1. Introduction
Lithium- ion batteries (LIBs) are energy-storage systems with excellent prospects. Since their successful commercialization in 1991, LIBs have been the focus of consistent efforts toward upgrading in various application scenarios due to their superior advantages of high energy density, high working voltages, and favorable cyclability [
1], [
2], [
3], [
4], [
5], [
6]. Great progress has been made in developing the cell-level energy density of LIBs—one of their major characteristics—from 200 W·h·L
−1 (80 W·h·kg
−1) to 700 W·h·L
−1 (280 W·h·kg
−1) by increasing the working voltages and specific capacity of lithium cobalt oxide (LiCoO
2, LCO) cathodes, thereby enabling an expansion of the standby time and advances in computer, communication, and consumer (3C) electronics [
7]. Aside from energy density, fast-charging capability—a key indicator of power density—has become a major far-reaching strategic goal for the evolution of LIBs, and one that will significantly improve customer convenience [
8], [
9], [
10], [
11]. For example, the US Department of Energy has set the clear goal of achieving extreme fast-charging performance (i.e., ≥ 6 C, where 1 C represents a charging time of 1 h), which would allow batteries to be recharged within 10 min [
12]. However, improvements in fast-charging performance will inevitably be accompanied by tradeoffs in other properties, such as capacity, voltage plateau, and lifetime, in most cases. Despite the great improvements that have been achieved in energy density, a stable fast-charging capability with high energy density is considered to be a major challenge of LIBs for meeting customer expectations in the rapidly expanding markets of portable electronics and electric vehicles (EVs). On a fundamental level, achieving fast-charging capability in LIBs involves modifying the diffusion rates of electrons and ions within the battery’s electrodes, electrolytes, and their interfaces [
13], [
14]. However, increasing the charging rate typically results in high-flux ions that cause increased voltage polarization, heat generation, safety risks, and reduced coulombic efficiency [
15]. Therefore, effective electrode and electrolyte materials are the cornerstone in building “double-high” LIBs with the double advantages of high energy density and high power density, and the physical/chemical reaction processes that occur inside these materials are the inherent factors determining the overall performance of LIBs.
To date, various LIB cathode materials have been commercialized (e.g., LiMn
2O
4 (LMO), LiFePO
4 (LFP), LiNi
1-x-yCo
xMnyO
2 (NCM), LiNi
1-x-yCo
xAl
yO
2 (NCA), and LCO) for various fields of application. Among these, LFP-, NCM-, NCA-, and LMO-based LIBs have penetrated the market of EVs and grid storage systems by reducing or eliminating the use of expensive cobalt [
16], [
17], [
18], [
19]. In portable application scenarios, such as laptop computers, mobile phones, and electronic watches, a greater proportion of energy is desired in a limited volume, highlighting the volumetric energy density rather than the gravimetric energy density of LIBs. LCO has remained the dominant cathode material in such applications, due to its comprehensive advantages of high volumetric energy density (∼2300 W·h·L
−1, 4.2 V vs Li/Li
+, ∼140 mA·h·g
−1), stable and high-voltage plateaus (3.7-3.9 V), favorable cycling stability (∼500 cycles), and easy synthesis [
1], [
20], [
21]. Moreover, nickel (Ni)-rich layered oxides (NLOs) and lithium (Li)-rich layered oxides (LLOs) are two potential classes of cathodes for next-generation LIBs in portable electronics due to their specific advantages of high discharge capacity (∼200 mA·h·g
−1 for NLOs and ∼250 mA·h·g
−1 for LLOs), high energy density, and cost effectiveness. Nevertheless, several challenges—including Ni oxidation, capacity fade, and irreversible capacity loss in initial cycles—remain unsolved [
22].
In the context of the accelerated development of the 5th generation mobile communication technology (5G) commercialization and the Internet of Things, greater requirements for lightweight, miniaturized, and smart 3C electronic devices with a longer standby time have increased the demand for LIBs with simultaneous high energy density and high power density (
Fig. 1). Through the joint efforts of researchers, the volumetric energy density of LCO cathode materials has been elevated from about 2300 to about 3000 W·h·L
−1 upon increased upper working voltages from 4.2 to 4.45 V, which will be further increased to approximately 3700 W·h·L
−1 when a voltage of 4.6 V is applied [
23]. Although encouraging progress has been achieved, more and greater challenges remain unsolved. The most critical challenge is the complex structural evolution, along with irreversible phase transition and anisotropic volume variation, that occurs during Li
+ deintercalation/intercalation, which results in the chemical and structural instability of LCO, with a state of charge (SOC) generally higher than 75% upon a high operating voltage [
24], [
25]. In addition, detrimental interface reactions at high voltages—including electrolyte decomposition and the formation of a cathode electrolyte interphase (CEI), deleterious surface phase transition, O
2 loss, and cobalt (Co) dissolution—augment the charge transfer impedance of the electrodes, crippling the cycling stability and rate performance [
26], [
27].
With the aim of obtaining high power density, great efforts have been devoted to investigating the relationships between the structural characteristics and fast-charging capabilities of LCO; it has been found that higher charging rates can be obtained with a decrease in the upper cutoff voltages, due to increased ion-transfer kinetics with less phase transformation at low voltages [
1], [
28], [
29], [
30], [
31]. However, a more inhomogeneous Li
+ distribution forms under fast-charging conditions, leading to anisotropic
c-axis variation and stress distribution throughout the LCO particles, eventually resulting in the structural failure of LCO [
32]. Even worse, the circumstances become more complex when the application scenario includes both a high voltage and fast charging.
Unfortunately, thus far, the incompatibility between the characteristics of fast charging and a high working voltage for LCO cathodes has neither been comprehensively summarized nor discussed in detail, which is a vacuum that we intend to fill. Herein, we provide an overview of recent progress in developing 4.6 V high-voltage fast-charging LCO cathodes, comprehensively summarizing the key challenges (e.g., structural and surficial degradation, inner inhomogeneity, and sluggish interfacial kinetics upon fast charging) and promising modification strategies while elucidating the underlying mechanisms (e.g., element doping, surface coating, nanosizing, their combination, and others) of modified LCO cathode materials (
Fig. 2) [
33], [
34], [
35], [
36], [
37]. Finally, future development directions and prospects of high-voltage fast-charging LCO cathode materials are proposed and discussed in a new light, including ➀ designing high-voltage (≥ 4.6 V) LCO cathodes; ➁ designing high-voltage (≥ 4.6 V) and fast-charging (≥ 50 C) LCO cathodes; ➂ developing advanced electrolytes to match high-voltage fast-charging LCO cathodes; ➃ obtaining a deeper mechanistic understanding of LCO cathodes via
in situ characterizations and simulation; and ➄ achieving high-voltage and fast-charging LCO in “double-high” cells.
2. Challenges and key issues in achieving high-voltage fast-charging LCO
2.1. Challenges presented by a high voltage
2.1.1. Structural degradation
In 1958, Johnston et al. [
38] first uncovered the crystallography of LCO, in which the O
2− ions take on a cubic close-packed (ccp) form, while the Li
+ and Co
3+ are alternately filled in (111) planes, with an α-NaFeO
2-type layered structure. Later, Delmas et al. [
39] denoted this structure as “O3-type LCO,” where the “O” represents the Li
+ ions located at octahedron sites in close-packed O
2− ions and the numeral “3” represents the “ABCABC…” packing type of oxygen (
Fig. 3(a) [
33]). The (111) alternating orders of Li
+ and Co
3+ ions result in a lattice distortion of the LCO crystals, with cell parameters of
a = 2.816(2) Å and
c = 14.08(1) Å, showing a hexagonal symmetry in the R3-m space group. As early as in 1980, Thackeray et al. [
40] realized reversible lithium storage in a LCO cathode material by using the electrolyte LiBF
4 in propylene carbonate; Sony Corporation then successfully commercialized the process for using LCO as a cathode in LIBs in 1991 [
33], [
41].
Thereafter, further studies focused on the electrochemical and structural behaviors of LCO [
41], [
42], [
43], [
44]. The electronic conductivity of Li
xCoO
2 was initially found to vary from semiconductive (
x = 1) to metallic (
x = 0.9-1.0) with the extraction of Li
+, which is further enhanced as the process continues, favoring the Li
+ transferal process (
Fig. 3(b)) [
42], [
45]. However, rearrangement of the oxygen layers occurs with deeper extraction of Li
+ (
x < 0.5), leaving a ccp oxygen lattice with different degrees of distortion (
Fig. 3(c) [
43]), and thus resulting in a series of mesophases [
46], [
47]. When the Li
+ ions are completely extracted from LCO, the oxygen develops into a hexagonal close-packed (hcp) CoO
2 with irreversible phase transition [
43], [
48], [
49]. Along with the violent phase transition, severe structural degradation and capacity fading inevitably occur, pushing researchers to make great efforts to investigate the underlying mechanisms.
As these investigations deepened, the phenomenon of excessive Co dissolution with a higher cutoff voltage rising to 4.2 V was found by Amatucci et al. [
43] to be another nonnegligible aspect of the degradation of LCO. Later, Ensling et al. [
44] and Chebiam et al. [
50] unraveled the fact that the lattice oxygen in Li
xCoO
2 tends to leak out, promoted by the unavoidable redox reaction between O
2− and Co
3+/4+ as a result of the overlapping of their density of state (DOS) with deep Li
+ extraction (
x < 0.5) (
Fig. 3(d)). Based on these fundamental understandings, successive studies further confirmed that the leakage of Co
3+/4+ and active oxygen ions (O
n−) was the underlying reason for the structural degradation of LCO at higher voltages [
51], [
52], [
53].
2.1.2. Interfacial degradation of LCO
As a key factor in LCO’s electrochemical performance, the surficial structural and chemical evolution upon lithiation/delithiation inevitably accounts for the failure of LCO [
54]. More specifically, deleterious surface phase transition, O
2 loss and Co dissolution, electrolyte decomposition, and generation of a CEI layer, which augment the charge transfer resistance in the electrode, have a detrimental effect on the cycling and rate performance of the LCO (
Fig. 4) [
42], [
55], [
56], [
57], [
58]. It is widely believed that the irreversible surficial phase transitions are an important determining aspect of the interfacial degradation of LCO [
59], [
60]. The surface Co
3+ of bulk LCO can even be immediately reduced to Co
2+ at the moment of contact with the electrolyte [
61], [
62]. Inconvertible surficial degradation of the layered structure to a spinel structure then occurs on deeply delithiated LCO, which intensifies and progresses to an interior bulk phase as more Li
+ ions are extracted [
63]. Further studies show that the detrimental surface phases are composed of CoO, Co
3O
4, and Li-inserted Co
3O
4 [
64], which cause increased activation energy and vacancy arrangement, exacerbating the diffusivity of Li
+ and increasing the interfacial resistance. Moreover, the resulting detrimental surficial layer exhibits limited stability, leading to the continuous transformation of the layered phase into a spinel phase—one of the fundamental reasons for the decay in performance [
65].
Oxygen loss is another key determining factor in interfacial degradation, as it induces structural deterioration from the surface to an interior bulk phase [
66]. A higher cutoff voltage (e.g., 4.4 V) causes the oxidation of O
2− to O
n- and the release of O
2 from the surface as a result of charge compensation; this can be attributed to the overlap between the Co 3d band and the O 2p band at a high voltage, which results in electrons transferring away from the O
2− [
27], [
67], [
68], [
69], [
70]. To gain insight into the charge compensation mechanisms, the electronic structure of LCO has been investigated. Some studies report that the Co
3+ and O
2− ions provide electrons simultaneously through redox reactions [
69], [
71], [
72], [
73], while others state that the Co
3+ ions are mainly responsible for the charge compensation below 4.4 V, and only the O
2− ions yield electrons for charge compensation above 4.4 V [
74]. Despite this controversy regarding the charge compensation mechanism, peroxide formation and O
2− loss have been validated to be the main factors in surface structure degradation, accounting for LCO’s superficial resistance augmentation and capacity decay [
63], [
75], [
76], [
77].
Co
3+/4+ dissolution also plays a key role in LCO degradation, occurring at a cutoff voltage higher than 4.5 V [
49], [
63]. It directly contributes to the reduction of active materials, which leads to blocked Li
+ transferal pathways due to the overgrowth of CEI film caused by catalytic decomposition of the electrolyte, the formation of corroded cavities, and Li
+-deficiency phases [
49], [
78]. The result is inevitable performance decay.
The properties of the CEI are another key factor in LCO degradation; they are greatly influenced by various electrolytes with different windows of electrochemical stability [
79], [
80]. The decomposition process occurs when the applied voltage exceeds the oxidizing potential of the electrolyte and the byproducts subsequently compose a CEI film. This electrolyte decomposition process happens continually, along with the release of the active radicals O
n- and Co
3+/4+, causing the CEI film to overgrow [
81], [
82], [
83]. Furthermore, overgrowth of the CEI layers hinders the transmissibility and invertibility of Li
+ during the subsequent cyclic insertion/desertion processes, impeding the kinetics and reducing the LCO’s capability for Li
+-storage [
84], [
85], [
86], [
87].
In short, deleterious phase transition, the oxidation of oxygen, the dissolution of Co3+/4+, and CEI overgrowth are inescapable surface issues that deteriorate the stability and safety of LCO under a working voltage higher than 4.4 V. Thus, persistent endeavors to develop all-encompassing strategies are urgently needed in order to overcome the challenges presented by high-voltage LCO.
2.2. Failure mechanisms during fast charging
To support the widespread deployment of multiplex applications, the ideal LIBs should possess a range of characteristics, including high energy density, high power density, and a long lifetime, enabling a long service time with a single quick charge [
12], [
88], [
89]. The fast-charging capability of LIBs fundamentally depends on the diffusion kinetics of the Li
+ and electrons within the composite electrodes, electrolytes, and their interfaces. The most deficient ionic and electronic transportation in the electrodes is the determining step in a high-rate battery, which will further lead to heterogeneity at the macroscale level (e.g., current, temperature, and ionic diffusion kinetics) [
90], [
91].
2.2.1. Inhomogeneity within electrodes under fast charging
The high-rate performance of LCO is of vital importance in satisfying customers’ expectations for fast-charging LIBs. LCO can reach a higher overpotential at higher charge/discharge rates to provide an extra driving force for Li
+ transferal; however, a large and uneven distribution of Li
+ occurs within the LCO particles under these conditions, resulting in structural defects and irreversible phase transformation [
92]. It is considered that Li
+ may energetically favor occupying the tetrahedral site of the oxygen lattice in Li
+-deficiency domains, leading to lattice O
2− charge compensation and the formation of an intermediate phase, which is detrimental to the cycling stability. For example, chemical mapping of the oxidation states of the Co ions in a single selected particle of LCO was performed at various rates after fully charging/discharging (
Figs. 5(a) and
(b)), and a heterogeneous SOC was clearly observed throughout the particle (especially at high rates) [
34]. More specifically, partial domains (highlighted in red) retain a charged state after being fully discharged, and the size of these domains increases at higher rates, triggering greater inhomogeneity at the particle level. Furthermore, other studies show that the phase-transition process of the cathode materials may differ at various rates, for example, a new intermediate phase may emerge at higher rates (
Fig. 5(c)) [
93]. This chemical inhomogeneity can also result in morphological cracks along with an inactive domain, as demonstrated by Diercks et al. [
94] and Gong et al. [
95], resulting in degraded electrochemical performance. Despite the valuable knowledge provided by these works, few studies have reported on the microscopic mechanism of the lattice oxygen redox reaction under high-voltage and fast-charging conditions.
Heat can be generated and accumulated at some locations in an LIB more easily than at others, such as at the core and near-tab regions. In addition, the thermogenesis rates may differ in diverse regions due to heterogeneous current densities, generating inhomogeneities that are unfavorable for the electrochemical reactions—a problem that is worse in large-format cells (
Fig. 5(d)) [
96], [
97]. For example, excessive current densities around the tabs may cause exorbitant regional charging or discharging rates, possibly promoting regional failure [
98].
Heterogeneities in temperature can also induce varying regional speeds that result in detrimental side reactions and structural failure rates, and plenty of degradation mechanisms in LIBs are dependent on temperature. For example, the rise in temperature during cycling causes the solid electrolyte interphase (SEI) layer to grow faster on the anode and to be more porous and unstable; it also results in binder decomposition, phase transformation, metal dissolution, and CEI overgrowth on the cathode electrodes [
63], [
78], [
99], [
100], [
101], [
102]. Moreover, elevated temperature can induce decomposition of the electrolyte and the generation of gaseous species, causing serious safety issues [
70]. Therefore, it is essential to gain deep insight into the inherent heterogeneity mechanisms of LCO and to develop novel and feasible strategies for improving the electrochemical performance of high-voltage fast-charging LCO.
2.2.2. Degradation due to sluggish interfacial kinetics
There are multifold specific reasons for the accelerated degradation of both the capacity and the power density of LIBs. In addition to the above mentioned inhomogeneity-based degradation mechanisms originating from insufficient Li+/e− transportation inside the electrodes, sluggish interfacial kinetics are fundamental reasons for the degradation that occurs from fast charging, which may be due to Li plating on the anode side, over-decomposition of the electrolyte, and the overgrowth of interfacial films on the electrodes.
Although the high ionic conductivity of organic liquid electrolytes (i.e., about 10
−2 S·cm
−1 at room temperature) can adequately satisfy the requirements for high-rate performance [
103], the interfacial kinetics are influenced by the CEI layer and by the SEI film generated on the electrodes after the first several cycles, which depend on the constitution of the electrolytes [
104], [
105]. In a common working scenario, the interfacial Li
+ transportation kinetics in an LIB are sufficient for the Li/Li
+ redox reaction inside the electrodes, exhibiting a redox reaction-controlled character. However, the diffusion of Li
+ on the interfaces is limited, and the overpotential thus increases under fast-charging conditions. Furthermore, the cathodes encounter a relatively higher working voltage during fast charging, making the degradation process of LCO likely to occur. On the other side, when the potential of the anodes falls below the Li/Li
+ redox potential, the deposition of metallic lithium and the formation of dead lithium are triggered, accelerating the failure of the LCO [
106]. Moreover, the overpotential caused by the passive surface layers on the anodes and cathodes can lead to consistent decomposition of the electrolyte and the formation of interfacial layers. This process consumes lithium irreversibly and increases the interfacial impedance, actuating a vicious cycle.
2.3. Challenges presented by high-voltage and fast-charging
When LCO cathodes are applied under the demanding conditions of a high voltage and fast charging, the degradation processes both inside the particle and on the surface are tremendously accelerated. Inside the particle, the violent phase transition that occurs as the high voltage multiplies the inhomogeneous reactions caused by fast charging will greatly intensify the oxygen/cobalt redox reaction and result in serious strain accumulation, leading to rapid crack formation and capacity decay. On the surface, the side reactions that occur at high voltage, such as cobalt dissolution and oxygen release, will aggravate the degradation when a high specific current is applied, causing the overgrowth of CEI films and an increase in surface impedance. Even worse, the growing impedance will increase the overpotential on the LCO surface, further facilitating the side reactions and structural degradation on the surface.
Key modification strategies for improving the high-voltage performance of the electrodes include elemental doping and surface coating, which aim to address the structural and interfacial degradation that occurs during a high-voltage cycling process [
1], [
7], [
107], [
108]. As a specific example, constructing a protective interfacial layer on the LCO electrode by means of surface coating is conducive to suppressing the surface side reactions—such as excessive electrolyte decomposition, degradation, and pulverization of the active material—and improving the interface compatibility [
60], [
109]. This process is generally favorable for improving the stability of advanced high-voltage LIBs [
110]. In contrast, strategies to improve the fast-charging capability should focus on the design of the macro/nanoscale structure, as high porosity, low tortuosity, more resilient electrodes, and accelerated Li
+ transportation and mechanical effects are beneficial for reducing the lithium concentration gradients induced by fast charging while maintaining overall homogeneous reactions [
36], [
108]. However, a more exposed surface area inevitably results in more serious surface side reactions, which will worsened at a high working voltage, posing an intrinsic dilemma.
Considering the issues described above, the key to fabricating advanced LCO electrodes with high-voltage and fast-charging performance lies in establishing fast pathways for the high-flux mass transportation of Li
+ and electrons and in promoting the structural stability of the materials. Moreover, tradeoffs must be considered while designing electrodes; for example, surface coating enhances the stability but impedes the transportation of Li
+ at the interfaces, and nanosized electrode particles facilitate fast ion-diffusion kinetics but reduce the volumetric energy density of the materials. Furthermore, the formation of CEI/SEI layers with sufficient ionic conductivity and stable properties is essential for the stability of the electrolytes [
111], [
112].
3. Promising strategies
3.1. Element doping
Element doping is an effective strategy to tune the inherent physics of LCO by adjusting the crystallize structure in terms of the dimensions of the atoms, such as the bandgap, arrangement of cations, concentrations of defects, and reallocation of charges [
113]. This is the earliest reported method for the enhancement of the high-voltage Li
+-storage capability of LCO; it increases the Li
+ extraction potential and the transportation kinetics of the electrons and ions, speeds up the redox reaction, and improves structural stability [
7]. Theoretical studies have demonstrated that transition metal dopants contribute to increasing the capacity, while non-transition metal ion doping is conducive to increasing the delithiation potential, both of which are favorable for achieving high energy density of LCO [
1], [
114]. Therefore, the element-doping strategy has received wide spread attention as a means of improving the electrochemical performance of LCO.
3.1.1. Lithium-ion sites
Doping heteroatoms at lithium-ion sites can not only enhance the transportation kinetics of the electrons and ions but also favor the structural reversibility of LCO, improving the electrochemical performance [
115], [
116], [
117]. For example, a handful of magnesium (Mg) ion dopants that are substituted for Li ions can act as “pillars” to prevent slab displacement upon a high SOC, thereby depressing irreversible phase transition and eliminating CEI overgrowth in the superficial region [
118].
Based on this design, Huang et al. [
115] prepared an LCO cathode material with Mg pillared at the Octa-3a sites in the Li slab (denoted as LMCO). The structure was well characterized by the distinctive signals observed on the Li
+ sites through scanning transmission electron microscopy (STEM) measurement with high-angle annular dark-field mode (HAADF) and the lower formation energy for LMCO based on theoretical calculation, as illustrated in
Fig. 6(a). Furthermore, it was confirmed from the
in situ XRD pattern that the position change in the (003) peak for LMCO (0.96°) was smaller than that for LCO (1.26°), suggesting that LMCO underwent a greatly eased detrimental phase transformation from O3 to H1-H3 and M2 at the charging state of 4.6 V (
Fig. 6(b)). Moreover, it was calculated that the band gap between the conduction band and the Fermi energy level was reduced from approximately 2.62 to about 0.49 eV, and the barrier energy was decreased fourfold from 2.73 eV in LCO to 0.63 eV in LMCO after Mg ion doping, demonstrating greatly improved electronic/ionic diffusion kinetics (
Fig. 6(c) [
115]). It was further revealed that an Li-Mg mixed structure was generated
in situ on the surface of the LMCO, which was favorable for suppressing the overgrowth of the CEI and phase transitions in the superficial regions. Thus, LMCO exhibited a high capacity (204 mA·h·g
−1, 0.2 C), improved capacity retention (84% after 100 cycles, 1 C) and fast-charging capability (138 mA·h·g
−1, 4 C) between 3-4.6 V.
Apart from the well-studied Mg ions, Ni is another representative element whose ions have a tendency to be positioned at lithium-ion sites in LCO; thus, Ni ions can be incorporated into Octa-3b sites in the lithium slab of LCO [
35]. After incorporation, the structural collapse that originates from the electrostatic repulsion between oxygen ions upon deep delithiation can be well mitigated through the “screen effect,” resulting in a greatly stabilized cycling performance (
Figs. 6(d) and
(e) [
35]). For example, Kim et al. [
119] introduced the entropy change (Δ
S) at different SOCs to explain the screening effect of Ni dopants in LCO. The Δ
S showed a simple monotonic decline as the vacancy concentrations at Li sites (i.e., lithium extraction) gradually increased, until the generation of a monoclinic mesophase. In contrast, the amplitude of Δ
S in the monoclinic region decreased greatly after Ni doping, which is indicative of stability enhancement of the crystal structure, leading to improved cyclability. In addition to Ni and Mg ions, copper (Cu), niobium (Nb), tungsten (W), and sodium (Na) ions can also serve as “pillars” upon substitution for Li ions in LCO [
120], [
121].
In addition to metal ions, polyanion dopants can decrease the energy of the top O 2p bands of LCO [
6], which is anticipated to depress the leakage of O
2− ions at higher voltages. The establishment of robust covalent bonds in the form of X-O-Co bonds through the connection between (XO
4)
n- (X = P, S, B, and Si) polyanions and CoO
x polyhedrons is deemed to be a favorable strategy to stabilize the crystal structure and O
2− ion skeleton of LCO [
122]. Moreover, the (XO
4)
n- polyanions can take over the tetrahedral sites, thereby suppressing the structural transformation from Li
xCo
2O
4 to spinel Co
3O
4 or even a rock salt structure, thanks to the favorable stability of (XO
4)
n- and the robust repulsive forces between the high-valence Co
3+ and X ions with a high valent state [
123].
3.1.2. Cobalt-ion sites
The manipulation of cobalt-ion sites through partial replacement by atoms (e.g., zirconium (Zr), aluminium (Al), and vanadium (V)) is considered to be a feasible strategy that has been widely demonstrated to enhance the electrochemical performance of LCO, especially under high-voltage or high-rate conditions [
124], [
125], [
126], [
127]. For example, by tuning the oxygen magnetic moment (OMM) via oxygen vacancies (O
v) and V dopants in cobalt-ion sites, Kong et al. [
128] developed an anionic redox-reversible LCO cathode material at 4.6 V (
Fig. 7(a)). The O
v in the LCO led to the generation of OMM; furthermore, the V dopants and the O
v stabilized the OMM during the cyclic process, enormously ameliorating the invertibility of the redox reaction for lattice O
2− by mitigating detrimental O
n- leakage (
Figs. 7(b) and
(c)). In addition, the O
v optimized the oxygen redox activity by lowering the band center of O 2p and decreasing the overlap of the electron clouds (
Figs. 7(d)- (f)). At the same time, the V dopants increased the energy barrier of Co migration, greatly increasing the reversible phase transition. Thus, the cyclic stability (93% capacity retention after 200 cycles at 5 C) and rate capability (80 mA·h·g
−1, 5 C) of LCO were greatly enhanced.
Al
3+ is another typical cobalt-ion site dopant. Both Al
3+ (0.535 Å) and Co
3+ (0.545 Å) have a similar ionic radius and the same valence, which ensures the substitution of Al for Co on a large scale in the LCO framework. The higher Al-O bonding energy and smaller crystal distortions in Al
3+-doped LCO improves the structural stability and reduces the lattice stress during the intercalation and extraction of Li
+ [
129], [
130], [
131], [
132]. For example, Er-Rami et al. [
129] found that an Al ion doping of 4 at% in LCO was helpful for the formation of a lithiated spinel-type phase on the surface and a stable CEI layer; it also delayed and increased the reversibility of the phase transformation process of H1-3 ↔ O1 (
Figs. 7(g)-(i)).
3.1.3. Oxygen-ion sites
The hybridization of O anions (O
2− → O
n−,
n < 2) and Co cation-redox (hybridization of anion and cation-redox) occur when LCO is charged to a voltage higher than 4.6 V [
73], [
107], [
133], [
134], [
135]. Undesirably, the decreased ion radius and electrostatic interaction result in the increased mobility and loss of O
n− and Co
4+ [
109], [
136]. A possible strategy for suppressing the escape of the oxygen ions and improving the stability of the Co cations in LCO is to dope nonmetal elements with weaker electronegativity than the O element into the oxygen-ion sites in LCO. For example, Zhu et al. [
60] prepared modified LCO using operando selenium (Se) treatment methods (Se-LCO;
Figs. 8(a) and
(b)), which exhibited an impressive “anti-aging” effect when cycled at 4.62 V. It was observed that the spinel domains were randomly distributed on the superficial layer of the commercial LCO (C-LCO) particles (
Fig. 8(c)), while stabilized layer lattices were formed on the surface of the Se-LCO particles after 120 cycles (
Fig. 8(d)), which favored rapid Li
+ transportation and low interface resistance during cycling. Furthermore, the oxygen’s K-edge curves from soft X-ray absorption spectroscopy revealed that the surface coating layer of Se was able to substitute for some mobile O
n− on the superficial layer of the delithiated Se-LCO, replenishing electrons to the oxidized O
n− (Se-Co → O) and recovering it to O
2−. This increased the stability of the lattice oxygen during cycling, mitigating oxygen migration by eliminating one prerequisite of the O
n--O
v exchange (
Fig. 8(e)). As a result, the average Li
+ diffusivity (
DLi+) in the 120th discharge process for Se-LCO was calculated to be approximately 10
−9 S·cm
−1—greater than that of C-LCO by three orders of magnitude (
Fig. 8(f)).
Elements with a stronger electronegativity than O have also been used to modify LCO. For example, surface doping with fluorine (F) ions can reduce the superficial side reactions of LCO with the electrolyte, as F takes part in the formation of much more stabilized Co-F bonds, which can effectively mitigate the oxidizing process of lattice oxygen at high cutoff voltages [
137]. Moreover, F doping can produce superficial transition metal fluoride (TMF
x) species as a positive component of the CEI to inhibit the corrosion of hydrofluoric acid (HF), thereby preventing HF corrosion on the surficial layer of LCO and promoting the stability of the CEI layer [
138], [
139], [
140], [
141], [
142].
3.1.4. Multiple sites
Multiple- element doping of LCO can result in a higher cutoff voltage of 4.5 V or even 4.6 V for achieving a higher energy density [
27], [
143], [
144]. For example, a combination of lanthanum (La) and Al dopants in LCO were reported with a high cutoff voltage of up to 4.5 V and even 4.6 V, where the La served as a “pillar” to increase the
c-axis and the Al functioned as a “center of positive charge” to facilitate Li
+ transportation, thereby stabilizing the structure and suppressing phase transformation. Liu et al. [
143] initially designed a La/Al dual-doped LCO (denoted as LA-LCO) by calcinating lithium carbonate with the La/Al concurrently doped precursors of La/Al-Co
3O
4 from La/Al-doped CoCO
3. The researchers found that a series of distinct phase-transition peaks at 4.1, 4.2, and 4.46 V (vs Li/Li
+) for pristine LCO (P-LCO) were substantially suppressed in LA-LCO, with the volumetric variation of the unit cell significantly decreasing from 3.63% to 2.97%, which was indicative of the suppressed structural distortion and mechanical decay of LA-LCO at 4.46 V. Further studies revealed that the LA-LCO also exhibited delayed emergence of the H1-3 phase and alleviated volume variation at 4.55 V, resulting in lower lattice strain and better reversibility of the O3-H1-3 phase transformation [
145].
Aside from the high voltage of 4.55 V that was achieved using La/Al coupling in a modified LCO, Zhang et al. [
27] obtained a stable cyclability of LCO at a higher cutoff voltage of 4.6 V (vs Li/Li
+) using a trace Ti/Mg/Al co-doping strategy (denoted as TMA-LCO). The doping of Mg and Al into the LCO framework effectively inhibited the irreversible phase transformation at a voltage higher than 4.5 V, while the dopant titanium (Ti) separated from bulk region at the grain boundary and promoted the formation of a thinner yet more stable CEI layer [
146]. Moreover, the mapping results from resonant inelastic X-ray scattering (RIXS) for the oxygen’s K edge demonstrated less oxidization of the lattice oxygen at the superficial layer of the TMA-LCO grains than bare LCO in a charged state (4.6 V), which suggested a significantly improved stability of the superficial lattice oxygen. This result agreed well with the density functional theory (DFT) simulation results, which showed that the dopant Ti decreased the electron deficiency of the O atoms upon the deep extraction of Li
+. Although remarkable cyclability and rate performance were achieved as well, different results reported by Wang et al. [
147] demonstrated that the Al and Ti ions were bulk co-doped, while Mg ions are doped at the particle surface in a gradient manner. This controversial result might be caused by different synthesis processes and heteroatom content of the LCO cathodes. In addition, it has been suggested that the size of LCO can be easily tuned with a trace amount of titanium oxide [
148], further improving the energy density and rate capability at high voltages.
More importantly, combining the doping of lithium-ion sites with the doping of oxygen-ion sites can significantly increase the working voltage to as high as 4.6 V for LCO by mitigating oxygen escape and the formation of a Li
+-insulating Co
3O
4 phase on the particle surface. For example, Kong et al. [
149] developed a zero-strain LCO by means of MgF
2 doping (denoted as LM
0.01COF
0.02), where the Mg at the lithium-ion sites functioned as a “pillar” to facilitate Li
+ diffusion and the F at the oxygen-ion sites helped stabilize the oxygen ions due to the greater electronegativity of the F ions in comparison with the lattice O ions (
Fig. 9(a)). Moreover, it was found that the dopant Mg and F increased the energy gap between O 2p and Co 3d by 0.1479 eV, reducing the redox activity of the O ions and improving the bond ionicity of Co-O (
Figs. 9(b) and
(c)). Therefore, the production of a lithium insulator phase (Co
3O
4) was suppressed (
Figs. 9(d) and
(e)), resulting in greatly improved cyclability at 4.6 V (92% capacity retained after 100 cycles at 1 C, and 86% capacity retained after 1000 cycles at 5 C) and excellent rate capability (140 mA·h·g
−1 at 5 C).
Combining the doping of oxygen-ion sites with the doping of cobalt-ion sites can also contribute to the electrochemical capability of LCO under a high voltage. As exemplified in
Figs. 9(f) and
(g), Huang et al. [
150] designed an oriented Al and F superficial gradient doping strategy to modify LCO (denoted as DG-LCO) for the purpose of reducing detrimental surficial side reactions between highly active Co
4+/O
n- and an organic electrolyte upon cycling to voltages higher than 4.5 V. The surface with a doping gradient exhibited a spinel structure with a coherent crystal lattice, which favored the diffusion of Li
+ at the interface and alleviated the structural degradation upon the deep extraction of Li
+. All these advantages led to a significantly enhanced cycling (86% capacity retained after 200 cycles at 1 C) and rate performance (108 mA·h·g
−1, 10 C) for DG-LCO at a high working voltage (4.6 V,
Fig. 9(h)).
Aside from these representative examples of multi-site-doped LCO [
151], [
152], [
153], [
154], other combinations of multi-site doping have demonstrated positive effects on the electrochemical performance (
Table 1). For example, Chen et al. [
118] prepared LCO with trace Nb/W co-intercalated into the Li slab and Al substituted into the cobalt-ion sites (denoted as ANW-LCO), resulting in increased interlayer spacing and decreased electronic coupling of Li
+, which assisted in increasing the effective utilization and diffusive kinetics of Li
+. Consequently, ANW-LCO exhibited a rate performance of 142 mA·h·g
−1 at 15 C and maintained a capacity of 85 mA·h·g
−1 after 1000 cycles at 10 C. This work further illustrates the importance of the synergistic doping of multiple elements in constructing high-voltage fast-charge LCO.
3.2. Surface coating
Surface-coating strategies have gained increasing attention for dealing with the severity of LCO surface issues [
155], [
156], [
157], [
158], [
159], as the surface of LCO is more susceptible than the inner bulk phase at high voltages. More specifically, lithium ions close to the surface are more depleted than those in the inner bulk phase during the charging process, due to the existence of a concentration gradient—an issue that is even worse when a high working voltage (i.e., deep extraction of Li
+) or high charge/discharge rate is applied [
160], [
161], [
162]. Co
3+ redox (Co
3+ → Co
4+) and oxygen redox will participate in charge compensation in sequence, because of the overlapping of the Co 3d and O 2p bands, resulting in oxygen leakage and detrimental phase transition (CoO
2 → Co
3O
4) on the outer shell [
59], [
60].
Furthermore, highly oxidized O and Co species will directly decompose the electrolytes on the interface, causing the overgrowth of the CEI film at the cost of Li
+, lattice Co, and O ions [
60]. Moreover, both O and Co leakage severely destroy the structural completeness, triggering the development of O
v from the outer shell into the interior, which is accompanied by structural collapse [
109]. The newly exposed particle surface will aggravate detrimental side reactions, such as superficial structural deterioration and decomposition of the electrolytes, considerably reducing the
DLi+ [
163]; this aggravates crystal pulverization, creating a vicious cycle. It is generally accepted that—except for related issues caused by residual lithium compounds on the electrode surface—other factors such as the oxidization and dilution of cobalt ions stem from the unstable/irreversible evolution of the lattice oxygen. Therefore, in principle, the key to achieving stability of the interface structure lies in stabilizing the lattice oxygen framework near the surface, and then realizing the optimization or improvement of other issues.
3.2.1. Dielectric materials
Dielectric materials, which function as electrical insulators, are relatively stable when placed in an electric field, even in the presence of organic electrolytes, providing an effective physical obstacle that protects the surface-modified LCO. It has been reported that a metal oxide—as a typical kind of dielectric material—that is surface-coated on an LCO cathode will further react with the byproducts of electrolyte decomposition. For example, Wu et al. [
164] synthesized a thin (1 nm) Al
2O
3 layer coating LCO by means of atomic layer deposition methods. When cycled in a LiPF
6-based electrolyte, the Al
2O
3 coating layer reconstructed
in situ into a Li
3AlF
6 layer that tightly bonded to the LCO, endowing the coating layer with fast
DLi+, decreased surface degradation, and improved electrochemical capability (
Fig. 10(a)). Thus, a capacity of 89%/88% was retained after 1000/200 cycles at 4.5/4.6 V, respectively, and an enhanced rate performance (165 mA·h·g
−1, 3 C) at 4.6 V was achieved (
Fig. 10(b)). It was further demonstrated that the Al
2O
3-coated LCO could actually reach a higher cutoff voltage. Zhou et al. [
165] proposed a facile and scalable wet-chemical Al
2O
3-coated LCO that reached high voltages of 4.5-4.7 V. The Al
2O
3 coating layer significantly depressed the production of the irreversible CEI component (i.e., LiF/Li
2CO
3, an organic species), thereby providing a kinetically favorable interface for the LCO cathode, which improved the Li
+ diffusion at the solid-liquid interface.
Partial coverage of dielectric materials on the outer shell of LCO can significantly facilitate Li
+ transportation at the multiphase interface formed by the dielectric materials, cathodes, and electrolytes. For example, Yasuhara et al. [
166] developed a BaTiO
3 nanodot (< 3 nm thick and 35 nm across)-covered LCO material with a coverage rate of < 5%. They demonstrated that the interfacial resistance between the electrolyte and the electrode of BaTiO
3-LCO was greatly decreased in comparison with that of LCO (
Figs. 10(c) and
(d)). Thus, the robust Li
+ transportation favored by the fast delivery channel in BaTiO
3-LCO contributed greatly to the enhanced rate (60 mA·h·g
−1, 100 C) and cycling performance (90% capacity retained after 800 cycles at 5 C;
Fig. 10(e)). In addition, Cheng et al. [
167] prepared a multilayer-coating-modified LCO consisting of a zinc (Zn)-rich surficial coating, a rock-salt buffering interphase, and an Al gradient doped superficial layer (
Fig. 10(f)). Interestingly, the multilayer structure quasi-epitaxially grew along the LCO crystal and greatly improved the lattice matching and uniformity of the coating layer, resulting in a smooth, dense coating layer and the original layered phase after cycling (
Fig. 10(g)). This synergistic effect of the multicoating layer ensured stable cycling stability up to 500 times with a capacity retention of 66% and a fast-charging capability of 10 C between 3-4.6 V.
3.2.2. Ionic conductive materials
It is well-established that traditional dielectric materials, such as Al
2O
3 [
165] and MgO [
168], can function as a coating layer for LCO to mitigate detrimental side reactions. However, dielectric materials are usually characterized as restricting the mechanical and chemical stability upon cycling and have a limited ability to protect superficial lattice oxygen under a high working voltage [
169], [
170], [
171], [
172]. In addition, the rate performance of LCO coated with dielectric materials not as obviously improved as the cycling capability. To address this issue, it is necessary to simultaneously promote
DLi+ in the coating layer, such as by using lithium-containing fluorides, which can enhance both the cycling and the rate performance of LCO under an elevated working voltage [
52], [
173]. For example, Qian et al. [
174] reported a Li/Al/F ternary doped LCO (denoted as LAF-LCO) produced by means of a facile hydrothermal treatment. The coating layer, which consisted of rich Al and F elements, significantly optimized the stability of the interface and the structure, despite the etching effect of HF [
175], [
176]. Moreover, the excess Li
+ ions granted the coating layer an enriched diffusion pathway [
177]. The elements of Al and F also diffused into the outer shell of the LCO, producing a Li-Al-Co-O-F solid solution. Benefiting from this unique design, the highly loaded LAF-LCO electrodes exhibited a retained capacity of 82% after 200 cycles and an improved rate capability. Other lithium-containing coating layers, such as LiAlO
2 [
176] and Li
2B
4O
7 [
177], have also been reported to boost the diffusivity of Li
+ with limited energy loss of the modified LCO along with increased impedance after surface coating.
Another typical material that can be used to coat LCO is a spinel-phase lithium ion conductor, due to its three-dimensional (3D) ion-diffusion pathways and its high structural stability, as its structural oxygen requires more energy to oxidize than that of LCO [
178]. It has been reported that a spinel-phase coating layer retained high stability at a working voltage as high as 4.75 V, owing to its specific effect on the delay of the O3 → H1-3 → O1 phase transformation and its depression of cobalt dissolution [
179]. As a typical example, Tan et al
. [
180] reported coating LCO with spinel Li
xCo
2O
4 and gradient doping it with SO
42- polyanions through an
in situ gas-solid interface reaction in a mixture of metal oxides and sulfur (
Figs. 11(a) and
(b)). The coated spinel oxide and the intense metal—O—S bond was able to eliminate the decomposition of the electrolytes and stabilize the oxygen at high voltages. Thanks to these advantages, the modified LCO exhibited an excellent rate capability (139 mA·h·g
−1 at 20 C) and cycling performance (97% capacity retained upon 100 cycles at 1 C) with a cutoff voltage of 4.6 V (
Fig. 11(c)).
Phosphate-based ionic conductors are another representative material that can be used as a coating layer to modify LCO; it has been reported that such a coating not only prevents the leakage of lattice oxygen from LCO but also facilitates an ideal pathway for fast Li
+ transfer at high voltages [
181], [
182]. For example, Yang et al. [
181] prepared a lattice epitaxially matched LiCoPO
4 coating layer that was completely grown on the surface of LCO via an
in situ chemical reaction (
Fig. 11(d)). The as-formed intense covalent P-O tetrahedron on the superficial layer increased the stability of the surficial lattice oxygen and restricted oxygen leaking and irreversible phase collapse, thereby ensuring a stabilized CEI layer, which resulted in a working voltage as high as 4.7 V. The modified LCO showed a capacity retention of 87% after 300 cycles at 4.6 V, as well as excellent rate performance (178 mA·h·g
−1 at 10 C at 4.6 V;
Fig. 11(e)). More importantly, other phosphate-based ionic conductor coating layers, such as LiAlGePO
4 [
183], LiZr
2(PO
4)
3 [
184], Li
1.5Al
0.5Ti
1.5(PO
4)
3 [
185], and Li
3Al(PO
4)
2 [
182], also delivered improved Li
+-storage capability with a stabilized surface, inhibited the evolution of oxygen, enhanced the Li
+ diffusion, and permitted high working voltages (> 4.5 V).
3.2.3. Electronic conductive materials
Electronic conductive materials are an important class of coating materials for LCO, as they can not only provide physical obstruction between the electrolyte and LCO for enhanced stability but also offer fast electronic pathways for improved performance. For example, Sharifi-Asl et al. [
186] prepared graphene-coated LCO via the self-assembly of graphene oxide with a negative charge on LCO particles with a positive charge in order to suppress lattice oxygen leaking under nonstoichiometric conditions (
Figs. 12(a) and
(b)). Notably, the surface film resistance was significantly less than that of bare LCO after 20 cycles, resulting in increased conductivity of the CEI film and better cycling stability (200 cycles with 86% capacity retained at 0.1 C between 3-4.2 V;
Figs. 12(c) and
(d)). This result was attributed to the improved structural/thermal stability provided by the graphene coating, which significantly hindered the release of oxygen at a high voltage. The greatly improved electrochemical performance was attributed to the formation of intense C-O
cathode bonds on the outer shell of the graphene/LCO, which was systematically explained by means of DFT and
ab initio molecular dynamics calculations. It should be noted that the intrinsic conductivity of ions and electrons through graphene/LCO could not be sufficiently improved, despite the greatly boosted surficial electronic conductivity and structural stability.
3.2.4. Multiple materials
Despite the great progress that has been achieved in coating LCO with dielectric materials and ionic/electronic conductive materials, it is still difficult to simultaneously manipulate LCO’s poor electronic and ionic conductivity, as well as its unstable surface/bulk structure, using a single material [
187], [
188], [
189], [
190]. To this end, Cheng et al. [
187] designed a hybrid layer (denoted as HC-LCO) evenly coated on the surface of LCO with Li
1.5Al
0.5Ge
1.5P
3O
12 and Al-doped ZnO as the Li
+ and electron conductor, respectively; this greatly accelerated the electron/ion transportation and depressed the side reactions (e.g., particle cracking, Co dissolution, and electrolyte decomposition) on the electrode/electrolyte interface (
Figs. 13(a) and
(b)). The as-prepared HC-LCO achieved an enhanced cycling capability (77.1% capacity retained upon 350 cycles at 1 C) and remarkable rate capability, exhibiting 139 mA·h·g
−1 at 10 C between 3-4.6 V (
Fig. 13(c)).
Mao et al. [
191] reported the use of a combination of ionic conductive Li
2CoTi
3O
8 and electrochemically stable LiF as a hybrid coating layer on LCO. The researchers found that the trace Ti and F atoms transferred into the LCO surface to form an oriented spinel-phase transition layer (
Figs. 13(d) and
(e)), avoiding mismatching on the interface and facilitating Li
+ diffusion (
Fig. 13(f)). These outside-in oriented nanostructures on modified LCO (denoted as OIN-LCO) served as a physical barrier inhibiting damage to the cathode and electrolyte (e.g., HF etching, Co ion dissolution, and CEI overgrowth), stabilizing the lattice oxygen, and preventing irreversible phase transition. With this protection, the modified LCO retained 81% capacity upon 200 cycles and exhibited an enhanced rate performance of 130 mA·h·g
−1 at 3 C between 3-4.6 V (
Figs. 13(g) and
(h)). Similarly, other combinations, such as Li
2SrSiO
4 and Al
2O
3, have been shown to greatly improve the electrochemical performance in
Table 2 [
192].
3.3. Nanosizing
Downsizing the electrode material to the nanoscale is one of the most powerful methods for addressing the sluggish kinetics that result from inferior
DLi+ in the particles; the increase in the contact area with the electrolytes and the shortening of the Li
+ transport distance that result from the downsized particles should be highly conducive to improving the rate performance. For example, the
DLi+ in LCO has been calculated to be 10-11.6 cm
2·s
−1 [
193]. A discharge duration (the insertion of Li
+) of 1 s requires a Li
+ diffusion pathway of less than 15 nm—such as a particle smaller than 30 nm [
36]. Moreover, the internal stress, local current density distribution, and local ion concentration can be optimized by nanosizing the LCO particles. Based on this principle, various nanosized electrode materials with high-rate performance have been designed.
Inspired by the unique physical and chemical characteristics of novel two-dimensional (2D) materials, Tai et al. [
108] designed a serious of nanosheets (e.g., LCO, LiMn
2O
4, and LiFePO
4) via a mechanical exfoliation strategy. They demonstrated that the lattice parameter of 2D LCO nanosheets showed a negligible variation of < 1% upon the insertion and desertion of Li
+, which was a representative zero-strain cycling characteristic of nanosized materials (
Fig. 14(a)). As expected, the as-prepared nanosheets delivered a remarkable rate (70 mA·h·g
−1, 10 C) and cycling performance with greatly reduced volume variations (
Fig. 14(b)).
Higher charging rates can be achieved more easily by further downsizing the crystal size of LCO. For example, Okubo et al. [
36] reported a series of size-controlled nanocrystalline LCO via a hydrothermal reaction, among which the nanocrystalline LCO with a particle size of 17 nm delivered a high rate capability with 65% capacity retention at 100 C between 3-4.2 V (
Figs. 14(c) and
(d)). Increased capacitor behavior was also observed on the intercalation of the Li ions, illustrating the importance of the nanosizing effect of LCO for high-power applications (
Figs. 14(e) and
(f)).
Specific exposed facets are also of vital importance for enhancing the fast-charging capability of LCO. To explore the effect of specific exposed crystal facets, Zhou et al. [
37] reported hierarchical fusiform LCO microrods with coordinated crystal orientation, which were characterized as having {010} facets dominating the exposed surface (
Fig. 14(g)). The total explosion of the {010} facets ensured a 2D channel for Li
+ diffusion, leading to a remarkable high-rate ability (111 mA·h·g
−1 even at 50 C) and high-rate cycling performance (113, 106, and 80 mA·h·g
−1 at 10, 20, and 50 C after 200 cycles, respectively) (
Figs. 14(h) and
(i)). Other nanosized LCO cathode materials with various nanostructure characteristics, such as LCO with a “desert rose” morphology [
194], LCO nanoplates with exposed {001} planes [
195], LCO nanowires with exposed {010} planes [
196], LCO nanorod arrays [
197], and LCO polyhedrons with twin boundaries [
198], have also been reported to shorten the Li
+ diffusion distance and construct fast Li
+-transferal pathways for high-voltage fast-charging LCO cathodes.
In fact, numerous works have discussed the effect of particle size of various electrode materials on high rates and energy densities. Smaller particles display a greater capability to accommodate volume variations and crack formations due to the decreased lithium gradients and energy release rates [
199], [
200]. However, the strategy of downsizing the particle sizes is a so-called double-edged sword. While this strategy increases the specific surface areas, it decreases the packing density, resulting in lower energy density and the sharp growth of electrode-electrolyte interfaces and interfacial impedance [
199], [
201], [
202], which are deleterious for the high-voltage performance of LCO cathodes. Thus, larger primary particles and secondary particles have been used to increase the packing density and energy density [
199], [
201]. For example, the particle sizes of industrialized LCO are centered around 16-20 µm for high-energy performance, while 6-8 µm is employed for fast-charging performance. Therefore, the determination of the boundary particle sizes for various high-voltage and fast-charging targets is of great importance for the real industrialization of LCO.
3.4. Combinations of multiple methods
Combining multiple strategies is regarded as having much more potential than the utilization of individual modifications, as it is difficult for individual modification methods to address every challenge presented by an LCO cathode at elevated cutoff voltages. Thus, applying various strategies provides complementary advantages [
51]. For example, Chen et al
. [
203] developed an innovated LCO cathode by combining elemental doping with surface coating strategies, thereby coupling the “pillar” effect (i.e., Mg doping) with an interface shielding strategy to obtain structurally stabilized cathode materials at high voltages. The “pillar” effect of Mg adequately suppressed the irreversible phase transition and permitted the reversible shuttling of Li
+.
In addition, it was recently reported that an amorphous Co
xB
y coating layer was beneficial for effectively reducing the leakage of lattice oxygen and restraining the irreversible phase transition through the strong bonding between the surficial oxygen in the LCO lattice and the amorphous Co
xB
y layer [
203]. This surficially coated Co
xB
y layer was also conducive to mitigating surficial side reactions following Co dissolution and eventually producing a compact CEI layer. These advantages facilitated an improved cycle stability, with 95% capacity being retained at a high working voltage of 4.6 V (
Figs. 15(a)-(d)).
Aside from amorphous Co
xB
y, it is noteworthy that a Se coating layer can also interact with O
2− on the surface of LCO through a bonding effect, which is helpful in mitigating the hybridization between the Co 3d and O 2p during deep delithiation. For example, Fu et al. [
204] designed a modified LCO cathode material with Mg doping and Se surficial coating (LCO-Mg@Se), which achieved 148 mA·h·g
−1 after 400 cycles at a working voltage of 4.65 V and 172 mA·h·g
−1 with 77% capacity retained at 10 C (
Figs. 15(e)- (g)). In addition to the representative Mg dopants, Mn and La co-doping has been employed to inhibit the inherent structural instability and promote Li
+ diffusion, while a Ti-based coating layer was applied to simultaneously stabilize the interface at high voltages. Encouragingly, the co-modified LCO delivered an enhanced rate capability (1.57 mA·h·cm
−2, 3 C) and cycling performance (
Figs. 15(h)-(j)) [
205]. When combining nanosizing with an element-doping strategy, the synergistic effects of the mesoporous structure, even atomic distribution, and enhanced Li
+ diffusion kinetics support the electrochemical capability for high-voltage LCO [
206].
3.5. Others
The CEI layer on the outer shell of the LCO that is formed from conventional ethylene-carbonate-based electrolytes suffers from disintegration and reconstruction prompted by the strong oxidization properties of the high-valent Co
4+ and O
n- species generated during a high-voltage and fast-charging process [
84], [
207], [
208]. Consequently, excessive interfacial side reactions and the consumption of electrolyte and Li
+ are inevitable, which further result in a decline in cycling performance. Therefore, it is crucial to explore advanced electrolytes to improve the properties of the CEI film. For example, Qin et al. [
209] constructed
in situ formed double-layer CEI film through the addition of hexamethylene diisocyanate (HDI), 1,1,2,2-tetrafluoroethyl-2,2,3,3-tetrafluoropropylether (FE1), and polymerized diphenylamine (DPA) to the electrolyte (1 mol∙L
−1 LiPF
6 in EC:DMC, 1:1 volume ratio;
Fig. 16(a)). The amide compounds formed during the charging process via a nucleophilic addition reaction between the HDI and the charged species decomposed from FE1 at 3.5 V (vs Li/Li
+), were able to form an inner passivation layer
in situ and increase the oxidizing voltage of the electrolyte to greater than 5.0 V (vs Li/Li
+) [
210], [
211]. Then, an outer layer with ultrahigh rigidity (a Young’s modulus ∼25.4 GPa) was formed through the
in situ electrochemical polymerization of DPA (
Fig. 16(b)). With its advantages of depressed electrolyte decomposition and the Co dissolving the
in situ formed CEI films, the LCO delivered enhanced cycling stability (75% capacity retained at 4.6 V upon 200 cycles) and rate capability at a high voltage (
Fig. 16(c)). Furthermore, other electrolyte additives, such as dimethyl phenylphosphonite for preventing the exfoliation of layered LCO [
212] and nitriles/fluoroethylene carbonate co-additives for the formation of an ultrathin and uniform CEI layer [
53], were demonstrated to greatly enhance the electrochemical capability of high-voltage LCO.
Aside from the influence of electrolytes for advanced LCO design, the components making up the electrodes have a major effect on the Li
+-storage capabilities of LCO cathode materials [
213], [
214], [
215], [
216]. For example, using dextran sulfate lithium (DSL) as a substitution for the traditional poly(vinylidene fluoride) (PVDF) as a binder can suppress the irreversible phase transition between O3 and H1-3 at voltages greater than 4.55 V by evenly coating the LCO particles
in situ, resulting in superior interaction effects (
Fig. 16(d)) [
217]. In addition, graphene [
218] and super-aligned carbon nanotube (SACNT) film [
219] were employed in the construction of 3D conductive networks in LCO electrodes (
Fig. 16(e)), which enabled fast electron transfer through the electrodes regardless of electrode thickness, contributing to an impressive rate capability (128 mA·h·g
−1, 5 C) of the sandwich-structured LCO cathodes (
Fig. 16(f)).
The diffusion of Li
+ becomes more significant as the electrode thickness increases. It has been demonstrated that a thick LCO electrode suffers from sluggish Li
+ transportation kinetics and an increased concentration of polarization during cycling [
220], [
221], [
222], [
223], [
224]. This issue becomes even more serious under high-voltage and fast-charging conditions by causing detrimental phase transformation due to the higher actual local operating voltage [
33], [
63], [
217], [
225]. As an example of the benefits of a thinner electrode, He et al. [
220] designed LCO electrodes with low tortuosity (denoted as LCO-LT) via an ice-templating method (
Figs. 16(g) and
(h)). The electrodes exhibited fast Li
+ transport, slight polarization, and reversible phase transition, achieving an enhanced rate capability (57 mA·h·g
−1 with a high loading of 25 mg, 5 C), improved cyclability (94% capacity retained after 200 cycles at 0.5 C), and a homogeneous actual operating voltage throughout the electrodes.
Other factors should also be carefully considered in the future. To be specific, charging protocols are classified into passive charging and active charging based on whether the charging profiles change dynamically. In fast-charging batteries, it is essential for the charge profile to be actively modified upon changes in the internal states, which favors efficient and healthy cycling with minimal degradation [
91], [
226], [
227], [
228], [
229]. In addition, prelithiation strategies (e.g., electrochemical prelithiation, chemical prelithiation, or adding sacrificial prelithiation reagent) compensate for the irreversible lithium consumption during the formation of the CEI and SEI film in the first several cycles, which is beneficial for the electrochemical performance of the LIB [
230], [
231].
4. Summary and perspectives
Driven by the motivation to develop advanced LCO to meet the requirements for LIBs with “double-high” capabilities in 3C electronic applications, researchers have widely accepted the urgent need to design advanced LCO cathode materials with elevated cutoff voltages (4.5 V or more, vs Li/Li+) and fast-charging (5/10 C or more) properties. This work provided a systematic and insightful summarization and discussion of the key fundamental challenges (e.g., structural degradation, interfacial instability, and inhomogeneity reactions) and latest advancements in modification strategies (e.g., element doping categorized by Li-/Co- or /O-ion site, surface coating categorized by ionic/electronic conductivity, nanosizing, the synergistic effects of combinations of methods, and others) toward high-voltage fast-charging LCO. Although encouraging progress has been achieved, greater and more significant challenges remain unsolved in meeting the demand for higher voltages (4.6 V or more) and faster charging capabilities (50 C or more). In order of their likely development, these challenges include: ➀ designing high-voltage (≥ 4.6 V) LCO cathodes; ➁ designing high-voltage (≥ 4.6 V) and fast-charging (≥ 50 C) LCO cathodes; ➂ developing advanced electrolytes matching high-voltage and fast-charging LCO cathodes; ➃ obtaining an elaborated mechanistic understanding of various forms of modified LCO via in situ characterizations and simulation; and ➄ achieving high-voltage and fast-charging LCO in double-high cells.
First of all, the prerequisite for designing high-voltage fast charging LCO cathodes is to achieve higher cutoff voltages of up to 4.6 V or more. Higher cutoff voltages of LCO increase the Li+-storage performance of the cathode and the overall energy density of the LIB, which inevitably involves more phase-transition reactions, more severe loss of surface lattice oxygen and dissolution of cobalt ions, more parasitic reactions of electrolyte decomposition and CEI overgrowth, and particle cracking. Thus, novel advanced and effective modification strategies—such as a combination of multi-site dopants, a combination of multi-functional coating layers, a combination of element doping and surface coating, and so forth—for advanced LCO are of primary importance for achieving the prerequisite of high voltages (≥ 4.6 V) for high-voltage fast-charging LCO cathodes.
Second, designing advanced LCO cathodes with fast-charging (5 or 10 C) or extremely fast-charging (≥ 50 C) and high-voltage (≥ 4.6 V) capabilities is the most promising strategy to meet the requirements for the double-high properties of next-generation LIBs in 3C electronic applications. Challengingly, an extreme fast-charging capability puts high demand on fast Li+ diffusion kinetics and homogeneous inner stress distribution in LCO crystals, which is difficult to meet through the modification strategies of surface coating and element doping. Decreasing the size of the LCO particles shortens the Li+ diffusion distance and reduces the inner stress accumulation, which is necessary for reaching the target of fast charging. However, the increased surface area exposure greatly exacerbates the structural/surficial degradation and parasitic reactions at high voltages. Therefore, the strategy of nanosizing can be desirably associated with strategies for surface/bulk stabilizing, such as surface-coating layers with high ionic/electronic diffusivity, to achieve advanced LCO with high-voltage and fast-charging capability. Moreover, as the tap density and energy density will decrease upon nanosizing, the balance between fast-charging and high-energy performance must be well considered for the successful commercialization of high-voltage fast-charging LCO. Furthermore, the large-scale fabrication of size-controlled Co3O4, as an important precursor to LCO, is of great significance for the commercialization of high-voltage fast-charging LCO, based on industrialization considerations.
Third, developing high-voltage advanced liquid-based electrolytes is essential in order to match high-voltage fast-charging LCO cathodes in LIBs. The electrolyte inevitably undergoes a decomposition reaction, and the byproducts further form CEI layers in situ on the LCO surface during the initial charging process, greatly determining the performance of high-voltage and fast-charging LCO cathodes. Developing well-matched electrolytes facilitates the resistance to electrolyte decomposition at high voltages, and the byproducts of decomposition can form even, thinner, more stable, and favorably ionic/electronic conductive CEI layers, which are essential for the application of high-voltage and fast-charging LCO cathodes. In practical applications, carbonate-based solvents have been most widely utilized in electrolytes due to their advantages of low cost, excellent solubility for a variety of lithium salts, low viscosity, and high conductivity. However, they must be modified in order to inhibit the oxidation of solvent molecules and introduce better film-forming components in high-voltage and fast-charging scenarios. Promising modification strategies mainly involve:
(1) The addition of various inorganic and organic electrolyte additives to improve the stability at high voltages (e.g., boron-, silicon-, sulfur-, F-, and phosphorus-containing additives; unsaturated carbonate derivatives; and nitrile additives);
(2) The development of novel electrolyte solutions with high ionic conductivity and chemical stability through the application of co-solvents for excellent rate performance (e.g., ester, sulfone, nitrile, and fluorinated co-solvents);
(3) The development of electrolytes with a high salt concentration for enhanced high rate capability and improved high-voltage stability.
Fourth, an elaborated mechanistic understanding via in situ characterizations and simulation is needed to accelerate the rational design of high-voltage, fast-charging LCO cathodes and well-matched electrolytes. However, mechanisms for charge transfer (e.g., lattice oxygen redox, Li+ diffusion, Co ion dissolution) and electrolyte evolution (e.g., electrolyte decomposition and the formation, solubility, stability, and ionic diffusion abilities of CEI layers) are still far from being accurately and definitively known based on traditional methods. Therefore, advanced methodologies, such as in situ characterizations and theoretical simulations, are urgently needed in order to obtain deep insights into mechanisms and to construct complete and detailed structure-performance relationships so as to further facilitate the development of high-voltage and fast-charging LCO cathodes.
Last but not least, the application of high-voltage and fast-charging LCO in a pouch cell level of ampere-hour capacity is of great significance in applications with increased critical requirements, such as the interface matchability of a limited electrolyte and cathode/anode, the kinetic and capacity matchability of a cathode and anode, an advanced separator with high ionic conductivity and heat dissipation, or advanced electrode design (e.g., highly oriented porosity, a 3D electronic conductive network, and a porous current collector). Thus, many more aspects of the successful application of high-voltage and fast-charging LCO in double-high pouch cells will need to be well overcome in the future.
In summary, this review provided a systematic summary and discussion of key fundamental challenges, the latest advances in modification strategies, and prospective insights into the development of high-voltage and fast-charging LCO. It also lays a foundation for advanced LCO cathode design, facilitating the acceleration of the research and development of double-high LIBs.
Acknowledgments
This work was financially supported by the National Key Research and Development Program of China (2022YFA1504100), the National Natural Science Foundation of China (22125903, 51872283, and 22005298), Dalian Innovation Support Plan for High Level Talents (2019RT09), Dalian National Laboratory For Clean Energy (DNL), Chinese Academy of Sciences (CAS), DNL Cooperation Fund, CAS (DNL202016 and DNL202019), Dalian Institute of Chemical Physics (DICP I2020032), Exploratory Research Project of Yanchang Petroleum International Limited and DICP (yc-hw-2022ky-01), and the Joint Fund of the Yulin University and the Dalian National Laboratory for Clean Energy (YLU-DNL Fund 2021002 and 2021009).
Compliance with ethics guidelines
Gongrui Wang, Zhihong Bi, Anping Zhang, Pratteek Das, Hu Lin, and Zhong-Shuai Wu declare that they have no conflict of interest or financial conflicts to disclose.
PDF
(8227KB)
