 is rapidly converted into iodine; finally, the endogenously generated I2 helps to further oxidize Sn2+ to Sn4+ to form SnI4 (Figure 6A).
Effect of Sn4+ in THP: Doping density, carrier lifetime, mobility and diffusion length are some of the most important factors affecting the performance of THP optoelectronic devices. The redox reaction of Sn2+ to Sn4+ leads to p-type behavior in THP, seriously affecting the performance of THP solar cells.
Research shows that the B-site alloying strategy is an effective method to prevent Sn2+ from oxidizing to Sn4+. Due to improved defect tolerance, germanium-based perovskites are considered as promising alternatives to THPs. In addition, the incorporation of large A-site ammonium cations into the perovskite structure can prevent moisture penetration at the grain boundaries of the LHP film, increase the crystal formation energy, prevent oxygen from diffusing into the perovskite, and protect Sn2+ from oxidation. Organic additives can be used to prevent Sn2+ oxidation and improve air stability and equipment performance.
Europium-based halide perovskites: Europium-based halide perovskites A common and stable form of the rare earth element Eu is Eu3+, whose electronic configuration is [Xe] 4f6. Due to Coulomb repulsion, spin-orbit coupling, crystal field perturbation and Zeeman effect, the 4f orbital can usually be split into 119 sub-levels of different energies. In this case, Eu3+ ions usually exhibit a radiative transition from f orbital to d orbital, characterized by narrow-band red emission. In addition, Eu3+ exhibits a strong electric field shielding effect in most cases, so even in a double perovskite matrix, it is still able to transfer charges from the conduction band to the Eu3+ orbital. Therefore, Eu3+-based double perovskites, such as Cs2NaEuCl6, exhibit red emission near 600 nm under UV light and high-energy x-ray irradiation. Interestingly, once Eu3+ is reduced to Eu2+, the emission channel changes completely. When metallic Eu is exposed to ambient air, the formation of complexes containing a mixture of Eu2+ and Eu3+ ions is often observed. Furthermore, the newly prepared CsEuCl3 nanocrystals show blue emission at 435 nm. After several hours of exposure to air, red emission from Eu3+ orbital transitions (peaks at 593, 615, 650 and 698 nm) was observed.
4. Conclusion and outlook
In this review, the authors discuss the role of metal cation oxidation states in halide perovskite semiconductors, which are generally divided into two categories based on oxidation and reduction trends. The origin of oxidation state changes has been discussed from a chemical perspective with reference to Frost diagrams. Characterization tools for detecting oxidation state changes are introduced. Furthermore, the effects of oxidation state on film morphology, charge transport, and device performance are also analyzed. In addition, the authors discuss strategies for controlling redox reactions, including molecular additives, crystal structure engineering, and device design. This is particularly important for the development of high-performance lead-free perovskite devices. Further development in this direction will require the joint efforts of experimentalists and theoreticians. To obtain ideal oxidation states, reliable experimental protocols are needed to reproducibly prepare high-performance perovskite materials and devices. The close connection between the oxidation state of metal cations in halide perovskites and their optoelectronic properties promises to broaden the horizons of this emerging field.
5. References
Tang W et al. The roles of metal oxidation states in perovskite semiconductors， Matter
Doi: 10.1016/j.matt.2023.07.019（2023）.
https://www.sciencedirect.com/science/article/pii/S2590238523003776
E-mail: info@chemborun.com
Tel: +86-574-87178138 
No. 1558, Jiangnan Road,, Ningbo, Zhejiang, China (Mainland)/31
