Recent advancements in batteries and photo-batteries using metal halide perovskites

The ongoing energy crisis and increasing pollution from fossil fuels have led to high demand for renewable energy sources. Among the various clean energy sources, solar and wind are the most abundant forms, and energy harvesting devices, such as wind turbines, and solar cells have become important technologies for future energy generation. To utilize the electricity produced, however, there must be some energy storage systems, such as batteries and capacitors.^1,2 Therefore, finding energy storage materials that are efficient, robust, and low cost is a priority for the scientific community. To aid in this endeavor, we must improve our understanding of the mechanisms of storage and the properties, challenges, and future directions associated with these materials.

Batteries are the most common form of energy storage devices at present due to their use in portable consumer electronics and in electric vehicles for the automobile industry.^3,4 During the “materials revolution” of the last three decades, battery technologies have advanced significantly in both academia and industry. The first successful commercial lithium-ion battery (LIB) in 1991 offered an energy density of ∼200 Wh/L, only barely outperforming the dominant technology of nickel-metal hydride (NiMH) cells.⁵ By 2021, the state-of-the-art commercial LIBs could offer energy densities over 700 Wh/L. The global lithium-ion battery market is estimated to increase from USD 41.1 billion in 2021 to USD 116.6 billion in 2030, and it is expected to grow at a CAGR (compound annual growth rate) of 12.3% from 2021 to 2030.⁶ At the present time, most portable devices are powered by rechargeable LIBs⁷ due to this high energy density as well as high open-circuit voltage; robust, low-weight design; and low self-discharge current.⁸ To put it simply, Li-ion batteries hold a lot of charge in a small package, and once charged, they stay charged. The success of Li-ion has also generated interest in improving new battery types, such as lithium-sulfur, lithium-air, and solid-state batteries, for similar applications and with the promise of lower cost. However, at the moment, each of these new batteries types may be suited for niche applications, and none is as widely used as the lithium-ion battery.⁸ 

In addition to novel battery types, researchers are also exploring next-generation materials in LIBs to replace graphite and LiFePO₄, as the as the anode and cathode, respectively.⁹ As a one potential candidate, metal halide perovskites have been recognized among the many accessible materials due to their unique properties along with solid-state characteristics across metallic, superconducting, insulating, and semiconducting nature.¹⁰ Properties such as tunable bandgap, high charge carrier mobility, and long charge diffusion lengths when combined with cost effective manufacturing strategy allow for a wide range of photovoltaic and optoelectronic applications.¹¹ Because of their strong ion diffusion capabilities, halide perovskites are now also utilized in energy storage devices.^12,13 In recent reports, 1D benzidine lead iodide (Bz-Pb-I) hybrid perovskite (C₆H₉I₃NOPb) reported a high initial discharge capacity of 1580 mAh g⁻¹ at a current density of 100 mA g⁻¹ with a high reversible capacity of 646 mAh g⁻¹ and showed good stability when tested up to 250 cycles. A Cu-based 2D perovskite halide (CH₃NH₃CuBr₄) also showed a remarkable performance with a high reversible capacity of 630 mAh g⁻¹ at 100 mA g⁻¹ even after 140 cycles, confirming the long-term stability and structural robustness of the perovskite.¹⁴ 

This perspective will first cover the basic properties of metal halide perovskites, including the interaction of lithium ions with perovskite crystals and the mechanism of lithium-ion storage in batteries. Following that, different kinds of perovskite halides employed in batteries as well as the development of modern photo-batteries, with the bi-functional properties of solar cells and batteries, will be explored. At the end, a discussion of the current state of the field and an outlook on future directions are included.

Since 2009, metal halide perovskites (MHPs) have gained significant attention as the active material in solar cells.¹⁵ With excellent optical and electronic properties, the power conversion efficiency (PCE%) of single-junction perovskite solar cells has grown from less than 4%¹⁵ to an astonishing 25%¹⁶ in merely 10 years, while other applications, such as high performance photodetectors, light-emitting devices (LEDs), lasers, and transistors, are also well-reported.^17,18 The MHP materials exhibit unique properties, such as bandgap tunability, high defect tolerance, high photoluminescence yield (PLQY), and excellent ambipolar charge carrier properties (high carrier mobility and long diffusion length). MHPs can be easily prepared via many solvent and gas phase deposition techniques that, along with composition engineering, can be used to produce materials with many different dimensions and functionalities.^11,19,20

Ion migration was observed in solar cells and identified as a de-stabilizing factor that gave rise to hysteresis and phase-segregation,²¹ as well as spectral instability in LEDs.²² Although this was initially considered a negative feature of the soft metal halide perovskite structure, it is the tolerance to host extrinsic ions, such as Li⁺ or Na⁺, and the diffusion of ions in and out of perovskite crystals that is needed for energy storage applications.

The basic formula of a metal halide perovskite (often referred to as hybrid organic–inorganic perovskite or HOIP) is ABX₃, as shown in Fig. 1(a). A site is a monovalent cation, such as methylammonium (MA), formamidinium (FA), and cesium (Cs). B site is a divalent cation, generally being Pb and Sn and X groups are the halides, Cl, Br, I, or a mixture of them. The continuous network formed by the extension of corner-sharing octahedral BX₆²⁻ is why this material is described as a 3D perovskite_. The bandgap of halide perovskite could be easily tuned by varying A, B, and X-site element or a combination of them, as shown in Fig. 1(b). Over the years, the classic definition of ABX₃ has been challenged and updated with the emergence of other related material classes, such as low-dimensional 2D-perovskite, i.e., “Ruddlesden–Popper” phase^23,24 and “Dion–Jacobson” phase,²⁵ double-perovskites,²⁶ anti-perovskites,²⁷ and lead-free perovskites,²⁸ to replace the toxic Pb with more environmentally responsible elements. Several examples of newer perovskites and perovskite-like (or perovskite-inspired) materials are shown in Figs. 1(c) and 1(d).

To understand the use of perovskites in batteries, it is important to understand how the LIB works. Generally, electric power in a battery is stored in the form of chemical energy. In the case of LIBs, anode, cathode, and an electrolyte are the three main components. The anode is the source of lithium ions, whereas the cathode is the sink of ions (during discharging). The chemical reactions between the two electrodes have two components: electronic and ionic.²⁹ The ionic part (lithium ion) travels through the separator dipped in electrolyte between electrodes, whereas the electronic part (electrons) takes the route of the external circuit.³⁰ Most electrode materials are good ionic conductors, but the addition of an electronically conductive material, such as carbon black or “super P,” is required to assist the flow of electrons via the external circuit. A binder is added to make the electrode mechanically stable; the most common binder for LIBs is polyvinylidene fluoride (PVDF). During discharge, lithium ions flow from the anode to the cathode. This permits electrons to move from the anode to the cathode via the external circuit, resulting in current generation. The cathode pulls the lithium ions back to the anode during charging, causing electrons to flow in the reverse direction [Figs. 2(a) and 2(b)].

Outside the electrochemical window of the chosen electrolyte, cathode and anode materials can operate and cause side reactions that compromise battery performance. In addition to using a functional electrolyte with a large electrochemical window, one can also employ an electrolyte that can react with the electrode material to generate a high-quality solid electrolyte interphase (SEI) layer.³¹ The SEI is a protective layer that develops on the negative electrode of lithium-ion batteries as the electrolyte decomposes during the first cycle. The batteries’ performance, irreversible charge “loss,” rate capability, cyclability, and safety are all heavily reliant on the quality of the SEI.³² The electrochemically active and inactive components that are chosen influence the cell’s performance metrics and overall characteristics, making lithium-ion batteries a very versatile technology.³³ 

An electrode must be able to reversibly store as many ions as possible. The most commonly used anode in present day LIBs is graphite with a theoretical capacity of 372 mAh g⁻¹, whereas electrolyte consists of a lithium salt in an organic solvent.³⁴ Graphite has a Li⁺ diffusion coefficient ranging from 10⁻¹² to 10⁻⁵ cm²s^–1.^35,36 Silicon is another potential anode for LIBs due to its superior capacity. However, Si exhibits particle pulverization, shedding, and electrochemical performance failures during lithiation and de-lithiation. This arises due to significant volume change during cycling, which may result in mechanical fracture and electrical contact loss with the current collector, thus resulting in capacity fade.^37,38 Due to the poor conductivity of Si, the chemical diffusion coefficient of Li in Si, ∼10⁻¹³ cm² s⁻¹, is low, which limits the performance.³⁹ In comparison, chemical diffusion coefficients for Li⁺ in the CH₃NH₃PbBr₃ perovskite lattice are ∼10⁻⁷ cm² s⁻¹, implying that highly lithiated electrodes have ionic conductivities in the range of 10⁻³ Ω⁻¹ cm⁻¹, confirming the fast ionic diffusion in MHPs.⁴⁰ MHPs have been employed as anodes for lithium-ion batteries because of their capacity to store multiple lithium ions in a single unit cell, leading to the formation of Li_xABX₃.⁴¹ The ion diffusion process in the MHP expands the possibilities for its use in batteries by allowing numerous lithium ions to be stored in a single unit cell. The key challenges here include improving electrode stability and capacity at a low cost, and thus far, perovskites have been able to achieve these goals.

The cathode, usually formed from a variety of lithium-containing materials, is also crucial to the performance of a battery. Lithium cobalt oxide cathodes, which exhibits a high energy density of 150–190 Wh/kg, are used in mobile phone and laptop batteries. Because of costs and stability concerns, the material is unsuitable for big battery packs in electric vehicles. Subsequently, researchers have developed stable cathode materials by substituting nickel and manganese for cobalt.⁴² 

Metal oxides make up the majority of lithium cathode materials.²⁹ However, high production and precursor costs act as limiting factors. Metal halides, such as AgCl and CuCl₂, were utilized as the cathode for the batteries to minimize the manufacturing temperature. These batteries have poor performance because of material dissolution, which occurs following the first discharge. To stabilize the metal chloride, (EDBE) polyether molecules were inserted between CuCl₂ sheets to construct a 2D perovskite halide (EDBE)[CuCl₄] 2D as LIB cathodes.⁴³ 2D perovskite halides have a layered metal halide structure similar to the topology of LiCoO₂ (the most common cathode material used in LIBs) and graphite (most common anode); therefore, it has the potential to act as either cathode or anode.

The voltage of the cell is determined by the energy difference between the anode and cathode redox energies.⁴⁴ The product of a battery’s voltage and capacity determines its energy density; a material with a higher voltage and capacity generates a battery with a higher energy density. As a result, when the anode material is the same, the greater the cathode potential and capacity of the cathode material, the larger the energy density of the battery.⁴⁵ 

LIB progress in the future is contingent upon the discovery of new and more efficient electrode and electrolyte materials. The following sections discuss the major findings on the use of metal halide perovskites and other soft metal halides in advancing LIB technologies.

Various investigations have looked into the interaction between lithium ions and halide perovskites. Adding trace amounts of extra elements to a target lattice without affecting the host crystal structure can alter semiconductor properties. Doping is an effective way to modify a semiconductor’s properties without altering its crystal structure. It modulates charge carrier mobility and density and can improve its charge transport properties. Jiang et al. discovered that in a variety of halide perovskites (e.g., CH₃NH₃PbCl₃, CsPbBr₃), reversible electrochemical intercalations of lithium ion (Li⁺) and sodium ion (Na⁺) could occurs, resulting in n-type doping, which leads to improved conductivity.⁴⁶ Due to lithium doping, a change in color of perovskites was also noticed, as well as a 40% drop in transmittance in the wavelength range of 450–850 nm. The photocurrent of the lithium-doped CsPbBr₃ crystal is almost four times compared to the orange CsPbBr₃ crystal that is not doped.⁴⁷ Doping levels, or the light absorption level of doped crystals, are more likely to alter photovoltaic characteristics of different crystals. Increased light utilization appears to improve the photovoltaic performance of doped CsPbBr₃ (darkened) crystals.⁴⁷ 

Given the irreversible nature of full Li-intercalation (n = 1) and conversion reactions in the MAPbBr₃ system, it is understandable why the capacity faded rapidly after 30 cycles.⁴⁸ In comparison, metallic Pb has a theoretical specific capacity of 550 mAh g⁻¹ with a lithiated phase of Li_4.25Pb. However, most Pb-based anodes have lower capacity and cycle stability than expected, possibly due to structural degradation caused by the volume difference between the pure and lithiated phases (233% expansion from Pb to Li_4.25Pb).⁵¹ The ease with which MHPs can be synthesized, together with composition engineering, enabling the production of materials with a variety of dimensions is useful in here. With the emergence of new and novel material class (for example, 2D-layered and lead-free perovskites) for energy storage applications, it is important to establish throughout studies in terms of simulations and in situ experiments to fully capture the working mechanism of Li-ion interactions with perovskites and further improve battery performance.

In less than a decade, perovskite halides have shown tremendous growth as battery electrodes for energy storage.^52,53 The first report on the use of organometal halide perovskite for Li-ion storage was published in 2015 by Xia et al., where the synthesis of the active materials, CH₃NH₃PbI₃ and CH₃NH₃PbBr₃, was done by a hydrothermal method.⁴⁸ CH₃NH₃PbBr₃ showed an overall better performance with the first discharge capacity value of 331.8 mAh g⁻¹ and the 200th discharge capacity of 121 mAh g⁻¹, a capacity retention of 76.9%. By comparison, the iodide counterpart, CH₃NH₃PbI₃, exhibited a first discharge capacity of 43.6 mAh g⁻¹ at a current density of 200 mA g⁻¹ [Figs. 3(a) and 3(b)].

The high tolerance for volume expansion during repeated charge–discharge cycles (i.e., the maintenance of structural integrity while incorporating Li⁺) is one of the most critical material properties for commercialization in Li-ion battery technology.⁵⁴ Several studies in the field have shown that low-dimensional metal halide perovskites (MHPs) have better long-term performance stability than higher-dimensional MHPs^55,56 and operate better as electrodes for lithium-ion batteries because they allow for more intercalation into the open spaces between layers.^23,57,58 As recent reports concluded the robustness of lower dimensional perovskite halides,⁵⁹ Tathavadekar et al. probed into the performance of 1D C₆H₉I₃NOPb, 2D (C₄H₉NH₃)₂PbI₄, and 3D CH₃NH₃PbI₃ perovskites, showing how the performance improves on going from 3D to 2D to 1D.⁵⁸ The 3D perovskite halide showed a first discharge capacity of 476 mAh g⁻¹ and a first reversible capacity of 202 mAh g⁻¹ at 100 mA g⁻¹. The findings for the 2D and 1D examples are more striking and fascinating, with first discharge capacities roughly four times higher than the 3D case. The improved performance showing greater capacity values can be due to the layered structure that allows for better insertion of Li ions. The 1D benzidine mediated Bz-Pb-I showed potential as an anode material for Li-ion batteries as well as Na-ion batteries with the capacity of 646 and 315 mAh g⁻¹ for Li and Na-ion batteries, respectively. There are irreversible processes in the electrode material during the initial charge–discharge and SEI (solid electrolyte interphase) production that cause substantial decreases in the first discharge capacity to the first reversible capacity in all 3D-2D-1D situations.⁵⁸ Changes in composition also affects the battery performance, and the capacity improved when the amount of Br content increased in MAPbI_(3−x)Br_x from x = 0 (pure I) to x = 1 (pure Br).⁵⁷ 

The morphology and crystal size of the active material influences its performance as an electrode. As shown by Wang et al.,⁶⁰ five different sizes of MAPbBr₃ crystals were synthesized by varying the precursor concentrations, and their electrochemical performance showed that decreased defects led to better performance and increased capacity values. A smaller crystal size also delivers a higher specific area, and this increases the number of active sites leading to a higher carrier concentration due to more charge transfer at the electrode/electrolyte interface.⁶⁰ 

Notably, the most used electrolyte for perovskite halide-based Li-ion battery is 1 M LiPF₆ in carbonate-based solvents, where ethyl carbonate (EC) and dimethyl carbonate (DMC) are the most common solvents. The first reported all-inorganic metal halide nanocrystals electrodes in Li-air batteries used aqueous lithium chloride (LiCl) as an electrolyte, and 100 nm sized Cs₄PbBr₆ nanohexagons on ITO films coated with TiO_x as electrode.⁶¹ These devices showed better performance than their organic–inorganic counterparts. The films that underwent five cycles of annealing showed the best electrochemical performance due to improved electrochemical stability unlike the non-annealed and/or uncoated nanohexagon layers that showed poor stability after the first scan.⁶¹ The galvanostatic charge–discharge displayed a specific discharge capacity value of 377 mAh g⁻¹, with the capacity retention of 75% and deterioration of Coulombic efficiency from 100% to 98% after 100 scans. The same group reported CsPbBr₃ as promising electrodes grown on ITO substrates using the anti-solvent method to obtain homogeneous microcubes that were then covered with TiO_x deposited via pulsed laser deposition (PLD).⁶² CV scans performed up to 1500 cycles showed that the redox peaks for Li-insertion and Li-extraction overlapped over 500 cycles up to 1500 cycles, showing the structural stability of the perovskite.⁶² The charge–discharge cycles obtained at a current density of 45 mA g⁻¹ over 1500 cycles showed excellent performance. A capacity of 549 mAh g⁻¹ is maintained from the 100^th cycle to the 1500^th cycle, displaying the robust nature of the electrode.⁶² 

Jiang et al. proposed CsPbBr₃ as an active material for the LIB anode in 2017 with a first charging capacity of 94.8 mAh g⁻¹ and the cyclic life of 32 rounds at 60 μA cm^–246 A report explored carbon nanotubes (CNTs) as a conductive carbon-based composite part for the CsPbBr₃ electrode to improve both the electrochemical behavior and lithium storage capabilities.⁶³ Predominant pseudo-capacitive effect leading to higher stability and improved rate capability was observed with the synergistic effects of CsPbBr₃ and CNTs (carbon nanotubes).⁶³ The pseudo-capacitive effect refers to the ability of lead perovskite halides to store charge in two ways: (1) via Faradaic electron transfer by accessing two or more redox states of the metal centers [e.g., Pb(II) and Pb(0)] and (2) via non-Faradaic charge storage in the electrical double layer present at the surfaces of these materials.⁶⁴ CsPbBr₃@CNTs synthesized via co-precipitation showed that CsPbBr₃ retained its structure followed by the first discharge due to the effects CNTs as compared to the degradation that occurred after the first discharge in the absence of CNTs. The first capacity value obtained with CsPbBr₃@CNTs was reversible with the value of 644.6 mAh g⁻¹ at the current density of 100 mA g⁻¹, which is six times higher compared to that obtained for CsPbBr₃ and maintained the value above 500 mAh g⁻¹ for 90 cycles [Fig. 3(c)]. The use of CNTs decreased the electrode resistance as the cross-linked CNTs provide short pathways for faster electron migration. The usage of cube-shaped all-inorganic CsPbBr₃ perovskites as potential electrodes for LIBs using a gel polymer electrolyte was demonstrated experimentally and theoretically for half and full cells.⁶⁵ DOS calculations based on density functional theory (DFT) were used to determine the effect of lithium-ion intercalation on the electronic characteristics of perovskites. The electrochemical results of the fabricated half-cell via galvanostatic charge–discharge cycles showed a first specific discharge capacity value of 376 mAh g⁻¹ that still maintains a value of 334 mAh g⁻¹ after the 10^th cycle at a current density of 30 mA g⁻¹. Because the lithium ions were held in the perovskite host via intercalation, the crystal structure was retained, giving the batteries extraordinary stability.⁶⁵ 

3D CsPbCl₃ has recently been reported as an efficient anode for Li-ion batteries and dual ion batteries.⁶⁶ Specific discharge capacities of 612.3 and 275.2 mAh g⁻¹ are obtained for half-cells at current densities of 50 and 250 mA g⁻¹ with an average Coulombic efficiency of 88%. Ex-situ XRD characterization done during the discharge cycle showed peaks corresponding to several lithium-lead alloys (Li_xPb_y) and metallic lead (Pb) and, hence, concluded that the conversion reaction is much more prominent than the intercalation reaction in this perovskite.⁶⁶ 

The effect of tuning the layering properties of the quasi-2D Ruddlesden Popper (RP) layered perovskite series (BA)₂(MA)_n−1Pb_nX_3n+1 (BA = butylammonium, MA = methylammonium, and X = I^– or Br^–) from n = 1 to n = 4 was also investigated. The n = 4 bromide species (BA)₂(MA)₃Pb₄Br₁₃ provides the best LIB performance among the iodide and bromide-based perovskite structures with a first discharge capacity of 108 mAh g⁻¹ compared to 32 mAh g⁻¹ for the n = 1 structure at a current density of 30 mA g⁻¹.²³ The overall gravimetric performance is influenced by the halide employed in hybrid perovskite electrodes, with bromide-based perovskites having a higher overall gravimetric capacity due to the lower atomic mass of bromine. By adopting a narrower potential window of 2.8–1.8 V, an irreversible conversion reaction at 1.4 V during the first discharge is avoided, allowing the first electrochemical process to be evaluated at 2.1 V as the primary source of reversible lithium storage.²³ In general, improved electrochemical performance was observed when transitioning from 3D to 1D perovskite halides due to the 1D organic soft matter (benzidine) enveloping the Li– Pb alloys on all sides and resulting in sustained high performance.⁵⁸ 

Thus, great strides have been made in using perovskites for electrodes in just the last few years (see Table I). Although lead-based perovskites are among the most popular perovskite materials for batteries, it is also notable that toxicity is a concern that must be addressed. Lead-free perovskite halides can also be used for electrodes, but this research is still in its early stages.

Lead-based halide perovskites, as previously indicated, have exceptional capacity to operate as electrodes in lithium batteries. However, the toxicity of lead to humans and the environment is an important issue for both consumers and businesses. The use of large quantities of lead in consumer products, outside of a few long-held exceptions, is banned in many regions, and this could limit the marketability of this technology.⁶⁷ In the past, regulatory concerns have impeded the widespread application of other technologies based on heavy metals.

As a result, it is critical to investigate lead-free metal halide perovskites with appealing properties that are both stable and environmentally acceptable.⁶⁸ To overcome this problem, Jaffe and Karunadasa were the first to report the application of lead-free (EDBE)[CuCl₄] perovskites in Li-ion batteries, acting as the cathode of the cell.⁴³ The first generation of lead-free halide perovskites only achieved an initial discharge of 26 mAh g⁻¹ at 28 mA g⁻¹ current density and maintains at a stable capacity of 14 mAh g⁻¹ per 700 cycles.⁴³ Another experiment employed copper-based all-inorganic perovskite halides as anodes, with substantially better results. Pandey et al. reported that (CH₃NH₃)₂CuBr₄ 2D perovskites exhibit the first discharge capacity at 1800 mAh g^–1 at 100 mA g^–1 with a reversible capacity of 480 mAh g⁻¹ [Fig. 3(d)].¹⁴ Cs₂CuBr₄, in comparison, works as a 3D perovskite anode, and it also reaches a noticeably good result of first discharge capacity around 800 mAh g⁻¹ with a reversible capacity of nearly 420 mAh g⁻¹ at 100 mA g⁻¹. Even after 100 cycles, there was no reported variation from the charge–discharge behavior.¹⁴ The durability of the material at higher current densities was demonstrated by the rate performance of the 3D lead-free perovskite anode.¹⁴ 

Another lead-free class of materials, bismuth-based perovskite halides have recently attracted attention due to their low toxicity, high efficiency, and the ease of processing.⁶⁹ Bi is a very promising lead alternative due to their similar electronic structures and characteristics. It is isoelectronic with Pb (presence of 6s² electrons)⁷⁰ and exhibits a greater absorption coefficient but similar ionic radius to Pb²⁺.⁷¹ Roy et al. proposed bismuth-based organic–inorganic hybrid perovskite-like iodobismuthates with reversible lithium intake and release as anode materials for Li-ion batteries.⁷² The perovskites, namely, [C₃H₅N₂]₃[Bi₂I₉], [C₂H₄N₃S][BiI₄], and [C₃H₅N₂S][BiI₄], were synthesized and employed.⁷² In their report, each of them showed a specific discharge of 450, 520, and 230 mAh g⁻¹, respectively, after 250 cycles at a current density of 250 mAh g⁻¹.⁷² The systems delivered exceptional power density when exposed to fluctuating current density and excellent stability over 250 cycles. Figure 3(e) shows the rate performance of [C₃H₅N₂]₃[Bi₂I₉]. In a recent similar publication, Wu et al. proposed the use of all-inorganic lead-free sodium bismuth chloride double-perovskites, Cs₂NaBiCl₆, as the anode of a Li-ion battery.⁷³ Halide double perovskite materials with the formula A₂M(I)M(III)X₆ or A₂M(IV)X₆ may be considered to be stable and environmentally friendly alternatives for optoelectronic and energy storage applications, in which two divalent lead ions (Pb²⁺) are replaced by a combination of one monovalent [M(I)] and one trivalent ion [M(III)] or one tetravalent ion [M(IV)] and one vacancy site.⁷⁴ Physical grinding produced ultrahigh content Li⁺ doped Cs₂NaBiCl₆ powders with varying Li⁺ concentrations, which were shown to be promising anode materials.⁷³ The best specific capacity was reported by the battery with Cs₂NaBiCl₆:6Li⁺ as the host material (see Table II). With almost 99% Coulombic efficiency (which is the ratio of discharge capacity to charging capacity), the first discharge curve can reach 775 mAh g⁻¹ and stabilize at around 300 mAh g⁻¹ after 25 cycles at a 75 mA g⁻¹ charge density [Fig. 3(f)].⁷³ 

Recently, thin films of lead-free phenethylammonium bismuth iodide (PEA)₃Bi₂I₉ on fluorine-doped SnO₂-precoated glass substrates were investigated as the anode in aqueous Zn²⁺ electrolytes as a proof of concept, with a specific capacity of 220 mAh g⁻¹ at 0.4 Ag⁻¹ and capacity retention of nearly 100% after 50 scans.⁷⁵ Because of its large capacity, low fabrication and operation costs, and improved safety, the anode is a good candidate for use in Zn²⁺ batteries.⁷⁵ The stability and performance of lead-free constructed perovskite halide batteries suggest that they have a bright future in energy storage.

Clean energy production and storage have become essential technologies for powering the modern world. Among renewable resources, solar energy is perhaps the most crucial and promising candidate for future energy generation.¹ Recent developments in photovoltaics have pushed solar power to the forefront of green energy harvesting technologies; however, to realize the full benefit of solar cells, generation must be coupled to storage. Currently, batteries and capacitors can be connected externally to solar cells to create a bi-functional generation and storage system. However, external connection of the two devices leads to increased weight, higher cost, extra wiring, and ohmic losses.^77,78 Hence, materials that provide bi-functionality by concurrently harvesting energy and storing it have tremendous potential for future solar batteries. Research on multi-field integration of solar cells and batteries in a single device has led to various innovations in the field of solar-rechargeable batteries.^79–82 

The basic working principle of a photo-battery does not change much with the device type or structure. The photo-active electrode under illumination generates electron–hole pairs due to the photovoltaic effect. These generated electrons and holes contribute to the reactions occurring through the charge/discharge cycle. In the various photo-battery structures, three-electrode structure consists of a common electrode configuration, which forms the bridge between the photovoltaic (PV) cell and the battery. The two-electrode system consists of a photo-active electrode that generates energy and stores it.⁷⁹ Most recently, photo-active electrodes are being explored in Li-ion^83–85 and Zn-ion batteries^86,87 that show huge capability for further development in the future. In electrode materials, such as LiMn₂O₄, light does not contribute to the electrochemical performance of the battery but does dramatically increase the charging rate in a conventional LIB.⁸⁸ 

Perovskite halides are already important to the fields of photovoltaics⁸⁹ and energy storage and are now also being considered as photoactive materials for photo-batteries. This is attributable to the same properties that make them ideal for PVs and batteries: a tunable bandgap, high charge carrier mobility, a low non-radiative recombination rate, a broad absorption spectrum, long charge diffusion lengths, and small carrier effective masses properties previously described.

Ahmad et al. demonstrated the use of 2D lead-based perovskites, namely, (C₆H₉C₂H₄NH₃)₂PbI₄, as a photo-active electrode material in a lithium-ion battery [Figs. 4(a) and 4(b)].⁹⁰ The battery with the iodide perovskite showed a specific capacity up to 100 mAh g⁻¹ at 30 mA g⁻¹. With reduced graphene oxide (rGO) as the conductive additive [Fig. 4(c)], it also exhibited photo-charging under illumination, without an external load, over the voltage range of 1.4–3.0 V. This was followed by discharge over 25 h using a 21.5 kΩ resistor as load, and thus, the device functioned as a true photo-battery.⁹⁰ When the voltage range was extended to a voltage lower than 1.4 V, an irreversible degradation of the perovskite due to the reduction of Pb²⁺ to Pb⁰ occurred. The battery also showed a higher potential output when discharged under light vs dark. With a 21.5 kΩ resistor as load, Ahmad et al. obtained a photo-conversion efficiency of 0.034%.⁹⁰ A recent study by He et al. provides a hypothesis using polarons to explain the mechanism of photo-rechargeability.⁹¹ Most notably, the displacement of the lithium ion is shown to be available upon the production of a hole polaron, which is thought to represent the photo-charging behavior that initiates the unidirectional movement of lithium ions on the electrode surface.⁹¹ This demonstrates that MHPs are effective electrode materials and can serve as the active layer for photo-charging in photo-rechargeable perovskite batteries.

While 2D and organic-inorganic hybrid perovskite are popular for lead-free electrodes, bulk perovskites and all-inorganic materials can also be used. Recently, Tewari and Shivarudraiah used an all-inorganic lead-free perovskite halide, with Cs₃Bi₂I₉ as the photo-electrode, to fabricate a photo-rechargeable Li-ion battery.⁷⁶ Charge–discharge experiments obtained a first discharge capacity value of 413 mAh g⁻¹ at 50 mA g⁻¹; however, the capacity declined over an increasing number of cycles due to the conversion of Bi³⁺ to Bi⁰ in the perovskite structure.⁷⁶ The photo-active electrode was then fabricated on either an FTO coated glass substrate or a porous carbon felt current collector to study the performance of the battery under light, and the photo-charging mechanism showed the perovskite to be the main contributor.⁷⁶ In this device type, the photo-batteries, under illumination by photons with energy greater than the bandgap of the perovskite, produces photo-generated electrons that flow via PCBM to the current collector [Fig. 4(d)]. This leads to an excessive number of holes on the perovskite matrix, leading to repulsion of Li⁺ ions back into the electrolyte.^84,90 The fate of photo-generated electrons is less clear. In the work of Ahmad et al., electrons are depicted as traveling back through the external circuit, during electrical charging, to the counter electrode where it reduces the Li⁺ ions back to Li metal. Another pathway for photo-generated electrons, in the absence of an external circuit, would be to the formation of a reactive oxygen species via electron transfer to the ethylene carbonate/dimethyl carbonate electrolyte, ultimately leading to the formation of SEI on the Li⁰ metal.⁹² In the work of Tewari and Shivarudraiah, the battery was discharged to 0.9 V and then photo-charged to ∼2.5 V under light, without an external load [Figs. 4(e) and 4(f)].⁷⁶ They reported a photo-conversion efficiency of 0.43% for the first photo-charge/discharge cycle followed by an average photo-charging efficiency of ∼0.1% for the next three cycles.⁷⁶ These findings suggest that Cs₃Bi₂I₉ perovskites have a promising future ahead and that further improvements in photo-efficiency can be achieved through better understanding of the charge separation and transport phenomena related to photo-charging. Photo-conversion efficiency is a key parameter to evaluate the performance of photo-batteries. For two-electrode configuration photo-batteries, it can be calculated by the following formula: η = E_output/E_light × 100%.⁹³  E_output refers to the discharging energy from the device and E_light refers to the input energy from light. Table III summarizes these findings, as well as recent reports on photo-batteries of a similar device type that employs alternate photo-electrodes.

Apart from the perovskites, Bouteau et al. studied the effect of illumination on the stability of standard battery electrolytes themselves, most commonly 1 M LiPF₆ in EC/DMC.⁹⁴ After 100 hours of exposure to light at A.M. 1.5 G, the electrolyte exhibited excellent photo-stability, with no discernible deterioration or modification. However, POF(OH)₂ and POF₂(OH) were detected by NMR after 200 hours of light exposure as a result of LiPF₆ hydrolysis.⁹² The UV absorption spectra highlight that solvated LiPF₆ ions reorganize with cyclic carbonate solvents, in this case ethylene carbonate and propylene carbonate, over time and that this stems from hydrogen bonding.⁹⁴ These experiments were conducted in the absence of any electrode; therefore, full battery tests are required to investigate the effect on photo-battery performance.

Much of the literature on light-assisted battery charging has focused on dye-synthesized solar cell (DSSC) technology, which can be paired with energy storage devices.^95,96 The most often used dye reported in photo-assisted battery applications is N719, one of several ruthenium dyes common to the field of DSSCs;^92,97,98 however, in terms of light harvesting and solar energy conversion capability, halide perovskite materials such as CH₃NH₃PbI₃ outperform organometallic dyes. A recent report replaced N719 with a perovskite material CH₃NH₃PbI₃ to act as a dye to activate the photo-charging properties of the lithium-ion battery electrode material LiFePO₄.⁹⁹ Prior to photo-excitation, lithium ion species diffuse from LiFePO₄ to CH₃NH₃PbI₃. The photo-active perovskite material was shown to generate electrons and holes due to photoexcitation, and the excess holes in the LiFePO₄ repel the diffused lithium ions to the counter electrode resulting in photo-charging.⁹⁹ 

Perovskites have shown considerable promise as energy harvesting and storage materials since they are multidimensional (0D, 1D, 2D, and 3D) materials. These structural variations result in the occurrence of a wide range of electrical, electronic, optical, and chemical properties.⁵³ Understanding crystallographic arrangements is important for the structure and stability of perovskite halides. Several strategies have been used to improve the perovskite material’s properties for energy storage applications, including tuning the dimensions, doping, and defect passivation. Low-dimensional MHPs perform better as lithium-ion battery anodes than high-dimensional MHPs due to increased intercalation between layers.^14,58 Controlling the morphology of the perovskites has been shown to be crucial for maximizing performance by enhancing Li-ion diffusion.¹⁰² Due to the increase in exposed area for the holes, porous morphologies with a larger surface area increase the intercalation affinity for Li ions and decrease the diffusion length of Li ions. Molecular simulations also suggest that macroscopic attributes, such as conductivity, strongly depend on the material’s morphology.⁴⁰ Simple defect modification, via cationic disordering, amorphization, doping, partial cation reduction, and manipulation of intrinsic defects, can considerably increase ion intercalation by directly altering chemical properties of the material.¹⁰³ MHP devices can have their bandgap, conductivity, charge carrier concentration, and mobility all adjusted via doping, which has a substantial impact on performance. The successful incorporation of the trivalent cation Bi³⁺ into the MAPbBr₃ lattice was reported using the retrograde solubility of perovskites where Bi could be utilized to modify a range of optical and electrical properties in the hybrid perovskite crystal.¹⁰⁴ Similarly, insertion of lithium ions leads to n-doping of halide perovskites, which results in increased conductivity in the perovskite structure.⁴⁰ The way ahead for this technology is clear: to enhance the charge/discharge capacity, to find new combinations of perovskites and analogs, and to find the optimum parameters to enhance the efficiency and stability.

Furthermore, regardless of the quick pace of development regarding perovskite devices, the actual mechanism for how these batteries function, in particular, photo-rechargeable batteries, requires further investigation. He et al. recently proposed that it is the distribution of polarons over the dipolar cationic sites that forces the lithium ions out during charging.⁹¹ Similarly, the dynamics of the electrons are somewhat ambiguous.^90,92 These mechanistic studies are critical to designing better devices with improved metrics. It improves understanding on how perovskite photo-batteries work microscopically, it also provides insights of feasible methodologies when investigating the mechanisms involved in other similar perovskite-based batteries.

Overall, many challenges remain for implementing perovskites in commercially viable Li-ion batteries or photo-batteries. LIBs are well studied and already commercially available, so disruptive discoveries are less likely than for other battery types, such as zinc or sodium ion. Fortunately, work done on perovskite LIBs applies well to many other ion and air battery types. Future innovations in perovskite batteries, at this time, hinge upon finding new perovskites with favorable activities. The discovery of materials that are feasible for photo-batteries, as opposed to normal batteries, has greatly improved the prospects of using perovskites for charge storage in these bi-functional generation and storage devices. However, the efficiencies, especially the PCE, must improve to the range of 5%–10% before this will be competitive with wired systems. Competitive technologies using non-perovskite oxide photo-electrodes also exist, such as vanadium pentoxide, vanadium dioxide, and molybdenum disulphide-zinc oxide currently boast PCEs of 1.2%, 0.18%, and 1.8% for zinc-ion batteries (under 455 nm light source).^87,100,101 Finally, with improvements in theory for this class of devices, we will be able to use simulated results to speed up the analog screening processes. Additionally, there may be opportunities for artificial intelligence (AI) based screening to design new active perovskite materials for perovskite-based, similar to work done on LEDs and solar cells. In any case, this is an interesting time for the nascent field of perovskite batteries, and one can expect the number of publications to grow substantially in the next few years.

All authors acknowledge support from the Research Grants Council/University Grants Committee (RGC/UGC) via GRF Grant No. 16306020 and from the School of Science and Department of Chemistry at HKUST via Grant Nos. IRS21SC07 and SBI19SC01.

The authors have no conflicts to disclose.

The data that support the findings of this study are available within the article.