In Situ TEM Studies on Electrochemical Mechanisms of Rechargeable Ion Battery Cathodes

Due to recent developments in secondary ion batteries for high‐energy‐density applications, a thorough understanding of the underlying mechanism of advanced cathode materials is of vital importance. In situ transmission electron microscopy (TEM) techniques capable of high spatial and temporal resolution in operando analysis of dynamic battery systems have attracted significant interest. However, the complex electrochemical reaction mechanisms of cathode materials have not been extensively investigated using in situ TEM due to technical difficulties in implementation. This perspective provides an overview on the recent development of in situ TEM for studying representative layered and olivine‐type cathode materials in secondary ion batteries; it further discusses the critical challenges and possible breakthroughs of this technique in deepening fundamental understandings in the near future.


Introduction
In the course of the development of today's society, environmental pollution and climate change caused by the rapid consumption of fossil and mineral fuels have prompted the urgent need for advanced energy conversion and storage. Among the many energy storage technologies, secondary ion battery can effectively realize the rapid conversion or storage of chemical energy and electric energy by virtue of its own energy storage and conversion function and can be flexibly applied to various scenarios in reality. [1] It is essential to have a thorough understanding of the complex electrochemical reaction mechanisms for the development of advanced electrode materials in secondary ion batteries. Of particular importance is the cathode materials, which play a critical role in determining the overall property/performance of secondary ion batteries. Among the many cathode materials developed for far, the layered material has the outstanding characteristics of high capacity and high energy density, [2] and the olivine-type material has the advantages of low cost, avirulent, and high safety. [3] There are a number of techniques available to characterize and analyze the charge/ discharge mechanisms of electrode materials. Recently, in situ characterization techniques such as in situ X-ray diffraction, [4] in situ X-ray absorption spectroscopy, [5] and in situ Raman [6] have been widely used to study the reaction mechanisms of electrode materials in batteries. These techniques could reveal either the structural or the compositional information of materials of interest at the bulk level but fall short in correlating such information with the microscopic characterization of high spatial resolution. In situ scanning electron microscopy has been extensively utilized to study multiscale phase transformation and electrochemomechanical topics in batteries, [7] which, however, suffers not only the intrinsically poor resolution but also the limited structure-analysis capability. In this sense, in situ transmission electron microscopy (TEM), capable of visualizing material's dynamic processes with high spatial and temporal resolution, is gaining growing interest. [8] In situ TEM has been applied in the field of catalysis. For example, He et al. tracked the catalytic behavior of nanoscale Cu in CO 2 atmosphere by in situ gas TEM holder, via which the working/failure mechanisms of Cu for CO 2 reduction were clearly disclosed at the microscopic level. [9] In addition, a variety of TEM techniques are widely used in solid-state battery field, among which cryo-TEM and in situ TEM have shown special roles in revealing the dendrite growth problem of Li metal anode [10] and understanding the mechanism of solid-solid interface, [11] respectively.
In the past decade, there have been hundreds of reports utilizing in situ TEM for battery material study, mostly on the anode side due to the electrode materials with nanowire morphology which are very suitable for in situ TEM studies. [12] In situ observation of the intercalation-based reaction in cathode material at microscale has high requirements for spatial resolution, sample thickness, and stability, while most cathodes used in rechargeable ion batteries are micron-scale bulk materials (large and thick particles). And the stress generated after applying bias will also affect the high-magnification imaging of intercalation-based reaction. Furthermore, the irreversible conversion reaction occurs immediately upon contact between the commercial DOI: 10.1002/sstr.202300001 Due to recent developments in secondary ion batteries for high-energy-density applications, a thorough understanding of the underlying mechanism of advanced cathode materials is of vital importance. In situ transmission electron microscopy (TEM) techniques capable of high spatial and temporal resolution in operando analysis of dynamic battery systems have attracted significant interest. However, the complex electrochemical reaction mechanisms of cathode materials have not been extensively investigated using in situ TEM due to technical difficulties in implementation. This perspective provides an overview on the recent development of in situ TEM for studying representative layered and olivine-type cathode materials in secondary ion batteries; it further discusses the critical challenges and possible breakthroughs of this technique in deepening fundamental understandings in the near future.
layered LiCoO 2 and Li metal without bias, [13] which also poses a challenge for in situ TEM electrochemical research. Due to possible technical difficulties, there are not many in situ TEM studies on the electrochemical behaviors of cathode materials of secondary ion batteries, leaving behind a huge gap between fundamental understanding and practical optimization. This essay reviews recent research on the technical advances of in situ TEM in studying the electrochemical mechanisms of cathode materials in secondary ion batteries, using layered and olivinetype materials as two representative examples, with an aim to provide perspectives of this technique in boosting further development of cathode materials in the near future.

In Situ TEM Technique for Cathode Material Research
In situ TEM technique can provide not only atomic-scale visualization of thin samples and diffraction analysis on local phase features of materials but also yield key chemical information using analytical tools such as electron energy loss spectroscopy (EELS) and energy-dispersive X-ray spectroscopy. Among many in situ TEM studies focusing on cathode materials, two kinds of in situ hardware setup have been developed, namely the solid-state and liquid-state in situ TEM, which are further discussed in the following paragraphs.

Solid Cell
At present, the solid cell in situ TEM technique monitoring the dynamic electrochemical reaction process of cathode materials can be roughly divided into two setups. One setup is to construct a microscopic solid cell in TEM using TEM-scanning tunneling microscopy probe holder. The cathode material coated with solid electrolyte was cut and polished by focused ion beam (FIB) and installed at one end of the holder. Then, the probe tip with anode was connected to the electrolyte with a subsequently applied bias to mimic the real-time dynamics (probe-type, Figure 1a). Another setup is to utilize chip-based microelectromechanical system (MEMS), which is widely used in electrical, thermal, [14] thermoelectric, [15] liquid, [16] and gas [9] in situ testing in TEM, to fabricate a microscopic solid cell between two electrodes of the thermal or electrical MEMS chips by FIB milling technology (chip-type, Figure 1b). Compared with probe-type solid cell, chip-type solid cell incorporates a more reliable interfacial contact between electrode material and solid electrolyte, which has a larger contact area and smaller interface impedance. The extraction of Li þ /Na þ and the migration of transition-metal ions in all-solid-state secondary ion batteries can be directly observed by using the two in situ TEM techniques mentioned earlier.

Liquid Cell
In general, most liquids (water and other organic solvents) would instantaneously gasify in the high vacuum environment of TEM due to their high vapor pressure. Huang et al. used ionic liquid with low vapor pressure as the solvent of lithium salt in TEM to build the world's first liquid-state nanobattery in 2010. [17] A similar in situ TEM technique can also be used for the electrochemical study of cathode materials in open liquid cell. As shown in Figure 1c, one drop of the ionic liquid electrolyte (ILE) was placed on the top surface of the Li 4 Ti 5 O 12 crystal anode. On the other side, LiMn 2 O 4 nanowires were bridged between Pt current collector and ILE. [18] In this way, the cathode material can be observed by in situ TEM during the charge and discharge cycles.
To explore the solid-liquid reaction mechanism of cathode materials in most conventional electrolytes, such as LiPF 6 in ethylene carbonate and dimethyl carbonate, a sealed liquid cell can also be used. As shown in Figure 1d, a sealed liquid cell requires the presence of a thin Si 3 N 4 film-coated window, which not only Figure 1. a,b) Experimental setups for solid-solid interaction. a) Schematic illustration of in situ probe-type solid cell by biasing to induce charge and discharge. Reproduced with permission. [24] Copyright 2020, Wiley-VCH. b) Schematic illustration of in situ chip-type solid cell. Reproduced with permission. [22] Copyright 2017, American Chemical Society. c,d) Experimental setups for solid-liquid interaction. c) Schematic illustration of in situ probe-type liquid cell by biasing to induce charge and discharge. Reproduced with permission. [18] Copyright 2013, American Chemical Society. d) Schematic side-view cross-section of the in situ electrochemical liquid cell. Reproduced with permission. [19] Copyright 2018, American Chemical Society. allows electron beam to transmit through with limited inelastic scattering but also seals the high vapor-pressure electrolytes to protect the high-vacuum TEM chamber. The cell is also built with a working electrode, a counter electrode, and a reference electrode, which are connected to an external potentiostat. [19] Liquid cell TEM has its unique advantages of direct observation of cathode materials transformation dynamics in liquids with high spatial resolution (down to the atomic range) and high temporal resolution (in milliseconds).

Layered
Layered cathode materials possess high specific capacity for secondary ion batteries in electric vehicle and grid storage applications. However, electrochemical cycling of layered cathodes often results in the accumulation of various defects such as dislocations, [20] cracks, [21] and the kinetic hindrance of ion intercalation induced by phase transition and grain boundary, [22] leading to capacity fade and premature failure of batteries. Understanding these degradation mechanisms is critical to improve the performance of secondary ion batteries. LiCoO 2 , as the first commercialized layered transition metal oxide material, is still one of the mainstream cathode materials in the field of portable devices. However, few studies have focused on the structural evolution of LiCoO 2 in all-solid-state lithium-ion batteries (LIBs) at high-voltage condition, and the physical mechanism for the high Li þ transfer impedance is still unclear. Electrochemical delithiation can be observed at atomic-scale in a solid cell using a MEMS chip-based in situ TEM holder and FIB milling to prepare a sample (Figure 1b  and 2a). It demonstrated the feasibility of LiCoO 2 delithiation in all-solid-state LIBs and the formation of nanosized polycrystal connected by coherent twin boundaries and antiphase domain boundaries after high voltage delithiation, which was assisted by an aberration-corrected scanning transmission electron microscopy ( Figure 2a). [22] The probe-type solid cell can be coupled with in situ EELS to analyze the interfacial impedance between LiCoO 2 and solid electrolyte. The in situ EELS spectrum revealed a disordered interfacial layer between LiCoO 2 and lithium phosphorus oxynitride (LiPON) that would eventually evolve to rock-salt CoO after cycling and proposed Li 2 O/Li 2 O 2 formation as an intermediate compound of oxygen evolution reaction (Figure 2b). [23] The presence and increase of disordered layer would lead to rapid capacity decay, so it is critical to solve interfacial issues for better safety and long-term cycling.
Ni-rich LiNi 1ÀxÀy Mn x Co y O 2 (NMC) and Li-rich layered compounds are both the cathode materials for LIBs with high energy density. However, these cathodes presently suffer rapid capacity fading and power loss upon cycling, and the understanding in the interfacial reactions between electrode and solid electrolyte is insufficient in all-solid-state LIBs. It thus demands a high spatial resolution observation from the same sample region, via which one can distinguish local atomic structure changes. The structural evolution of a Ni-rich NMC cathode during delithiation can be investigated by building a solid cell inside TEM with a biasing TEM holder. The atomic structure imaging confirmed the formation of secondary phase (spinel, rock-salt) and grain boundary (antiphase boundary, twin boundary) inside the particle Figure 2. Recent advances of layered cathode on energy storage mechanism by in situ TEM. a) Grain boundary of LiCoO 2 during delithiation studied by a chip-type solid cell. Reproduced with permission. [22] Copyright 2017, American Chemical Society. b) Chemical and electronic structure evolution at the interfaces between LiCoO 2 and LiPON by in situ EELS. Reproduced with permission. [23] Copyright 2016, American Chemical Society. c) Interior structural changes before and after delithiation of Ni-rich NMC studied by a probe-type solid cell. Reproduced with permission. [24] Copyright 2020, Wiley-VCH. d) The delithiation-induced nanovoid formation in Li-rich LNMO studied by in situ TEM. Reproduced with permission. [25] Copyright 2022, American Chemical Society. e) In situ electrochemical cycling of a Na-NMC cathode. Reproduced with permission. [26] Copyright 2021, American Chemical Society. of Ni-rich NMC, which was seen to be aided by crystallographic defects in all-solid-state battery (Figure 2c). [24] These indicate that the inhomogeneous lithium extraction and transition metal migration caused by defects inside the bulk play an important role in the degradation of the Ni-rich NMC material. In addition, the same in situ technique was also used to observe the interfacial reaction of Li-rich Li 1.2 Ni 0.2 Mn 0.6 O 2 (LNMO) and LiPON (solid electrolyte) in a solid cell (Figure 2d). [25] Nanopores are inborn near the LNMO/LiPON interface because of the inherent chemical instability between the cathode material and electrolyte. Such chemical reactions also spontaneously triggered oxygen loss, structural reconstruction, and severe cation mixing, resulting in high interfacial impedance at the cathode surface. The biasing-induced nanovoids are formed inside LNMO during delithiation and grow bigger by merging with others when more Li þ are removed, indicating that the porosity development without cation mixing can be a possible degradation mechanism in Li-rich-layered cathode materials. Layered cathode materials are also commonly used in sodiumion batteries, such as P2-type Na 0.7 -Ni 0.3 Mn 0.6 Co 0.1 O 2 (Na-NMC). It is well known that cracks have a great effect on the degradation mechanism of cathode materials, but the generation mode of cracks under electrochemical cycling during battery operation remains unknown. A microscopic probe-type solid cell assembled using Na-NMC as cathode and Sn as anode was utilized to analyze the crack generation process at atomic-scale (Figure 2e). [26] The in situ TEM tests directly demonstrate that delamination cracks and kinks form during desodiation rather than sodiation, and wide cracks were not formed immediately upon cycling but resulted from cumulative growth over many cycles.

Olivine Type
Olivine-type LiFePO 4 has attracted much attention for its application as a high-safety cathode material in LIBs, and its performance is also being gradually optimized. Furthermore, various opinions and explanations on the electrochemical behavior of LiFePO 4 were proposed from the extensive investigation over the past years. Thereinto, a major debate focuses on the role of the metastable solid solution reaction and its competing counterpart, the phase transition mechanism, during battery charging and discharging. Obviously, real-time atomic-scale observation of lithiation/delithiation of LiFePO 4 by in situ TEM is critical to elucidate the underlying mechanisms. The mechanism of the phase boundary migration and the anisotropic lithiation of FePO 4 microparticles can be clearly observed at atomic-scale by assembling a probe-type solid cell. The phase boundary between FePO 4 and LiFePO 4 was clearly shown to align along the (010) plane and move towards the [010] direction, which is the same as the Li þ diffusion direction (Figure 3a). [27] In addition, in situ high-resolution TEM can also observe the disordered metastable solid solution formed Figure 3. Recent advances of olivine-type cathode on lithiation/delithiation mechanism by in situ TEM. a) Phase boundary migration of FePO 4 during lithiation studied by a probe-type solid cell. Reproduced with permission. [27] Copyright 2013, Wiley-VCH. b) Dynamic evolution of the solid solution zone was captured during the delithiation of a LiFePO 4 crystal. Reproduced with permission. [28] Copyright 2014, American Chemical Society. c) Inhomogeneous competitive reactions among LiFePO 4 particles revealed by time-sequential energy-filtered TEM in a chip-type liquid cell. Reproduced with permission. [29] Copyright 2014, American Chemical Society. d) Mechanism among LiFePO 4 crystal revealed by electron diffraction tomography throughout the charge/discharge cycle in a sealed liquid cell. Reproduced with permission. [19] Copyright 2018, American Chemical Society. rapidly and existed stably during the delithiation of LiFePO 4 and further reveal there are no permanent dislocations occurred in solid solution zone, which means that such coherent interfaces have higher Li þ diffusion mobility (Figure 3b). [28] Compared to other in situ or ex situ techniques, real-time dynamic observations of phase transition and phase boundary migration indicate powerful and unique capability of in situ TEM in detecting the localized reaction mechanisms. Sealed liquid cell TEM technique combined with spectroscopy also has unique advantages to observe and study the degradation mechanism of LiFePO 4 and the kinetics of Li þ insertion and removal in real-time. For example, quantitative electrochemical analysis in sealed liquid cells can be realized by combining energy-filtered TEM (EFTEM) spectroscopic imaging at nanoscale processes in electrochemical reaction. Elastic 0 eV EFTEM images revealed that LiFePO 4 gradually suffers mass loss during the course of five cycles, and it was also found that different LiFePO 4 particles show the competing delithiation mechanisms under the same conditions (Figure 3c). [29] Furthermore, Karakulina et al. have successfully demonstrated that electron diffraction tomography data can also be collected in situ in sealed liquid cell ( Figure 3d). [19] This opens many possibilities for studying the structure solution of nanoparticles that undergo structural changes or crystallization in a liquid environment using in situ TEM techniques.

Opportunities and Future Outlook
In situ TEM technique has been verified as a powerful tool to detect the electrochemical reactions of cathode materials in real-time. These nanobattery configurations can monitor the dynamic reaction process of ion and electron transport in realtime at high spatial resolution and simultaneously conduct structural and chemical analyses. However, whether in all-solid-state or liquid batteries, the in situ TEM techniques still have many issues demanding further modification/advancement toward more intensive and extensive understanding of cathode materials.

Intergranular Charge/Discharge Behaviors of Secondary Particles
For nickel-rich secondary particles that are polycrystalline, it is difficult to study the interacting behaviors (such as intergranular ion transport/crack formation) between primary particles by in situ solid cell TEM. This difficulty arises from the intrinsically weak bonding between the internal primary particles, which can hardly maintain an intact interface during the in situ TEM sample preparation. Particularly, the indispensable use of FIBmilling method for sample thinning down could significantly damage the originally weak particle-particle contact within one secondary particle by eliminating most of the contact area. This problem results in the challenging characterization of any crack propagation across the grain boundaries, which is critical to understand the structural failure of cathode materials upon cycling. To mitigate this issue, either more delicate in situ TEM sample preparation techniques should be developed, or the secondary particles themselves should somehow gain enhanced interfacial contact to make the grain-grain bonding more sustainable.

Cathode-Electrolyte Contact Limitation and Future Design
The limitation of studying the electrochemical mechanism of cathode materials in probe-type nanobattery is that the interface contact mode is based on point contact, which has limitation on the ion diffusion pattern. This brings an obstacle to explore the material degradation mechanism caused by possible ion intercalation/deintercalation in the battery system during the actual operation. Chip-based solid cell can only study the delithiation process rather than the full lithiation-delithiation cycling process for cathode, since Li metal is very active and can be hardly prepared as the anode of a nanobattery by conventional FIB milling. The cryo-FIB is a potentially useful tool for microfabrication of active materials that are beam-sensitive. Moreover, to minimize air exposure of the Li metal, the vacuum sample-transfer technique should also be developed when transferring specimens into TEM. These techniques provide the possibility for the assembly of microscopic all-solid-state Li metal nanobatteries to study the electrochemical mechanisms of cathode upon cycling.

Spatial Resolution Improvement for Liquid Cell TEM Imaging
At the current stage, it is still difficult to observe dynamically the internal structure evolution of cathode materials in liquid battery by in situ TEM. First, in situ chip-type liquid cell TEM technique has some limitations on the size and morphology of cathode materials. Large size and irregular morphology of materials are not conducive to TEM imaging and spectroscopic imaging of nanoscale processes. In addition, the manufacture and development of thinner windows of chip-type liquid cells are highly rewarding for observing the electrochemical mechanism of cathode materials in a real experimental environment. In addition to imaging artifacts, EELS and electron diffraction pattern are also compromised by multiple scattering events in thick liquids. In this regard, Yang et al. have recently completed a groundbreaking work in which they have successfully optimized SiN x imaging window of the chip to 35 nm and reduced the liquid layer to 150 nm. [30] It brings hope for exploring the reaction mechanism and formation mechanism of cathode electrolyte interphases of cathode materials at the atomic scale in liquid cell. In contrast, an imaging artifact deserving attention in chiptype liquid cell is electron beam radiation damage. Electron beam irradiation may result in complex precursor solution reactions such as bubble formation or PH changes. Electron beam damage can be avoided by decreasing the total dose rate, electron beam current, and exposure time by varying the accelerating voltage. However, these operations can directly generate a significant reduction in the spatial resolution of TEM to some extent. The development of high sensitivity electronic detector and low-dose image processing technology will be helpful to achieve the acquisition of better resolution images in liquid cell.
Although many crystal defects, such as phase transitions and cracks, are well-known issues in secondary ion batteries, the detailed microscopic understanding is still unclear in the literatures. Due to the rapid electrochemical reaction in practical batteries, it is difficult to capture the structural evolution of the cathode at the reaction instant. Exploring ion deintercalation/ intercalation behavior of cathode materials in real time by in situ www.advancedsciencenews.com www.small-structures.com TEM observation, revealing their electrochemical reaction mechanism, which has important implications for the electrochemical properties enhancement in real battery system. With more sophisticated hardware design and technological development, we believe that in situ TEM techniques can address more critical issues of cathode materials in secondary ion batteries and lead to the advancement of secondary ion batteries with high power density, high energy density, and long cycle lifetime.