A Review of Nanocarbon-Based Anode Materials for Lithium-Ion Batteries

Abstract

Renewable and non-renewable energy harvesting and its storage are important components of our everyday economic processes. Lithium-ion batteries (LIBs), with their rechargeable features, high open-circuit voltage, and potential large energy capacities, are one of the ideal alternatives for addressing that endeavor. Despite their widespread use, improving LIBs’ performance, such as increasing energy density demand, stability, and safety, remains a significant problem. The anode is an important component in LIBs and determines battery performance. To achieve high-performance batteries, anode subsystems must have a high capacity for ion intercalation/adsorption, high efficiency during charging and discharging operations, minimal reactivity to the electrolyte, excellent cyclability, and non-toxic operation. Group IV elements (Si, Ge, and Sn), transition-metal oxides, nitrides, sulfides, and transition-metal carbonates have all been tested as LIB anode materials. However, these materials have low rate capability due to weak conductivity, dismal cyclability, and fast capacity fading owing to large volume expansion and severe electrode collapse during the cycle operations. Contrarily, carbon nanostructures (1D, 2D, and 3D) have the potential to be employed as anode materials for LIBs due to their large buffer space and Li-ion conductivity. However, their capacity is limited. Blending these two material types to create a conductive and flexible carbon supporting nanocomposite framework as an anode material for LIBs is regarded as one of the most beneficial techniques for improving stability, conductivity, and capacity. This review begins with a quick overview of LIB operations and performance measurement indexes. It then examines the recently reported synthesis methods of carbon-based nanostructured materials and the effects of their properties on high-performance anode materials for LIBs. These include composites made of 1D, 2D, and 3D nanocarbon structures and much higher Li storage-capacity nanostructured compounds (metals, transitional metal oxides, transition-metal sulfides, and other inorganic materials). The strategies employed to improve anode performance by leveraging the intrinsic features of individual constituents and their structural designs are examined. The review concludes with a summary and an outlook for future advancements in this research field.

1. Introduction

Energy storage devices are a vital technology for the widespread use of electric vehicles and renewable energy harvesting. The demand for Internet of Things (IoT) and wearables continues to grow, with all domains of our lives being increasingly connected to the internet, including every gadget that requires a secondary energy storage device. Health sensors, in conjunction with WiFi and a miniaturized energy storage device, can be used to monitor health remotely [1,2]. Many health sensors do not require high power to operate. At present, a temperature sensor can consume 71 nW of power [3]. Motivated by increasing healthcare costs, implementations of wearable health monitoring systems ranging from simple pulse monitors to activity and sleep quality monitors, and even implantable sensors, have all stimulated the design and fabrication of low-cost secondary batteries [4,5]. As a result, these energy storage devices must be available in a variety of shapes and sizes to complement the style and functionality of their host devices. Lithium-ion batteries (LIBs) are among the most promising energy storage systems because of their excellent energy storage efficiency. The LIB is a type of secondary battery (i.e., rechargeable battery). Lithium (Li) has a high theoretical specific capacity (3860 mAh/g), a low redox potential (−3.04 V), and a low density (0.59 g/cm³). A primary battery provides energy until its storage capacity is depleted and then discarded. The anode plays a key function in LIBs and has an impact on battery performance. The physical and chemical properties of the anode material must be optimized as they influence the battery’s performance [6]. Anode materials must meet three main requirements: (a) the potential for Li insertion and extraction in the anode versus Li must be as small as possible; (b) the amount of Li that can be accepted by the anode material should be as high as possible to attain a high specific capacity; and (c) the anode hosts must withstand repeated Li insertion and extraction without structural damage to achieve a long cycle life.

Graphite has been a common choice for anode materials in LIBs because of its superior cycling stability, flat discharge plateaus, and high electronic (in-plane) conductivity owing to delocalized π-bonds and an appropriate structure for Li-ion intercalation and diffusion [7,8]. In graphite, six carbon atoms can hold one Li to form LiC₆ with the intercalation of Li between carbon layers. Based on the reaction to form the LiC₆ compound, Li⁺ + 6C + e⁻ → LiC₆, LIB anodes made of graphite material have a theoretical specific capacity of 372 mA h/g, an observed capacity of 280–330 mAh/g, and a low charging–discharging rate based on the type of graphite employed [9]. Therefore, it is limited in its ability to meet growing energy demand. Further, since graphite’s theoretical capacity has been nearly reached in practice, there is a strong interest in developing other anode replacements.

In the past, metals/metalloids (Sn, Si, Ge, Sb), metal oxides, and metal-phosphides/sulfides/nitrides (P, S, N) have been used as anode materials in LIBs. However, these materials have low rate capability due to weak conductivity [10], dismal cyclability, fast capacity fading owing to large volume expansion and shrinking leading to pulverization of the electrode, and severe electrode collapse throughout the cycle operations. Carbon is the most commonly used anode material. This is the principal material of choice, and it is the most commercially successful anode material. Further, carbon in the anode provides better electronic conductivity, a large surface area, and mechanical strength, allowing for superior electron transport and rate capability [11]. It has been shown that the reversible capacity of graphene-based materials can be as high as 1000 mAh/g [12,13]. Nanocarbons such as mesoporous graphene (MPG), carbon tubular nanostructures (CTNs), and hollow carbon nanoboxes (HCBs) are promising possibilities for LIB anodes. These nanocarbon anodes exhibit high capacities of around 1100, 600, and 500 mA h/g at 0.1 A/g for MPG, CTN, and HCB, respectively. MPG, CTN, and HCB anodes retain their capacities of 181, 141, and 139 mA h/g at 4 A/g, respectively [14]. Furthermore, these nanocarbon anodes exhibit remarkable cycling stability beyond 1000 cycles, indicating the formation of a passivating layer (a solid electrolyte interphase) that facilitates long-term cycling [14]. Nanocarbons with graphene layers that undergo the lithiation/delithiation process, a high ratio of graphite edge structure, and a large surface area that allows capacitive behavior provide high capacity and increased rate capability. High in-plane conductivity in Gr and the vertical interlayer spacing promotes improved ion intercalation, enhancing capacity and rate capability. Carbon shells in carbon nanospheres allow for volume expansion and contraction during charge storage while also assisting electron transport. The combination of these two factors results in increased cycling stability and better rate performance. The 1D carbon also supports Li-ion transport because of its high conductivity and 1D tubular architecture. Carbon shields the Li-ion from dendritic development during recharging [15]. The creation of an ideal porous structure in nanocarbons would provide the chemistry and structure needed to store Li-ions and guarantee the electrolyte ions’ accessibility while also reducing transport pathways [16].

Given the advantages and disadvantages of nanocarbons and other high-Li storage materials such as metals and transitional metal oxides, the synergetic combination of these materials as nanocomposites can produce superior outcomes when compared to individual constituents. This review examines the recently reported synthesis methods of carbon-based nanostructure composite materials and the effects of their attributes on high-performance anode materials for LIBs. These include composites with various nanocarbon architectures in combination with nanostructured materials with significantly increased Li storage capacity. In particular, 1D structures (CNTs, carbon nanofibers, and carbon nanorods), 2D structures (graphene oxide and reduced graphene oxide, doped and undoped graphene), and 3D structures (carbon nanospheres, carbon-coated metal oxide nanospheres, core-shell structures, hollow hierarchical porous carbon spheres, doped-carbon-coated transition-metal oxides, and porous carbon-based anodes) are covered.

Basics of Lithium-Ion Batteries

Traditional LIBs are made up of four parts: positive electrodes (cathodes, e.g., LiCoO₂, LiFePO₄), negative electrodes (anodes, e.g., carbon), a separator, and an electrolyte, e.g., lithium salt/LiPF₆ (Figure 1a,b). During the charge and discharge processes, Li-ions move from the cathode to the anode. The charging process happens when Li-ions travel through the separator with the help of a conductive electrolyte from the Li metal oxide cathode to the carbon-based anode. When the cell is subjected to an external load, a discharging process occurs in which Li atoms intercalated in the anode lose electrons and transform into Li-ions, which then return to the associated counter-cathode [17]. Electrons that complete the half reaction traverse an external wire since electrolytes do not contain free electrons. The composition and structure of electrodes (both cathode and anode) are important aspects influencing battery performance. To create an electrode, conductive additives and active materials (e.g., carbon black, CNTs, carbon nanofibers (CNFs), graphene, and transitional metal oxides) are attached to a current collector made of metal (e.g., Cu or Al foil) with binders based on polymers. Between active materials and the current collector, conductive additives build a network that facilitates electron transfer, and the binder maintains the electrode’s stability and integrity during battery cycling. Figure 1c depicts the procedures in sample preparation and electrochemical property characterization.

LIB performance is often measured using parameters, e.g., charging/discharging rate, specific/gravimetric capacity, specific energy, cycle stability, volumetric capacity, safety, and voltage profile. These parameters are impacted by several kinds of chemical and physical properties related to the material and electrode assembly. The specific (or gravimetric) capacity (mAh/g) relates to the quantity of energy capacity per unit mass of the battery. The volumetric capacity (mAh/cm³) is the quantity of energy capacity per unit volume of the battery. The total energy within an electrode material can be expressed as the energy or charges per unit mass (specific energy, in Wh/kg) or energy per unit volume (energy density, in Wh/L). Power density is the rate at which an electrode material can deliver energy, and it is commonly expressed in per unit mass as specific power (W/kg) or per unit volume as power density (W/L). It determines how quickly a battery can provide energy. High-energy density batteries can store a larger amount of energy but may not deliver it rapidly.

The number of charge–discharge cycles before the battery significantly loses its energy is referred to as cyclability. The average capacity retention across multiple cycles is commonly referred to as cycle stability. Cycle stability is linked to the structural and chemical stability of the electrode materials [18]. A battery’s rate of charge or discharge (C-rate) is defined as the time it takes to charge or discharge completely. For example, at 1C, it takes 1 h for the battery to fully discharge or release its maximum capacity [19], whereas at 2C, it takes 0.5 h.

Coulombic efficiency is defined as the ratio of discharge capacity to charge capacity throughout the same cycle. Initial coulombic efficiency (ICE) is an essential metric that describes the quality of the anode. It is a critical factor in full battery design since it influences the use of active materials and the weight of the assembled battery. The ICE of anode materials in half cells is determined by the ratio of the initial charge capacity to the initial discharge capacity, indicating their ability to prevent irreversible reactions and loss of capacity. This property is crucial for the practical use of anode materials in batteries, as it affects the available energy density. A low ICE value results in a low energy density of the battery. The formation of the solid–electrolyte interface (SEI) layer and the irreversible absorption of Li-ions within the anode materials contribute significantly to the low ICE of the intercalation type of anode [20]. A number of factors, including the basal to edge plane ratio, the chemical composition of the carbon surface, and the surface characteristics of the carbon (surface area and porosity), affect the formation of the SEI layer in carbon-based anodes like graphite and Gr. The SEI layer develops quickly on the edge plane and in carbon with the most surface area and defects. As a result, such characteristics can be tuned in order to obtain a low-capacity loss.

The electrical potential E (or voltage in V) at which electrode materials charge and discharge does not remain constant. Graphite usually has a rather flat discharge profile. For high-performance electronics, a flat working potential, comparable to graphite, is preferred.

2. One-Dimensional Carbon Nanocomposites Comprising Carbon Nanofiber (CNF), Carbon Nanotube (CNT), and Carbon Nanorod (CNR) Nanostructures

Nanostructured electrodes overcome the constraints of bulk crystalline expansion. It has been demonstrated that lowering the particle size of the material to the nanoscale can change the crystal structure and alter the process of volumetric expansion during lithiation [21]. CNTs have extremely high thermal conductivity, which improves composite electrode safety. CNTs have been found to have significantly higher Li capacity than graphite because of their unique structures and characteristics; they exhibit high conductivities (10⁶ S/m and 10⁵ S/m for SWCNTs and MWCNTs, respectively) and great tensile strength (~60 GPa) [22,23] and flexibility, which may improve battery cyclability. Because some carbonaceous materials can reversibly react with Li in LIB, CNTs have been proposed for use as Li storage materials [24,25].

When CNTs are utilized as anode materials, their morphology (number of defects, lengths, and diameters) has a significant impact on the electrochemical performance of LIBs. CNT defects include sidewall holes, cap removal, and edge fragmentation. The presence of holes on CNT surfaces promotes improved intercalation and diffusion of Li-ions into the tubes, enhancing their capacity. Ball milling and chemical etching are two of the most frequent ways used to alter the morphology of CNTs. Short and defective CNTs can be produced via chemical etching, mechanical cutting, or a combination of the two. Chemical etching of CNTs is the treatment of CNTs with strong acids (such as HNO₃ and H₂SO₄). This procedure generates a large number of oxygen-containing groups, which destroys the CNT structure in the form of many defects [25]. Ball milling for a suitable period of time may boost CNTs’ reversible capacity and coulombic efficiency [26]. However, both the mechanical ball-milling process and the chemical etching approach typically lead to a large loss of raw materials, inhomogeneous cutting with lengthy CNTs left, and the degradation of graphitic structure.

The mechanism by which Li-ions are stored in CNTs has been the subject of numerous investigations, including theoretical treatments [27,28,29,30,31]. Yang et al. investigated the electrochemical intercalation of Li into pristine end-closed CNTs and proposed a surface mechanism for storing Li species on the bare surface of CNTs and carbon nanoparticles. Because the CNTs investigated were end-capped and no other treatments were applied, the Li adsorbed was only discovered on the CNT surfaces [27]. Senani et al. [28] used ab initio quantum chemical computations to investigate the adsorption of Li-ions on the surface of (12,0) SWCNTs. The adsorption of a single Li-ion on the inside of SWCNTs is favored compared to the outside, as the Li adsorption energy is −0.98 eV on the inside and −0.86 eV on the outside. They also demonstrated that, following Li attachment, charge is transferred from Li-ions to CNTs, and the bonds between Li atoms and CNTs have ionic characteristics. They hypothesized that the amount of charge transfer should be proportional to the radii of curvature of CNT. Klink et al. presented a morphology modification method for gas-phase oxidation of CNTs that introduces more surface functional groups than liquid oxidation procedures [32]. MWCNTs were oxidized for varying amounts of time with liquid or gas-phase nitric acid. Gas-phase oxidized CNTs (g-CNT) have a substantially lower initial charge loss (172 mAh/g) than liquid-phase oxidized CNTs (l-CNT) (283 mAh/g) (Figure 2a,b). This is because exfoliation is less prominent in the former case, which is most likely due to an increase in surface carbonyl groups. As a result, gas-phase oxidized CNTs are thought to be better suited for battery-related applications because they have a lower specific initial charge loss, are easier to produce, and exhibit less mechanical degradation.

First-principles computations have demonstrated that Li trapped inside SWNTs has significant mobility along/around the axial direction, while radial motion is restricted. A clear link between the Li/C ratio and tube diameter was found. As the tube diameter increases, the intercalated Li atoms tend to form a multi-shell structure (Figure 2c) at equilibrium, formed of coaxial tubes with a linear chain in the axis, which improves Li capacity [33]. Li et al. have experimentally studied the Li insertion/extraction properties of nanocarbon material anodes, including SWNTs, MWNTs, C₆₀@SWNTs (C₆₀-peapod), graphite nanoflakes, and graphite nanoparticles containing some MWNTs [34]. According to the galvanostatic charge–discharge experiment, the Li storage capacity, coulombic efficiency, and cyclability of nanocarbon materials are highly influenced by their subtle architectures and subsequent treatments. Cyclic voltammetry demonstrated that Li oxidation potentials rise with the length in the Li reversible insertion/extraction route. Raman spectroscopy revealed a mechanism of Li insertion/extraction in SWNT and C₆₀-peapod electrodes, indicating that Li can enter the inner space of the tubes, the bundle’s trigonal interstitial channels, and the intercalation site between nearby tubes, with the trigonal interstitial channels being the dominant reversible sites for Li storage.

Being 1D carbon nanostructures, carbon nanofibers (CNFs) and carbon nanorods (CNRs) have similar physical characteristics to CNTs. CNFs have a small diameter and excellent conductivity, enabling efficient charge transport of both Li-ions from the electrolyte and electrons from the current collector. Furthermore, because these composite nanofibers create free-standing, flexible, and electrically conductive mats that may be employed directly as binder-free electrodes, no binders or additives are required during the battery manufacturing process [35]. Therefore, CNFs have been developed as anode materials in Li-ion batteries due to their high surface area, compact structure, and numerous pores. Such features are ideal for huge volumes of Li-ion diffusions [36]. Numerous studies have supported and published 1D carbon-based nanostructure composites as anode materials for Li-ion batteries. Recent research has focused on CNT-based anodes for LIBs, with varied degrees of efficacy depending on the treatments used.

There have been several reports of 1D nanostructured carbon composites containing metalloid elements and post-transition metals: Si-CNT [37], Ge-CNF [38], Sn nanoparticles/CNT [39]. SnO₂ nanocrystals/MWCNT [40], Si/C@CNF [41], SiO nanoparticles/CNT [42], Si@C–SiO₂ [43], Sn/CNF, Si/CNF, Ge/CNF [38,44], Sn/SnO₂@CNF [45], and SnO₂@CNF [46]. Further, 1D nanostructured carbon-metal oxides composites also have been reported: Fe₂O₃ nanorods/CNFs [47], GeO₂/multichannel CNF [48], Co₃O₄ nanoparticles/MWCNT [49], porous FeCO₃/CNF [50], Zn₂SnO₄ nanoparticles/CNT [51], TiO₂ coated SnZnO/CNF [52], FeCO₃ NRs/CNT [53], MgFe₂O₄/CNF [54], Mn₃O₄ nanoparticles/CNT [55], Fe₃O₄ nanoparticles/CNF [56], MnO₂/CNF [57], MnO₂@CNT fabric [58], NiCo₂O₄ nanoparticles/CNT [59], and MnO/CNF free-standing fabric [60]. CNTs can be employed as a support matrix to create CNT/metal composites that take advantage of metals’ higher capacities [61,62]. This strategy has many advantages: it permits an anode to benefit from the high Li capacity without the issue of pulverization. When metal nanoparticles abruptly alloy with Li and grow in size, an anode can remain intact in its structure because the CNT works as a flexible network, keeping the metal nanoparticles linked to the anode current collector. When metal nanoparticles alloy and dealloy, the CNTs can transport electrons to and from them. Furthermore, CNTs can store any additional Li that has not been alloyed with metallic nanoparticles [24].

Si, with a theoretical capacity of ~4200 mAh/g, has long been recognized as an appealing material for anodes in energy storage systems. However, the structural deterioration and instability of the SEI, as well as the low electrical conductivity, hampered commercialization. The volume expansion and poor conductivity can be remedied using CNTs. Zhu et al. used modified CVD to produce a neuron-like Si-CNT composite electrode (Figure 3a–c) [37]. The neuron-like Si-CNT composite not only contributed to the Si anode’s excellent electronic conductivity and mechanical stability, but it also efficiently compensated for the Si particles’ substantial volume change. When compared to the physical Si + CNT mixture, the Si anode’s rate capability and cycling stability are greatly improved. The neuron-like Si-CNT composite nonetheless delivered a reversible capacity of 1161.5 mAh/g at a high current density of 20 A/g. After 200 cycles, capacity retention reached 96% (Figure 3b). The electronic conductivity of the Si + CNT mixed electrode laminate was 0.81 S/cm, compared to approximately 0.52 S/cm for the bare Si electrode. The conductivity of the as-prepared neuron-like Si-CNT composite electrode was 0.95 S/cm (Figure 3c).

Emerging nanostructure such as Si nanowire has been shown to be capable of repeated discharges of >3000 mAh/g [64]. The nanowires resist anode crumbling because their small width reduces the effect of volume expansion, allowing them to expand and contract without breaking. A paper-like ultrathin and flexible Si/CNT composite anode for LIBs was created via conformal electrodeposition of a thin layer of Si on CNTs at room temperature (RT) [65]. The flexible Si/CNT composite had a high volumetric capacity (~1400 mAh/cm³ at 200 mA/g) due to its active material and current collector. Furthermore, prelithiation treatment can effectively increase the Si/CNT composites’ low initial coulombic efficiency, and a red LED can be easily lit by a full pouch cell with a Si/CNT composite as a flexible anode in flat or bent states. Similarly, carbon-coated Si nanoparticles (30–100 nm in diameter) dispersed in CNT (6–13 nm in diameter, 2.5–20 μm in length) networks were tested as an anode material for LIBs. This material showed a capacity retention of 70% after 40 cycles [66]. The carbon layer in Si@C-CNT improved electrical conductivity while allowing for volume variations. The CNT network strengthened the electrode, stabilized the electric network for active Si, and improved cycle performance. In contrast, without the CNT network, the Si@C anode developed cracks, pulverization, and disintegration due to volume expansion.

A chemical reduction approach was used to develop a Sn/carbon nanotube (CNT) composite anode material for LIB [39]. The Sn/CNT composite displayed consistent cycling performance with a reversible capacity of 413 mAh/g at the 100th cycle and a high initial reversible capacity of 630.5 mAh/g. The increased electrochemical performance of the Sn/CNT composite might be due mostly to the excellent dispersion of Sn nanoparticles on CNTs and the partial filling of Sn nanoparticles within the CNTs. It was proposed that chemical treatment of CNTs with concentrated nitric acid, which cuts CNTs into small fragments and increases the number of oxygen-functional groups on the surface, serves an important role in positioning Sn nanoparticles on CNTs and preventing Sn nanoparticle clustering during the charge-discharge process.

Sn-based oxide materials, such as SnO₂, with high theoretical reversible specific capacity (e.g., 781 mA h/g), have been widely studied as prospective alternatives for graphitic anode materials (theoretical capacity: 372 mA h/g). However, the main disadvantage of its industrial application is poor cycle performance, which stems from the significant specific volume change caused by repetitive charging and discharging of the battery, resulting in mechanical failure and loss of electrical contact at the anode [46,67]. To improve cycle performance, anodic capacity, and rate capacity, SnO₂ anode materials must be morphologically and structurally modified. The introduction of carbon components into SnO₂@C composites is an effective technique for improving their electrochemical performance.

Electrospinning was used to fabricate SnO₂@C nanofibers (NFs), which were then tested as anodes in LIB half cells. The NF anodes had initial discharge capacities of 1375.5 mAh/g at 80 mA/g current density [46]. This high Li-ion storage capability of SnO₂ NFs is most likely due to the amorphous carbon protection and the synergy created by the ultrafine SnO₂ particles implanted in the CNF matrix. The nanometer-sized SnO₂@C NFs can not only provide minimal ion diffusion durations, resulting in faster phase transitions, but also ample room to buffer volume changes during the Li insertion and extraction process.

Gao et al. developed a method for producing designable and well-controllable multi-wall Sn/SnO₂@C hollow nanofibers: wire-in-tube (WIT), tube-in-tube (TIT), and wire-in-double-wall-tube (WIDWT) nanofibers (Figure 3d) [63]. WIDWT nanofibers can increase the contact area between the electrolyte and electrode while also providing active areas for redox processes. Throughout the charging and discharging process, the twin tube inner layers provide a suitable but not excessive gap between the core and shells, thus reducing volume expansion and guaranteeing favorable transport kinetics for both Li-ions and electrons (Figure 3d). The Sn/SnO₂@C composite outperforms bare SnO₂ in terms of electrochemical performance (Figure 3e,f). The collaboration of multi-structure and composition optimization contributes to the great rate capability (812.1 mAh/g at 1 A/g and 566.5 mAh/g at 5 A/g) and better cycling performance (986.3 mAh/g at a current of 1 A/g and 508.2 mAh/g at 5 A/g after 2000 cycles).

As stated previously, Si has the highest theoretical capacity (4200 mAh/g), but its volume changes by 400% during the anode operation [68]. Ge has a lesser theoretical capacity of 1600 mAh/g, and its volume change is reduced to 370% [69]. Sn has the lowest theoretical capacity of 990 mAh/g, with a volume change of only 259% during the alloying process [70]. In order to enhance electrode conductivity—which is progressively lost owing to the significant volume change—and enable strain relaxation, nanocomposites of Si, Ge, and Sn containing confinement materials (such CNFs) have been produced. They have shown promising specific capacities and cyclability for use in LIBs [71,72]. Electrospinning and carbonization were used to create Si/C, Ge/C, and Sn/C composite nanofibers with varying nanoparticle content [38]. The resulting composite nanofibers were employed as anodes in Li-ion half cells, where they demonstrated reasonably high reversible capacities and good cycle performance. CNFs have a small diameter and high conductivity, allowing for effective charge transport of Li-ions from the electrolyte and electrons from the current collector. When Si/C, Ge/C, and Sn/C composite anodes were compared, Si/C nanofibers had the largest initial coulombic efficiency due to the minor effect of SEI. Nevertheless, Ge/C and Sn/C anodes excelled compared to Si/C anodes in terms of cycle performance, especially at higher filler concentrations, owing to superior nanoparticle distribution in carbon nanofibers and a lesser volume change during the Li insertion and extraction processes.

Because of its high theoretical capacity, adequate lithiation potential, low cost, and ecologically friendly nature, Mn₃O₄ is regarded as a promising LIB anode material [55]. However, the low electronic conductivity and rapid volume increase during cycling make improving Li-ion storage qualities difficult. Mn₃O₄ nanoparticles (~18 nm) anchoring on CNTs (CNTs@Mn₃O₄) are created by employing CNTs@MnO₂ as a precursor, which is made in a simple room-temperature technique. Taking advantage of the features provided by the microstructure and hybrid material components, the as-prepared CNTs@Mn₃O₄ electrode has a reversible capacity of 895.0 mAh/g at 0.5 A/g after 200 cycles and 743.7 mAh/g at 5.0 A/g when utilized as a LIBs anode [55].

Cyclic voltammetry (CV) is a popular technique for studying the oxidation, reduction, and phase transformation processes in electrode reactions. The electrochemical performance of the CNTs@Mn₃O₄ was assessed using CV at a scan rate of 0.25 mV/s as a LIB anode (Figure 4a). The first lithiation step generates two separate cathodic peaks. The cathodic peak at ~0.15 V is related with the reduction of Mn₃O₄ into Mn metal nanoparticles, whereas the left peaks at ~0.05 V are attributed to CNT lithiation. Three peaks at around 0.20, 1.25, and 2.10 V were seen during the anodic process, corresponding to CNT delithiation, Mn metal oxidation into Mn₃O₄, and higher valence Mn oxides. In subsequent cycles, the reduction potential of Mn₃O₄ changes to 0.30 V, and the voltage hysteresis between the reduction and oxidation processes decreases, implying that the kinetics of the electrochemical reaction improve after the first lithiation–delithiation event. Figure 4b shows the rate capability of the CNTs@Mn₃O₄ electrode.

Because of their high capacity, low cost, and natural availability, FeCO₃ nanomaterials have been extensively researched as anode materials for LIBs. Nonetheless, the sluggish diffusion of Li-ions in bulk materials and their low electronic conductivity limit FeCO₃’s power performance. The FeCO₃/carbon nanotube composites (FeCO₃/CNTs) were created employing CNTs, ferrous sulfate, and urea as raw ingredients in a single-step in situ crystallization method. The inclusion of modest CNTs can significantly improve the electrochemical performance of FeCO₃. At a current density of 200 mA/g, FeCO₃/CNT-15 (FeCO₃/CNT-x, where x is wt%) has an initial discharge capacity of 1436 mAh/g and retains a capacity of 694 mAh/g after 140 cycles [53]. Figure 4c shows plots of discharge–charge capacities against cycle number for voltages ranging from 0.05 to 3 V and current densities of 200 mA/g. The as-prepared FeCO₃ and FeCO₃/CNT composites can achieve an initial discharge capacity of more than 1400 mAh/g. The generation of SEI films and electrolyte deterioration are the primary causes of the first capacity reduction. The capacities decreased during the first twenty cycles, but subsequently stabilized at 451, 560, 550, 694, and 481 mAh/g for FeCO₃, FeCO₃/CNT-5, FeCO₃/CNT-10, FeCO₃/CNT-15, and FeCO₃/CNT-20, respectively, throughout the next 140 cycles. FeCO₃/CNT-15 demonstrated the best electrochemical performances. After 10 cycles, the coulombic efficiency of FeCO₃/CNT-15 approached 100%, and the maximal reversible capacity of 694 mAh/g was maintained even after 140 cycles. The structure of FeCO₃/CNT-15 after 140 cycles stayed almost identical to that of FeCO₃/CNT-15 prior to the cycling stability test [53].

Transitional metal oxides have been observed as suitable anode materials for LIBs [73]. The spinel NiCo₂O₄ (NCO) has a theoretical capacity of 890 mAh/g and a two-fold higher electronic conductivity than nickel oxides or cobalt oxides [74]. However, NCO has limited cyclability and rate capability because of its slow electron/ion transport rates and significant volume expansion/contraction during the charging–discharging process [75], restricting its potential usage in high-energy storage systems. To solve this issue, by lowering the size of nanoparticles and building hybrid nanostructures of NCO and a carbon matrix, one may achieve good electrochemical performance due to the shorter ion or electronic diffusion path [76]. Microwaves have attracted a lot of attention in a variety of fields, including microwave-absorbing applications [77], thermoelectric applications [78], and supercapacitors [79], because of their ability to shorten reaction times, increase yields, and allow for the fabrication of nanostructures with smaller and more uniform sizes, which results in improved performance.

Zhang et al. used a novel and rapid microwave-assisted approach to successfully fabricate porous NiCo₂O₄@CNTs 1D structure nanocomposites [59]. The hybrid material was composed of ultrafine NiCo₂O₄ nanoparticles measuring 6–10 nm that were securely anchored on the surface of CNTs. When compared to pristine NiCo₂O₄, the NiCo₂O₄@CNTs composite material demonstrated more intriguing electrochemical properties, such as a greater specific capacity, an excellent current charge–discharge capacity, and improved cycle stability. As a negative electrode for LIBs, a flexible carbon electrode was developed by impregnating magnetite nanoparticles (Fe₃O₄-NPs, ~32 nm) on electrospun CNFs [56]. The Fe₃O₄-NPs particles were found to be well adsorbed on 400 nm CNFs. The addition of Fe₃O₄-NPs increased the electrochemical performance of the CNFs, resulting in an initial high discharge capacity of 1146 mAh/g—higher than that of pure CNFs (480 mAh/g). The composite active material also demonstrated a higher coulombic efficiency (90%). However, because of the reduced electrical conductivity of Fe₃O₄-NPs, the composite showed a modest drop in capacity retention at high C rates.

Using a combination of hyper-crosslinking, liquid phase oxidation, and a hydrothermal (HT) technique, three types of carboxyl modification tubular CFs (CMTCFs) and MnO₂ composite materials (CMTCFs/MnO₂) were created [57]. The HT reaction period was adjusted (6, 8, and 12 h) to control the morphology and crystal phase of MnO₂ in CMTCFs/MnO₂. CMTCFs wrapped in δ-MnO₂ nanosheets (CMTCFs@MNS) were employed as anodes and compared to the other two CMTCFs/MnO₂. According to electrochemical study, the CMTCFs@MNS (which underwent a 6 h HT treatment) electrode has a large reversible capacity of 1497.1 mAh/g after 300 cycles at 1000 mA/g and a long cycling reversible capacity of 400.8 mAh/g after 1000 cycles at 10,000 mA/g. At the same time, following twelve modifications in current density, CMTCFs@MNS exhibits an avg. reversible capacity of 256.32 mAh/g at 10,000 mA/g.

Tzeng et al. improved anode cycling performance by partially encapsulating Si flakes in silicon dioxide and covering them with conductive nanocarbon sheets and CNTs [80]. A silicon dioxide surface layer on a Si flake improves the physical integrity of a Si-based anode. The exposed Si surface facilitates the quick passage of Li-ions and electrons. CNTs and nanocarbon sheets operate as electrical contacts between the Si flakes and the current collector [80].

In carbonaceous materials, heteroatom doping (donor or acceptor atoms) is useful for controlling the active site, ion transport, and electronic transfer for electrochemical energy storage [81,82,83,84]. N and B doping in carbon can result in increased conductivity and pseudocapacitance [85,86,87]. Further, due to their comparable atomic radii, both B and N atoms are capable of reaching a high doping concentration. A composite, Li₄Ti₅O₁₂ (LTO)/MWCNT-co-doped with nitrogen (N) and boron (B), has shown a reversible capacity of 120 mAh/g at a rate of 20 C [88]. After 150 cycles at 20 °C, 97.5% of the capacity was preserved, with minimal capacity fading. The co-doped CNTs’ improved electrochemical performance compared to LTO could be attributed to the separate forms of N and B. This not only maintained the benefits of the N and B additives, such as good electron-donating and electron-accepting capabilities, but also resulted in a synergistic effect due to their unique electronic structure [89]. To obtain the composite, a mixture of N, B-CNT, and LTO was calcinated.

Transition-metal sulfides (TMSs) involving conversion and/or alloying reactions are regarded as promising anode materials for advanced LIBs and sodium-ion batteries (SIBs) [90]. However, the poor electronic conductivity and significant volume expansion severely limit their practical application. Hou et al. proposed a covalent coupling technique for TMS-based anode materials that employs amide linkages to bind TMSs and carbon nanotubes (CNTs) [90]. During the synthesis, thiourea serves as both the capping agent for morphological control and the linking agent for covalent coupling. In this covalently coupled ZnS/CNT composite, the ZnS nanoparticles (∼10 nm) securely anchored to CNT bundles. The compact ZnS-CNT heterojunctions are highly beneficial for promoting electron/ion transmission and maintaining structural integrity. Because of the strong coupling interaction between ZnS and CNTs, the composite exhibits prominent pseudocapacitive behavior and highly reversible electrochemical processes, resulting in reversible capacities of 333 mAh/g at 2 A/g over 4000 cycles for LIBs and 314 mAh/g at 5 A/g after 500 cycles for SIBs.

3. Two-Dimensional (2D) Carbon Nanostructure (GO and Gr) Composites

Because of their unique structures, 2D carbon nanomaterials have attracted a lot of attention for use in flexible LIBs due to their large specific surface area, excellent charge mobilities, inherent flexibility, and high Li storage capacity.

Figure 5. (a) Schematic of the most common Gr production methods. Each method has been evaluated in terms of Gr quality (G), cost aspect (C; a low value corresponds to high cost of production), scalability (S), purity (P), and yield (Y) of the overall production process [91]. (b) Phosphorus covalently crosslinked SnO₂ and Gr. Illustrations of formation process of SnO₂/GO and SnO₂@[email protected] with permission from ref. [92]. © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. (c–f): Carbon layer encapsulated Fe₃O₄@rGO. (c) Schematic diagram of the Fe₃O₄/rGO and C/Fe₃O₄/rGO composites; (d) cyclic performance at 0.2 A/g of the Fe₃O₄, Fe₃O₄/rGO and C/Fe₃O₄/rGO electrode; (e) CV curves with capacitive contribution at 2.4 mV/s; and (f) separation of contributions from capacitive effects and diffusion controlled capacities at different scan rates. Adapted with permission from ref. [93]. © 2020 Elsevier Ltd. and Techna Group S.r.l.

Figure 5. (a) Schematic of the most common Gr production methods. Each method has been evaluated in terms of Gr quality (G), cost aspect (C; a low value corresponds to high cost of production), scalability (S), purity (P), and yield (Y) of the overall production process [91]. (b) Phosphorus covalently crosslinked SnO₂ and Gr. Illustrations of formation process of SnO₂/GO and SnO₂@[email protected] with permission from ref. [92]. © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. (c–f): Carbon layer encapsulated Fe₃O₄@rGO. (c) Schematic diagram of the Fe₃O₄/rGO and C/Fe₃O₄/rGO composites; (d) cyclic performance at 0.2 A/g of the Fe₃O₄, Fe₃O₄/rGO and C/Fe₃O₄/rGO electrode; (e) CV curves with capacitive contribution at 2.4 mV/s; and (f) separation of contributions from capacitive effects and diffusion controlled capacities at different scan rates. Adapted with permission from ref. [93]. © 2020 Elsevier Ltd. and Techna Group S.r.l.

Graphene (Gr), which has atoms organized in two dimensions to form a sheet structure, has robust mechanical flexibility, its surface area is large, its chemical stability is outstanding, and its electrical conductivity is excellent [94]. However, its coulombic efficiency is low. It is proposed that Gr’s double atomic surfaces allow it to store more Li atoms than graphite. The interaction between Li and Gr is facilitated by charge transfer and the van der Waals (vdW) effect. The coulombic interaction is governed by charge transfer and is thus mostly determined by the electronic structure. As a result, the coulomb interaction in Gr is similar to graphite’s. With the vdW effect, Li adsorption on Gr is somewhat weaker than in graphite. Introducing vacancies, pores, or defects on Gr enhances the adsorption of Li, leading to an increase in capacity [95]. Additionally, Gr has derivatives that include graphene oxide (GO) and reduced graphene oxide (rGO), both of which include different kinds of oxygen functional groups. GO and rGO are particularly useful for wrapping and dispersing nanoparticles, reducing volume strain, and facilitating Li storage and electron transport. Figure 5a depicts a wide range of ways used to produce Gr [91]. Gr may be built into a 2D thin film using vacuum filtering and a 3D network using solvothermal/CVD techniques to create suitable supporting substrates for the direct development of active cathode materials. This presents numerous prospects for Gr and Gr derivatives to develop binder-free nanostructured electrodes with high reversible capacity in LIBs. To maximize the performance of Gr as an anode material for Li-ion batteries, Gr composites with supporting material are preferable to Gr as a single component. The quantity of ions per gram of material controls the battery’s capacity, and consequently its energy. Gr, like graphite, can be utilized as an anode for hosting Li⁺ as a carbonaceous matrix in composites with other Li-storing materials. Numerous studies on Gr-based materials for electrochemical energy storage have been conducted in the last few years [96]. Several Gr composites with post-transition-metal oxide and phosphor have been found in a series of studies in the literature: SnO₂/Gr [97], 3D SnO₂/rGO [98], carbon-coated SnO₂/rGO foam [99], cobalt-doped SnO₂/Gr sheets (Co-SnO₂@GS) [100], MOF/SnO₂/Gr [101], rGO/Fe₂O₃/SnO₂ ternary nanocomposite [102], SnO₂/NiFe₂O₄/Gr [103], and SnO₂@P@rGO [92]. Further, Gr/transition-metal oxide composites have also been reported: Gr anchored with Co₃O₄ nanoparticles [104], CuO/Gr [105], MnO/Gr [106], Zn₂GeO₄ nanorod/Gr composites [107], CoO-rGO [108], hollow 3D mesoporous CoFe₂O₄ NSs/Gr [109], core shell structured ZnFe₂O₄@C/G nanocomposite [110], CNF/MnO/rGO thin film [111], nanocubic CoFe₂O₄/Gr [112], Mn₃O₄/rGO [113], hollow α-Fe₂O₃ nanotube-Gr [114], C/Fe₃O₄/rGO [93], hollow Co₃O₄ nanoparticles/porous rGO [115], CoO NW@Gr cloth [116], ZnFe₂O₄/rGO [117], porous α-Fe₂O₃ nanoparticles/Gr [118], hierarchical nanospheres Fe₂O₃/Gr [119], TiO₂ nanotubes/rGO [120], and self-supporting Co₃O₄/Gr films [121].

Metal oxide/Gr composite material structures can be generally classified into the following models: widely employed anchored type (SnO₂ [122], Fe₂O₃ [123]), wrapped (Fe₃O₄ [124], MoO₂ [125]) encapsulated (Fe₃O₄ [126], Fe₂O₃ [127]) sandwich-like (TiO₂ [128]) layered stacking (SnO₂–NiO–MnO₂ [129], MnO₂ [130]), and mixed (Li₄Ti₅O₁₂ [131], Fe₂O₃ [132]) (Figure 6).

Table 1. The electrochemical performance of 2D nanocarbon-based anodes documented in the literature. Abbreviations: Gr = graphene; HT = hydrothermal.

Composites	Synthesis Method	1st Discharge–Charge Capacity (mAh/g)	Current Density (mA/g)	Cycling Performance (mAh/g): Capacity Retention (Number of Cycles)	Current-Density Applied (mA/g)	Reference
Zn2GeO4 Nanorod/Gr	HT	1508/1020	200	768 (50)	200	[107]
CuO/Gr composite	solution method	900	65	600 (100)	65	[105]
SnO2/Gr	Chemical method	1719/852	100	634 (50)	100	[97]
Co3O4 nanoparticles@Gr	Solution phase dispersion	1097/753	50	935 (30)	50	[104]
Gr/Fe2O3/SnO2	In situ precipitation followed by reduction	1179/746	168	700 (100)	400	[102]
SnO2/Gr	In situ reduction, spray drying, thermal treatment	1542/837	500	704 (150)	500	[98]
3D C-Coated SnO2/rGO foam	One-pot HT method	2544/960	100	717 (130)	100	[99]
SnO2@P@GO	Solvothermal	1278/405	100	550 (200)	1000	[92]
3D CoO–Gr hydrogel	HT	1230/984	100	1010 (100)	100	[108]
CNF/MnO/rGO thin free- standing binder-free film	Electrospinning, Electrospraying	1582/1107	100	574 (3000)	5000	[111]
Hexagonal Mn3O4@RGO	Chemical reduction	1086/884	—	638 (250	123	[113]
3D sandwich-like C/Fe3O4/rGO	Solvothermal	1315/721	200	832 (300)	200	[93]
Hollow-Co3O4/Porous-rGO	Facile etching strategy and the Kirkendall effect	1164/813	200	1016 (200)	200	[115]
ZnFe2O4 nanoparticle/rGO	Metal reduction	1939/1285	1000	1022 (500)	1000	[117]
Cu2Sn3S7/Cu2SnS3/SnS2/rGO	Liquid phase	1043/859	500	965 (300)	500	[134]
Mesoporous Fe2O3/N-doped Gr	HT	1426/1143	0.1 C	905 (250)	1 C	[135]
SnO2/G coated with polyacrylic acid	In situ polymerization, In situ carbonization	1652/987	100	480 (500)	1000	[136]
Co-doped SnO2 and Gr sheets	Hydrolysis	2102/1280	100	828 (200)	500	[100]
Cuboid particles MOF/SnO2/Gr	Solvothermal	1332/660	100	450 (1000)	1000	[101]
SnO2/NiFe2O4/Gr	HT	1465/905	200	613 (100)	800	[103]
MnO on Gr	Solution based, subsequent reduction	891	200	843 (400)	2000	[106]
ZnFe2O4@C/Gr	Solvothermal, mussel inspired	1059/688	232	705 (180)	232	[110]
Hollow mesoporous CoFe2O4 nanospheres/Gr	HT	1459/942	100	805 (300)	2000	[109]
Nanocubic CoFe2O4/Gr	HT	1473/953	200	835 (200)	1000	[112]
MoS2 and N-doped Gr (MoS2/NG)	HT	1884/818 at 200	200	817 (120)	200	[137]
3D Gr aerogel/hollow α-Fe2O3 nanotubes	HT	1973/1320	100	1512 (65)	100	[114]
Nanoporous CoO NW/3-D porous Gr	HT; annealing	1493/1231	200	1246 (350)	500	[116]
Black phosphorus/Gr	Solvothermal	2587/1836	-	1401 (200)	100	[138]
Mesoporous manganese sulfide spheres/Gr sheets	In situ solvothermal	1444/1231 at 50	50	735 (300)	200	[139]

summarizes the performance of 2D carbon nanostructure composites as LIB anodes mentioned in the literature.

Glucose was utilized as a crosslinking agent in the HT process to improve the bond between SnO₂ and rGO. The authors developed a carbon-coated SnO₂/rGO (SnO₂/rGO/C) foam by constructing a 3D foam structure of distributed SnO₂ nanoparticles on rGO sheets and covering it with a thin layer of amorphous carbon using a HT approach and then freeze drying [99]. The carbon layer may aid in the binding of active SnO₂ nanoparticles to rGO sheets, increasing the composite’s electrical conductivity. The electronic conductivity of the SnO₂/rGO/C foam was found to be 2.44 S/cm, which is significantly greater than that of the SnO₂/rGO foam (0.24 S/cm) without the carbon layer. The rGO and carbon layer collaborate to successfully preserve the structural arrangement’s stability, preventing SnO₂ detachment from the rGO substrate. The as-prepared SnO₂/rGO/C film electrode outperforms SnO₂/rGO and bare SnO₂ as anode materials for LIBs in terms of specific capacity, rate performance, and cycling performance due to its 3D porous nanostructure and excellent mechanical flexibility. This outstanding performance is due to the synergistic action of SnO₂, rGO, and the carbon layer. As a free-standing anode for LIBs, the foam demonstrated high capacity and outstanding cycling stability (717 mAh/g after 130 cycles at 100 mA/g).

Covalent bonds, which are chemically stable, can give great mechanical integrity and an electrical connection between two materials [140,141]. SnO₂ cycle performance can be improved by creating a nanostructure with covalent bonds. Zhang et al. created a phosphorus covalently crosslinked SnO₂ and graphene (reduced from graphene oxide (GO) and indicated as SnO₂@P@GO, Figure 5b) [92]. Phosphorus covalently connects both Gr and SnO₂ nanocrystals in this unusual structure via P-C and Sn-O-P bonding, respectively, and acts as a buffer layer during charging and discharging (Figure 5b). As a result, SnO₂@P@GO can support the large volume variation of SnO₂ while retaining outstanding structure integrity and exhibiting long-life, high-rate cycling. As a result, when used as the anode of a LIB, SnO₂@P@GO demonstrated excellent stable performance, preserving 95% of the second capacity after 700 cycles at a current density of 1 A/g. An elastic conductive network and fast electron transport were provided by Gr. Even after 700 cycles, the SnO₂ particles are completely wrapped up by Gr rather than being exposed to the electrolyte, improving the structure’s stability even further.

Pseudocapacitance arises when quick, reversible faradaic reactions occur on or near the surface of an electrode material, causing charge accumulation in the interfacial region. The surface area affects the pseudocapacitive storage of transition-metal oxide electrodes. If the surface area grows, the pseudocapacitive becomes more significant. Most carbon nanomaterials with a large surface area, such as porous Gr, simply provide large pseudocapacitance for delithiation. Consequently, pseudocapacitive storage has a direct impact on rate capability and cycling performance. Wu et al. [93] found that increasing the capacity of C/Fe₃O₄/rGO for 300 cycles resulted in pseudocapacitive behavior, as illustrated in Figure 5d. This is related to the transition of Fe₃O₄ nanoparticles into smaller ones after 100 cycles. Figure 5e,f show a comparison of C/Fe₃O₄/rGO’s capacitive effect and diffusion effects. This study adopts a two-step strategy to developing a double carbon layer 3D network composite anode material of C/Fe₃O₄/rGO for LIB. First, GO is utilized to hydrothermally produce interlayer cauliflower-like Fe₃O₄. Then, a layer of carbon source is applied to the surface of Fe₃O₄/GO and the result is annealed to create the C/Fe₃O₄/rGO composite. Annealing increases the surface interaction of Fe₃O₄, rGO, and carbon (Figure 5c). Carbon material applied to the surface of Fe₃O₄/rGO can improve conductivity and electron transport in the electrode. This 3D nanostructure limits the volume growth of the C/Fe₃O₄/rGO composite throughout charging and discharging. After 300 cycles, the C/Fe₃O₄/rGO composite had a specific capacity of 844 mAh/g (at 0.2 A/g cycling) and outstanding rate capability (Figure 5d). Furthermore, cyclic voltammetry analysis of the pseudocapacitance of the C/Fe₃O₄/rGO composite at various scan rates demonstrates that the material contains a significant amount of pseudocapacitance. At 0.1, 0.4, 1.2, 2.4, and 5 mV/s, pseudocapacitance constitutes 31, 37, 49, 57, and 77% of total capacity, suggesting that the C/Fe₃O₄/rGO composite exhibits considerable pseudocapacitance behavior during high rate performance (Figure 5f). Figure 5e depicts the separation of capacitance and diffusion currents at a 2.4 mV/s scan rate. The electrochemical performance of Fe₂O₃/rGO produced by oleic acid-assisted HT in situ is characterized by a first-cycle discharge capacity of 1431 mAh/g at a current density of 100 mA/g, which decreases to 600 mAh/g after 30 cycles [142]. The cycle reversible capacity remains at 478 mAh/g after 50 cycles. Gr has a conductive network that greatly improved conductivity and electrolyte transfer. The steric hindrance of Gr may limit the crystal formation of iron oxide, enhance or decrease its active site, and reduce the route of Li-ions in the crystal [142].

In addition to binary compounds, ternary compounds were employed to form Gr composites. Ternary nanocomposite, as an alternative material, demonstrated increased capacity and reduced volume expansion. Zn₂GeO₄ nanorod/Gr composites (ZGCs) were produced using a two-step HT procedure [107]. A single Zn₂GeO₄ nanorod had both crystalline and amorphous areas. Gr sheets were used to compactly cover and anchor the Zn₂GeO₄ NR surface. As an anode for LIBs, partially crystalline ZGC containing 10.2 wt% Gr exhibited excellent electrochemical performance: reversible capacity of 1020 mAh/g in the first cycle, cyclic performance of 768 mAh/g after 50 cycles, and excellent rate capability of 780 mAh/g at a current density of 0.8A/g. The amorphous zone in partly crystalline Zn₂GeO₄ NRs and the elastic Gr sheets allowed for volume changes during the charge and discharge processes. Xia et al. reported a ternary nanocomposite of rGO/Fe₂O₃/SnO₂ formed by homogenous precipitation of Fe₂O₃ nanoparticles onto GO, followed by reduction of GO with SnCl₂ [102]. SnO₂ nanoparticles were deposited on the GO surface when the GO was reduced. Fe₂O₃ nanoparticles (20 nm) were uniformly disseminated in the Gr nanocomposite, surrounded by SnO₂ nanoparticles. Because of the differing Li insertion/extraction potentials, the primary role of SnO₂ nanoparticles during cycling is to inhibit Fe₂O₃ aggregation. Gr can act as a matrix for Li⁺ and electron transport, alleviating stress that would otherwise accrue in the Fe₂O₃ nanoparticles during Li absorption and release. The electrochemical performance of a ternary nanocomposite of rGO/Fe₂O₃/SnO₂ as an anode in LIBs has been greatly enhanced, with initial discharge and charge capacities of 1179 and 746 mAh/g, respectively. Nearly 100% charge–discharge efficiency was maintained over the next 100 cycles with a specific capacity of >700 mAh/g.

As with CNT, Gr has been frequently alloyed or mingled with Si to increase Li storage capacity. Mn, like Si, is an alloying anode material having an excellent capacity; however, its performance is limited by large volume expansion. A ternary nanocomposite of Si, Mn, and rGO (Si–Mn/rGO), wherein the Si–Mn alloy offers high capacity properties and embedded rGO nanosheets provide structural stability, was prepared by mechanical complexation and thermal reduction of Si nanoparticles, MnO₂ nanorods, and rGO nanosheets in combination [143]. The LIB anode material had a specific capacity of 600 mAh/g and retained around 90% capacity after 50 cycles at 100 mA/g. The improved performance is due to facilitated Li-ion interactions with the MnSi alloy phase, as well as the creation of a structurally reinforced electroconductive matrix composed of rGO nanosheets.

CoFe₂O₄, a binary transition-metal oxide, is an attractive option due to its high theoretical capacity (916 mA h/g) and thermal/chemical stability [144,145]. CoFe₂O₄, like other transition-metal oxides (TMOs), undergoes significant volume change during cycling, causing pulverization and aggregation. This destroys the electrode’s structural integrity and causes a loss of electrical connection with the current collector, resulting in rapid capacity fading [146]. Furthermore, CoFe₂O₄’s innately low conductivity makes it difficult to achieve excellent electrochemical performance [147]. Combining CoFe₂O₄ electrodes with carbon materials including carbon nanofiber [148], nanotubes, nanosheets [149], and Gr [150] is a typical technique to improve its electrochemical performance.

A simple one-pot technique was used to create a nanocomposite of CoFe₂O₄ nanocubes (edge length of ~5–15 nm) that were evenly distributed and adhered to Gr [112]. The results reveal that the inserted Gr oxide can cause nanocubic CoFe₂O₄ to form as well as close the junction between CoFe₂O₄ and Gr (Figure 7a). The plane-to-plane contact mode between CoFe₂O₄ nanocubes and Gr nanosheets in the composite nanostructure not only facilitates electron and ion interfacial transport, but also ensures the overall structural integrity of the CoFe₂O₄/Gr electrode during cycling. The CoFe₂O₄/Gr electrode has high cycling performance (1109 mA h/g after 100 cycles at 0.2 A/g and 835 mA h/g after 200 cycles at 1 A/g) (Figure 7b) and rate capability (420 mA h/g at 5 A).

Wang et al. used a one-step HT method to create an α-Fe₂O₃ hollow nanotube (HNT)/Gr aerogel (GA) hybrid that enhances hematite characteristics by controlling the shape and mixing with 3D Gr [114]. There was no polymerization agent added to the Gr polymerization conditions. The α-HNTs/GA demonstrated remarkable electrochemical characteristics, with a specific capacity of ~1500 mAh/g and a consistent rate performance of 280 mAh/g at 5 A/g after 55 cycles. Furthermore, the high capacity retention, reversibility, and coulombic efficiency could be attributed to the synergistic effect of Gr sheets with hematite and the specific hollow nanotube morphology of Fe₂O₃, both of which buffered the large volume change caused by the insertion/extraction reaction of Li⁺ ions. Furthermore, the 3D porous framework structure of GA, with a large BET area and a high hematite load on GA, contributed to its high initial specific capacity. The electrochemical properties of α-HNTs/GA as an anode material were studied using CV at a scan rate of 0.2 mV/s for the first two cycles (Figure 7c). Figure 7d depicts the specific capacity obtained during galvanostatic discharge–charge testing with a current density of 100 mAh/g. During the first cycle, the specific discharged and charged capacities were ≈1973 and 1320 mAh/g, respectively, resulting in an initial coulombic efficiency (ICE) of around 67%. In subsequent cycles, the specific capacity and CE tended to stabilize and then increase. After 65 cycles, the specific capacity reached 1512 mAh/g discharged and 1406 mAh/g charged, with a CE level of 97% to 99% and a rising trend. Figure 7e depicts the material’s rate performance at varying current densities, culminating at 100 mA/g.

In contrast to conventional electrodes, a freestanding binder-free electrode film eliminates the usage of electrochemically inactive components such as conductive additives, and/or current collectors, which account for around 20% of the total battery weight. Zhou et al. used a simple HT reaction and subsequent annealing to create nanoporous CoO nanowire clusters on 3D porous graphene fabric (CoO-NW@GC) [116]. The self-supported Gr cloth has a large surface area, high porosity, and superior electric conductivity, all of which can contribute significantly to fast electron and ion transport. The CoO-NW@GC hybrid as an anode for LIBs achieves 1190 mAh/g and 429 mAh/g under current densities of 0.2 A/g and 3.2 A/g, respectively, due to the nanoporous CoO nanowire clusters uniformly depositing on the solid 3D skeleton of the GC substrate. The capacity was still 1100 mAh/g after more than 200 cycles at a current density of 0.5 A/g.

SnO₂/Gr nanocomposites were created using a chemical technique in which surface charge control leads echinoid-like SnO₂ nanoparticles to form and be uniformly decorated on Gr [97]. Under regulated pH, the electrostatic attraction between a Gr nanosheet (GNS) and echinoid-like SnO₂ particles results in a novel nanostructure with extremely small SnO₂ particles uniformly spread on the GNS. After 50 cycles, the anode preserved a reversible capacity of 634 mAh/g and a coulombic efficiency of 98%. The great reversibility can be attributed to the GNS’s mechanical buffering against the substantial volume change in SnO₂ during the delithiation and lithiation reactions. Additionally, the nanostructure considerably improves power capability by allowing easy electron transport across the GNS and rapid delithiation and lithiation processes within the echinoid-like nano SnO₂.

Using a combination of processes, including in situ metal tin reduction, spray drying, and thermal treatment, a SnO₂/rGO composite was fabricated [98]. Under moderate HT conditions, metal tin acts as a reducing agent for few-layered graphene oxide (GO) for generating restacked 2D rGO sheets and a metal source for homogeneously dispersed nanostructured Sn and Sn oxides (Sn/SnO/SnO₂). After heat treatment, a SnO₂/rGO composite is generated, in which Sn and SnO species are converted to SnO₂, and the micro-sized 3D porous rGO supporting framework effectively stabilizes the nanosized SnO₂. When compared to its 2D counterpart, the 3D SnO₂/rGO composite has a greater reversible capacity of 708 mAh/g under a current density of 500 mA/g over 150 cycles, as well as a long and stable cycling life under a high current density of 1000 mA/g.

Recent investigations show that element doping improves the electrical conductivity and chemical stability of SnO₂ materials [151,152,153]. SnO₂-based anodes with transition-metal doping have significantly improved electrochemical performance [154,155]. A hybrid material comprising cobalt-doped SnO₂ and Gr sheets (Co-SnO₂@GS) was created using a co-hydrolysis method to improve the electrochemical performance of SnO₂-based anodes [100]. In this hybrid structure, several Co-SnO₂ nanoparticles (4–7 nm) are uniformly positioned on the GS. Additionally, Co doping can improve the interface interactions between SnO₂ and GS. The Co-SnO₂@GS hybrid, which benefits from the synergistic effects of Gr hybridization and Co doping, has dramatically improved Li storage capability when compared to undoped SnO₂@GS and bare SnO₂. Figure 8a compares the cyclic performance of Co-SnO₂@GS, SnO₂@GS, and SnO₂ electrodes at 0.1 A/g. As demonstrated, the cyclic performance of Co-SnO₂@GS is clearly superior to that of undoped SnO₂@GS and bare SnO₂. Figure 8b examines the dealloying and conversion capabilities of SnO₂@GS and Co-SnO₂@GS. After 200 cycles, a steady capacity of ~828.5 mAh/g is still possible, with 83.2% retention from the sixth cycle (Figure 8c). Furthermore, this hybrid material has an excellent rate performance with a capacity of 674.7 mAh/g at 1 A/g.

A one-step HT synthesis approach was used to create SnO₂/NiFe₂O₄/graphene (SNG) nanocomposites for LIBs [103]. The nanocomposites exhibit good electrochemical performance due to the synergistic effect of Gr, NiFe₂O₄, and SnO₂ nanoparticles. The electrochemical performance improved as the SnO₂ level increased. After 50 cycles, a nanocomposite containing 50% SnO₂ has a high reversible capacity of 731.5 mA h/g at a current density of 200 mA/g, which can retain 80.9% of the initial cycle. After 100 cycles, the material has a comparatively high discharge capacity of 613 mA h/g at 800 mA/g. These increased electrochemical performances are related to the potential of NiFe₂O₄/SnO₂ buffer layers, which prevent electrode material pulverization during Li insertion and extraction. Furthermore, Gr sheets can enhance electrical conductivity and retard volume change to some level, allowing the structure to remain integrated and improving cycling stability and rate capability.

The particle size of SnO₂ plays a crucial role in controlling the reversibility of the conversion reaction from SnO₂ to Sn (SnO₂ + 4 Li⁺ + 4e⁻ ↔ Sn + 2Li₂O) for Li storage [156]. That is, the size of the SnO₂ particles is important to the anode material’s capacity. Nanosized SnO₂ can shorten electron and Li-ion diffusion paths, endure substantial volume fluctuations, and increase electrolyte–anode contact [157]. Several strategies for adjusting the particle size of SnO₂ have been proposed [158,159], including altering temperature, pressure, and additives. Metal–organic frameworks (MOFs), a new type of porous material, have sparked widespread interest in recent years [160]. MOFs, with their enormous specific surface areas and variable pore diameters, have shown great promise in gas adsorption and separation, catalysis, and drug delivery [161]. Furthermore, because MOFs are made up of metal ions and organic ligands, they are thought to be good precursors or templates for the development of TMOs/carbon hybrid materials [162].

Gao et al. developed the SnO₂@MOF/Gr composite anode, in which SnO₂ nanoparticles are packed into an Al-MOF. A wet impregnation technique traps SnO₂ nanoparticles inside the pores of Al-MOF. As a result, Al-MOF can tolerate SnO₂ volume variations during charge and discharge cycles and retain a stable SEI film. During the electrochemical alloying–dealloying process, Sn nanoparticles generated from SnO₂ tend to coalesce into bigger Sn clusters, resulting in capacity decline [163]. The Al-MOF layer in the SnO₂@MOF/Gr composite, on the other hand, will be capable of efficiently preventing Sn clustering. Due to the low electronic conductivity of SnO₂ and MOF, Gr sheets were used to surround the SnO₂@MOF composite to improve electronic conductivity. Furthermore, Gr has a reasonable theoretical capacity as an anode material, in addition to a large surface area and excellent mechanical flexibility [164]. After 1000 cycles, the SnO₂@MOF/Gr anode had a capacity of 450 mAh/g at a current density of 1000 mA/g, whereas the pure SnO₂ anode had essentially no capacity after 50 cycles (Figure 8d). Figure 8e depicts the rate capability of SnO₂@MOF/Gr at 500, 800, 1000, and 2000 mA/g, with high stable capacities of 565, 450, 385, and 324, respectively. After 100 cycles, the capacity recovers to 410 mAh/g as the current density drops to 500 mA/g.

ZnFe₂O₄ exhibits a high theoretical capacity (1000.5 mAh/g), making it a viable anode material for LIBs. However, rapid capacity fading and low rate capability pose practical challenges to its utilization [165,166]. ZnFe₂O₄@C nanoparticles with a core-shell structure were uniformly bonded to Gr nanosheet surfaces using a method inspired by mussels and calcination. Among the reported ZnFe₂O₄ anodes, the resulting ZnFe₂O₄@C/G nanocomposite exhibited one of the best Li storage properties, delivering a reversible capacity of 705 mA h/g at 0.25 C after 180 cycles (with a capacity retention of 99.4%) and an elevated rate capability of 403.5 mAh/g at 5 C. The excellent electrochemical performance is mostly owing to the conductivity, buffering, and shielding properties of carbon shells and Gr nanosheets on ZnFe₂O₄ nanoparticles.

Manganese monoxide (MnO) is a particularly appealing manganese oxide due to its high theoretical capacity (755 mAh/g), high density (5.43 g/cm³), comparatively low electrochemical motivation force (1.032 V vs. Li/Li⁺), low cost, and good natural abundance [167]. Despite these major characteristics, MnO is still limited by its low rate capability, which stems from weak conductivity and rapid capacity fading caused by large volume expansion and severe electrode collapse during the cycle operations [168]. Wang et al. used a simultaneous electrospinning and electrospraying method, followed by heat treatment, to create a CNFs/MnO/rGO film for high-performance LIB anodes [111]. The 2D reduced Gr sheets are dispersed on each of the 1D CNFs/MnO composite filaments to obtain freestanding and binder-free composite film electrodes, which not only improves the material’s electrical conductivity and flexibility but also increases the specific surface area, resulting in a significant increase in specific capacity. Because of the large elastic buffer region, it was revealed that the CNFs/MnO/rGO framework film could successfully handle volume changes in MnO particles. The composite electrode has a high discharge capacity of 1118 mAh/g at 0.1 A/g and then returns to 0.1 A/g after 80 cycles, with capacity retention of up to 98%; at high rates (e.g., 5 A/g), the composite maintains a high reversible capacity of 574 mAh/g after 3000 cycles with a substantial coulombic efficiency of 99%. Furthermore, this method offers a simple way of combining 1D fibers/2D nanosheets in a composite film with a controllable fiber diameter using simultaneous electrospinning and electrospraying methods.

The Kirkendall effect [169] (Section 4) and a simple etching technique were used to create a hollow Co₃O₄ (H-Co₃O₄) NP-embedded porous rGO (H-Co₃O₄/P-rGO) nanocomposite, which served as a model study for other TMOs [115]. Structure wise, one nanoparticle (NP) had one carved pore around it. It has outstanding reversible capacity (1016 mAh/g after 200 cycles at 0.2 A/g, Figure 9b), strong rate performance (810 mAh/g at 0.5 A/g, 510 mAh/g at 10 A/g), and good cycling endurance when used as an anode for LIBs. As a control sample, a Co₃O₄ NP-embedded porous reduced graphene oxide (Co₃O₄/P-rGO) composite was also made to highlight the benefits of hollow structured Co₃O₄ in H-Co₃O₄/P-RGO. Figure 9a compares the rate performance of H-Co₃O₄/rGO and Co₃O₄/P-rGO at rates ranging from 0.1 to 10.0 A/g. The discharge capacity of H-Co₃O₄/P-rGO is 840, 810, 760, 620, and 510 mA h/g at rates of 0.1, 0.5, 1.0, 5.0, and 10.0 A/g, respectively. Co₃O₄/P-rGO has discharge capacities of just 580, 530, 490, 360, 260, and 90 mAh/g at the same current densities.

Among the currently studied anode materials, ZnFe₂O₄ is a promising binary transition-metal oxide (TMO) anode material with a high theoretical capacity (1000 mAh/g) due to its combination of conversion and alloy Li-ion storage mechanisms, cost-effectiveness, abundance, eco-friendliness, and a low discharging voltage plateau of ~1.5 V [170]. Furthermore, the ZnFe₂O₄ crystal has an AB₂O₄ spinel structure, in which oxygen is cubically close-packed to establish a 3D framework for Li-ion diffusion. As a result, the binary ZnFe₂O₄ anode can exhibit increased electrochemical activity [171]. The quick reversible capacity fading throughout each cycle, induced by both pulverization caused by massive volume changes and the ZnFe₂O₄ crystals’ weak inherent electrical conductivity, is undesirable [172]. As a result, hybridizing nanostructured ZnFe₂O₄ with a conductive and flexible carbon supporting framework is one of the most effective ways.

Immobilizing nanosized electrochemical active materials with a 3D porous framework made of conductive Gr sheets improves Li-ion storage characteristics [117]. A simple approach is used to create a rGO-supported zinc ferrite (ZnFe₂O₄) composite anode material (ZnFe₂O₄/rGO). To begin, a redox reaction occurs between oxygen-containing functional groups on multilayered graphene oxide (GO) sheets and a regulated quantity of metallic Zn atoms. ZnO nanoparticles are nucleated in situ and grow directly on GO sheets. Second, numerous Fe atoms totally reduce the GO sheets, resulting in the formation of γ-Fe₂O₃ nanoparticles next to the ZnO nanoparticles. This stage involves building a 3D porous rGO supporting framework with γ-Fe₂O₃@ZnO nanoparticles efficiently contained between the rGO layers. Finally, the γ-Fe₂O₃@ZnO/rGO intermediate product is thermally treated to allow for a solid-state reaction, resulting in the ZnFe₂O₄/rGO composite. The ZnFe₂O₄/rGO composite has a high reversible capacity of 1022 mAh/g for 500 consecutive cycles at a high current rate of 1.0 A/g (Figure 9c).

Wang et al. presented a high-performance anode material made of tungsten-doped SnO₂/rGO composite synthesized using the one-pot HT technique [173]. Tungsten ions were doped inside the unit cells of tin dioxide rather than just sticking to its surface. Tungsten-doped SnO₂ was formed in situ on Gr sheets, creating a three-dimensional conductive network that improved electron transport and Li-ion diffusion. Tungsten-doped SnO₂ nanoparticles were evenly placed on a Gr sheet, preventing potential problems with aggregation and volume expansion. As a consequence, the nanocomposite electrodes comprising tungsten-doped SnO₂ and reduced graphene oxide demonstrated long-term cycling properties. After 100 cycles, the residual capacity remained as high as 1100 mAh/g at 0.1 A/g. It stayed at 776 mAh/g after 2000 cycles with a current density of 1 A/g.

Heteroatom-doped Gr can improve cycle performance, first cycle charge–discharge capacity, and Li storage capacity through facilitating Li-ion contact with active sites. Heteroatom doping (B, N, F, S, or P) increases electrochemical performance because Li-ion storage is affected not only by heteroatom contents, but also by synergistic coupling effects between heteroatoms [174,175,176]. Recent theoretical investigations have shown that doped Gr is more efficient at storing Li-ions due to increased defects in the Gr plane [177,178].

First-principles calculations were performed to investigate the effects of electron deficiency in N-doped Grs on their application in LIBs. Three different defect models, graphitic, pyridinic, and pyrrolic Grs, were used [177]. The pyridinic Gr was found to be the most suitable for Li storage with a high storage capacity, while the graphitic structure was the weakest of the three. It was discovered that the reversible capacity of the pyridinic structure can reach 1262 mAh/g, which is greater than the experimental 1043 mAh/g. Figure 10a shows a comparison of the most stable sites after Li atoms adsorbed on Gr sheets having different defects. Figure 10b depicts the hybridization peaks of the N and Li atoms in pyridinic and pyrrolic N-doped Gr. The binding energy between Li and pyridinic Gr is higher than for graphitic Gr. Figure 10c shows the avg. intercalation voltage of Li in the pyridinic, pyrrolic, and defect Grs, indicating the highest voltage when nine atoms of Li are embedded.

Graphite oxide (GO), tri-block co-polymer P123, and thiourea were used as the N and S sources in the HT technique and heat treatment to synthesize N and S co-doped mesoporous graphene (NSMG) [179]. XPS analysis of the NSMGs revealed the presence of active pyridinic-N, pyrrolic-N, graphitic-N, pyridinic N-oxide, thiophene, and -SO_x groups in the structure. Annealing temperature was used to adjust the N and S content. Super capacitors and LIBs showed enhanced electrolyte ion mobility and Li-ion diffusion, respectively; this could be related to structural characteristics such as numerous pores and defects, wrinkles, and N/S co-doping. NSMG treated at 600 °C had the maximum capacitance of 261 F/g at 0.5 A/g in supercapacitors, whereas NSMG treated at 800 °C had the best performance in LIBs with a discharge capacity of 460 mAh/g at 100 mA/g, good cycle stability (440 mAh/g), and higher rate capability. The high capacity of the NSMG (800 °C) is due to the stable and desired residual oxygen, such as the carbonyl group (C-O), which increases with heat treatment [180]. Precursor-assisted CVD was used to prepare hierarchically porous graphene-like microspheres (3D NS-GSs) doped with N and S [181]. By employing (NH₄)S₂O₈ as a N and S dopant, the 3D NS-GSs have a larger surface area than pure Gr-like microspheres. The combination of large surface area and proper heteroatom doping resulted in a high reversible capacity, good rate capability, and long cycling stability in 3D NS-GS as the LIB anode. Doped Gr has an electron-adopting propensity in pyrrolic and pyridinic structures, and their sites may attract more Li-ions while adjusting the electronic characteristics of neighboring carbon atoms over the cycles [177,182]. As a result, N- and S-co-doped Gr sheets of 3D NS-GSs can accept more charge than pristine Gr sheets.

The addition of F atoms to the Gr surface effectively adjusts its electrical structures and creates surface defects [183]. N and F co-doped graphene (NFG) with N and F concentrations of 3.24 and 10.9 at%, respectively, was synthesized using the HT reaction of trimethylamine tri(hydrofluoride) [(C₂H₅)₃N·3HF] with aqueous-dispersed graphene oxide (GO) as an anode material for LIBs [184]. N and F co-doping in Gr raised the framework’s disorder and defects, increased the interlayer space, wrinkled the nanosheets with multiple open-edge sites, and so improved Li-ion diffusion across the electrode compared to sole N or F-doped Gr. XPS study of NFG revealed the existence of active pyridine and pyrrolic type N, significantly higher electrically conductive graphitic N, and a semi-ionic C-F bond in the structure. The N and F doping content, as well as the component types of N and F functional groups, may be regulated by the high temperature. The NFG produced at 150 °C had a coulombic efficiency of 56.7% in the first cycle, reversible specific discharge capacity of 1075 mAh/g at 100 mA/g, rate capability of 305 mAh/g at 5 A/g, and excellent cycling stability (capacity retention of ∼95% at 5 A/g after 2000 cycles).

4. Three-Dimensional (3D) Carbon Nanostructured Composites

4.1. Carbon Nanospheres, Carbon-Coated Metal Oxide Nanospheres, Core-Shell Structures, and Doped Carbon-Coated Transition-Metal Oxides

Carbon nanospheres (CNSs) and carbon nanoporous (CNP) structures are also promising carbon nanostructures, alongside CNTs, nanofibers, nanorods, and Gr. As Li-ion battery anodes, CNSs and CNPs have excellent cycle stability and coulombic efficiency. If 3D carbon nanostructures are combined with proper electrolytes, they will produce a stable SEI. However, 3D carbon nanostructures alone have a poor capacity. To overcome this drawback, carbon nanospheres/nanoporous structures can be mixed with other materials to create composites. Carbon shells not only allow for volume expansion and contraction during charge storage, but they also assist electron transport. The combination of these two variables results in extended cycling stability and exceptional rate performance. Furthermore, the broad surface can improve the contact area between the electrode and the electrolyte, which is advantageous for the use of active materials. The porous and hollow construction results in a short diffusion length of ions in the electrode, which improves electrochemical performance. One of the key factors for C hollow sphere (CHS)-based anodes is to boost the anode’s areal and volumetric energy density by utilizing all CHS voids [185]. The carbon sphere’s porous structure is divided into three types: micropores (avg. pore size < 2 nm), mesopores (2–50 nm), and macropores (>50 nm).

To synthesis diverse hollow micro/nanospheres, self-templated methods have been devised, including those based on the nanoscale Kirkendall effect (Figure 11a) [186], Ostwald ripening (Figure 11b) [187], galvanic replacement (Figure 11c) [188], and surface-protected etching [189]. These methods are particularly appealing since the hollow shells are created through a chemical reaction, and the template can be depleted spontaneously during the hollowing process. Methods based on the nanoscale Kirkendall effect offer a variety of benefits: high-quality colloidal nanocrystals with consistent sizes and shapes are readily produced, and their reaction with appropriate materials, together with the nanoscale Kirkendall effect, can produce extremely crystalline hollow nanocrystals with a high yield even in the quantum domain [190]. Ostwald ripening is a phenomena that occurs in solid or liquid solutions when small nanoparticles dissolve and are redeposited on larger nanoparticles [78]. This phenomenon is caused by small particles having a higher average surface energy, which causes material to evaporate (or dissolve) more quickly. At a constant amount of material, Ostwald ripening induces the evolution of a population of nanoparticles, resulting in less and fewer big aggregates surviving for long periods of time. The main step in galvanic replacement reactions is the replacement reaction between a suspension of nanoscale metal templates and a salt precursor containing a less active metal. Figure 11c depicts the schematic formation of a nanostructure made up of an Au/Ag alloy core and an Au/Ag alloy shell. In step (A), a conformal coating of pure silver is electrolessly deposited on the surface of an Au/Ag alloy nanoparticle; and (B), the resulting particle is reacted with an aqueous HAuCl₄ solution to change the pure silver coating into a bigger Au/Ag alloy shell. For a detailed discussion on the templating methods of HCS, readers are urged to refer to an article by Liu et al. [185].

The following is a list of some carbon nanospheres (CNSs) and nanoporous composites that contain metalloids and post-transition metal: xarbon-Si-SiO₂-CNS [191], Si/CNS [192], SiO₂-CHS [193], mesoporous C-SnO₂-CNS [194], double-shell structured Si@SnO₂@C [195], carbon-encapsulated SnO₂ hollow spheres [196], Si-TiO₂-CNS [197], and carbon-coated Si/carbon nanotube/GO (C-Si/CNT/GO) [198]. Several studies also detailed the effective production of composites containing metal oxide and heteroatoms: hollow carbon spheres with encapsulation of Co₃O₄ nanoparticles (HCSE-Co₃O₄) [199], core-shell porous Fe₃O₄@C nanospheres [200,201], core-shell structured ZnO@C nanosphere [202], TiO₂/CNS [203], ball-in-ball ZnO/ZnFe₂O₄@carbon nanospheres [204], α-Fe₂O₃/CNS [205], porous Mn₃O₄/C nanospheres [206], Co₃O₄-doped hollow hierarchical porous carbon spheres [207], Cu₂O/CNS [208], nitrogen-doped carbon-coated TiO₂ (TiO₂/N-C) [209], and Mn₃O₄/N-doped porous carbon sphere (NPCS) [210]. The electrochemical performance of a few hollow carbon sphere-based materials for LIBs that have been reported in the recent years is listed in

Table 2. Electrochemical performance of some hollow carbon sphere-based materials for LIBs. Abbreviation: HCS = Hollow carbon sphere.

Material	Highest Reversible Capacity (mAh/g)	Current Density (A/g)	Reversible Capacity (mAh/g)	Current Density (A/g)	Cycles	Ref.
Tin embedded in nitrogen-doped carbon shells	917	0.2	600	0.2	250	[211]
Yolk shell Sn@C microspheres	829	1.0	691	1.0	500	[212]
Tin nanoparticles encapsulated in HCS	1250	0.2 C	650	0.2 C	100	[213]
Ultra-small Tin embedded in carbon shells	743	0.5	423	4.0	6000	[214]
Fe3O4/C open hollow spheres	917	0.09	587	1.0 C	30	[215]
Fe3O4 in macropores of carbon HCS	1092	0.2	645	2.0	1000	[216]
α-Fe2O3@C hollow spheres	1136	0.1 C	1400	0.1 C	50	[217]
Fe2O3 nanoparticles embedded in N-doped HCS	1573	0.1 C	1142	1.0 C	100	[218]
F-doped Fe3O4/C hollow sphere	930	0.1 C	620	1.0 C	200	[219]
Yolk–shell structured Fe3O4@HCS	1010	0.15	750	1.0	70	[220]
Fe2O3 nanoparticles anchored on nitrogen-doped HCS	1120	0.1	702	0.5	65	[221]
Ge encapsulated in HCS	1270	0.4 C	1020	0.4 C	100	[222]
Coaxial SnO2@C hollow spheres	520	0.32 C	475	0.8 C	100	[223]
NiSb embedded in carbon shell	667	0.05	497	0.1	100	[224]
Double-shelled SnO2/C spheres	1100	0.25 C	911	0.25 C	100	[225]
Yolk-shelled SnO2@C spheres	780	0.1	512	0.1	30	[226]
N-doped carbon/SnO2 hollow sphere	1003	0.16 C	910	0.16 C	100	[227]
SnO2/C hollow core–shell spheres	640	0.05	576	0.05	100	[228]
SnO2/C hollow nanospheres	983	0.1	473	0.1	50	[229]
SnO2/C hollow microspheres	700	0.2	550	0.5	100	[230]
SnO2/C/MnO2 hollow sphere	800	0.1	644	0.1	200	[231]
Nanosized SnO2 embed in HCS	627	0.2	501	0.2	100	[232]
SnO2/Fe2O3/C hollow sphere	1147	0.2	475	2.0	1000	[233]
Si nanocrystallites sealed in HCS	1570	0.25	1020	0.25	250	[234]
Mesoporous SiO2/carbon hollow sphere	624	0.1	560	3.0	180	[235]
Core-shelled SiO2@C hollow sphere	153	0.11	653	0.11	160	[193]
Carbon-coated CuO hollow sphere	670	1 C	750	1 C	300	[236]
CoO nanoparticles encapsulated in HCS	584	0.1	554	0.1	50	[237]
Co3O4 nanoparticles encapsulated in HCS	732	0.074	732	0.074	20	[199]
Triple-shelled MoO2/C hollow sphere	867	0.2	580	0.5	200	[238]
MnO2/HCS with 3D network	900	0.1	692	0.1	100	[239]
HCS/MnO2 nanosheets	947	0.1	826	0.5	140	[240]
Carbon-coated CuFe2O4 hollow sphere	815	0.1	550	0.1	100	[241]
ZnCo-oxide/C hollow sphere	992	1.0	823	4.0	1000	[242]
CeO2/C hollow sphere	400	1.0	313	1.0	500	[243]
MoS2 nanosheets coated HCS	1190	0.1	750	0.2	100	[244]
Bowl-like SiC encapsulated in HCS	1345	0.6	742	3	1000	[245]
C@TiO2@MoS2 hierarchical nanospheres	1030	0.2	993.2	0.2	100	[246]
WS2-decorated nitrogen-doped HCS	801	0.1	631	0.5	150	[247]
SeS2 impregnated in double-shelled HCS	1070	0.1 C	600	1 C	1000	[248]
Ni3S4/C hollow sphere	1003	0.1	480	1.0	500	[249]

.

The template approach and carbon thermal vapor deposition were used to create an interconnected sandwich-type composite of carbon/Si-SiO₂/carbon nanospheres [191]. The carbon conductive layer enhances the electronic conductivity of Si-based anodes and reduces volume expansion during discharge or charging. The interconnected sandwich carbon/Si-SiO₂/carbon nanospheres composite has strong battery characteristics, with a reversible capacity of 1094 mAh/g and a reversible capacity of 1050 mAh/g after 200 cycles, and a capacity retention of 96%. A simple method for creating SiO₂@C hollow spheres involves hydrolyzing tetraethyl orthosilicate (TEOS) to generate SiO₂ shells on organic spherical templates [193]. The shells are then calcined in air to remove the templates and carbon layers are used to cover them (Figure 11d). SiO₂@C hollow spheres with varying SiO₂ concentrations show a steady rise in specific capacity with discharge–charge cycling. The SiO₂@C with 67 wt% SiO₂ content has discharge–charge capacities of 653.4/649.6 mAh/g across 160 cycles at a current density of 0.11 mA/cm².

Figure 11. Hollow micro/nanosphere synthesis methods. (a) Schematic representation of the nanoscale Kirkendall phenomenon in the synthesis of hollow nanocrystals; J stands for diffuse flux [169]; (b) Ostwald ripening; (c) galvanic replacement. Illustration of the procedure for preparing a nanostructure consisting of Au/Ag alloy core and Au/Ag alloy shell [188]. Copyright © 2004, ACS. (d) Fabrication process of the SiO₂@C hollow spheres. SiO₂ shell is formed on the PAA (polyacrylic acid) sphere template, and then a C coating layer covers the shell after calcination of the template. Adapted with permission from ref. [193]. Copyright © 2017, The editorial office of Journal of Materials Science & Technology. (e–g): C@SnO₂@C composite. (e) fabrication steps of porous dual carbon shell-coated SnO₂ hollow nanospheres (C@SnO₂@C); (f) rate performance of the hollow SnO₂ and C@SnO₂@C electrodes at varying current density from 0.2 A/g to 3.2 A/g; (g) the long-term cycling performance of C@SnO₂@C at 5 A/g for 10000 cycles. Adapted with permission from ref. [194]. © 2019 Elsevier B.V.

Figure 11. Hollow micro/nanosphere synthesis methods. (a) Schematic representation of the nanoscale Kirkendall phenomenon in the synthesis of hollow nanocrystals; J stands for diffuse flux [169]; (b) Ostwald ripening; (c) galvanic replacement. Illustration of the procedure for preparing a nanostructure consisting of Au/Ag alloy core and Au/Ag alloy shell [188]. Copyright © 2004, ACS. (d) Fabrication process of the SiO₂@C hollow spheres. SiO₂ shell is formed on the PAA (polyacrylic acid) sphere template, and then a C coating layer covers the shell after calcination of the template. Adapted with permission from ref. [193]. Copyright © 2017, The editorial office of Journal of Materials Science & Technology. (e–g): C@SnO₂@C composite. (e) fabrication steps of porous dual carbon shell-coated SnO₂ hollow nanospheres (C@SnO₂@C); (f) rate performance of the hollow SnO₂ and C@SnO₂@C electrodes at varying current density from 0.2 A/g to 3.2 A/g; (g) the long-term cycling performance of C@SnO₂@C at 5 A/g for 10000 cycles. Adapted with permission from ref. [194]. © 2019 Elsevier B.V.

For carbon-coated SnO₂ nanocolloids [250] and SnO₂/carbon hybrid hollow spheres [251], capacity deterioration occurs due to the aggregation of SnO₂ nanoparticles after repeated cycling. Furthermore, the SEI layer generated atop the typical core-shell structure is unstable because it cannot resist massive volume changes over long periods of time [252]. The dual carbon shell-coated SnO₂ hollow nanospheres (C@SnO₂@C) were created for stable LIBs by combining the benefits of porous carbon and hollow structured SnO₂ [194]. The double carbon shells served as a protective layer to stop the aggregation of nanosized SnO₂ and structural damage, in addition to acting as a conducting medium (Figure 11e). The C@SnO₂@C hollow nanosphere anode showed noticeably improved cycle stability and rate performance when compared to its counterparts, the pure SnO₂ hollow nanospheres and the carbon-coated SnO₂ nanospheres [250,251]. With the aid of a dually coated structure-based space-bound protection approach, the C@ SnO₂@C hollow nanospheres demonstrated an initial specific capacity of 1483.9 mAh/g at 200 mA/g and a reversible capacity of 713.8 mAh/g after 300 cycles, corresponding to a capacity fading of 0.071% per cycle from the 2nd to the 300th cycle. The rate performance of the C@SnO₂@C and pure SnO₂ electrodes (Figure 11f) shows that even at a low current density of 0.2 A/g, SnO₂ capacity rapidly decays, with almost no capacity remaining at 1.6 A/g and 3.2 A/g. Conversely, the C@SnO₂@C electrode outperforms pure SnO₂ hollow nanospheres in terms of rate. When the tested current density is reset to 0.2 A/g, the SnO₂ electrode has a negligible capacity of 62.3 mAh/g after 100 cycles, whereas the dually coated C@SnO₂@C electrode can recover a high reversible capacity of 675 mAh/g while remaining completely stable in subsequent cycles. The high rate capacity of the C@SnO₂@C composite is most likely attributable to its spatial double protection nanostructure and refined SnO₂ nanocrystals. The uniformly porous carbon layers can act as a highly conductive medium. Figure 11g displays C@SnO₂@C’s long-term cycling performance at 5 A/g over 10,000 cycles.

Carbon-encapsulated SnO₂ hollow spheres for LIBs were created using a self-templating process from Sn spheres [196]. High-temperature annealing (for example, 700 °C) invariably resulted in the production of Sn on the outer surface, which was detrimental to long-term cycling. In contrast, annealing at an ideal temperature of 600 °C followed by acid etching totally encased the SnO₂ within a carbon shell while maintaining a stable hollow structure (Figure 12a). After 120 cycles, the as-prepared carbon-encapsulated SnO₂ hollow spheres (SnO₂@C-6H) demonstrated a reversible capacity of 898 mAh/g at 200 mA/g (Figure 12b) and a promising rate capacity of 549.8 mAh/g at a high current density of ~2000 mA/g (Figure 12c). The observed excellent outcomes might be related to the development of a unique hollow structure with an increased specific surface area of 215.18 m²/g and a reasonably high degree of graphitization of the carbon shell (i.e., Raman intensity ratio of D to G band, I_D/I_G = 0.84). The reason for this is that the acquired hollow spheres not only provide a barrier effect for volume expansion and SnO₂ aggregation, but also increase Li storage [196].

Many metallic oxide coatings on Si particles have been investigated to increase electrochemical performance, including Co₃O₄, MnO, and NiO [253,254,255]. However, due to their inherent shortcomings, most metallic oxides frequently experience large volume variations and poor structural stability [106]. In this sense, TiO₂ coating, with its simple synthesis procedure and strong cycling stability, stands out among the metallic oxide possibilities listed above and receives a lot of attention. TiO₂ particles have a fast (de)lithiation process with negligible volume expansion (<4%) and a high operating voltage from 1 to 3 V (vs. Li/Li⁺), resulting in high electroconductivity and restrained exothermic reactions. This ensures an attractive and reliable battery by preventing the generation of dendritic Li and SEI [256]. One-step sol–gel coating and magnesiothermic reduction techniques were used to create the dual stabilized core-shell-shell Si-based composites with hollow spherical morphologies (Figure 13a,b) [197]. The Si@TiO₂@C nanospheres, as synthesized, have a high specific surface area of 280.9 m²/g, a homogeneous carbon layer of ~3 nm thick, a TiO₂ interlayer of about 6 nm, and a hollow structured Si core with a shell of 124 nm. For this designed architecture, the hollow Si@TiO₂@C nanospheres achieve an initial discharge specific capacity of 2557.1 mAh/g at 0.2 A/g and a reversible steady capacity of 1262 mAh/g after 250 cycles at 1 A/g. Furthermore, after 400 cycles, the Si@TiO₂@C nanospheres maintain a reversible discharge capacity of 1058.3 mAh/g at 4 A/g (Figure 13c). When evaluated at greater current densities of 0.5 and 8 A/g, the Si@TiO₂@C nanospheres maintained discharge capacities of 2457.2 and 718.2 mAh/g (Figure 13d). The two shells, a rigid TiO₂ interlayer and a conductive outer carbon layer, can drive expansion into the cavity rather than the exterior, acting as a compact fence to keep the central Si core from breaking and pulverizing. They also provide numerous conductive pathways for rapid Li⁺ transport and electron transfer, improving structural stability and electrical conductivity.

Carbon coating is a highly effective approach for improving the electrochemical performance of nanostructured TMOs. A scalable approach for fabricating hierarchical Fe₃O₄@C nanospheres has been developed [201]. The method of production consists of a one-step solvothermal synthesis of the Fe₂O₃/MIL-100(Fe) precursor coupled by an annealing treatment in Ar. The Fe₃O₄@C nanospheres (avg. size of ~200 nm) are composed of several interconnected nanoparticles with an amorphous carbon layer on the surface of Fe₃O₄, leading to a distinct hierarchical nanostructure with a greater specific surface area of 176.5 m²/g. The comparison of rate capabilities of the hierarchical Fe₃O₄@C and hollow Fe₂O₃ electrodes (Figure 14a) reveals that the Fe₃O₄@C electrode provides high reversible specific capacities of 852.2, 736.5, 529.8, and 362.1 mAh/g at current densities of 200, 400, 800, and 1600 mA/g. In comparison, the hollow Fe₂O₃ electrode has lower specific capacities than the Fe₃O₄@C electrode at each rate studied. The hierarchical Fe₃O₄@C nanosphere electrode achieves a high capacity of 757.9 mAh/g after 500 cycles at a current density of 500 mA/g, indicating better high-rate Li storage performance (Figure 14a). This might be explained by the distinct hierarchical nanoarchitecture, increased specific surface area, and homogenous in situ carbon layer produced by MIL-100(Fe).

As already mentioned, in Si-based anode materials, the considerable volumetric variability of Si has resulted in severe chalking and quick capacity loss. To tackle these problems, a carbon-coated Si matrix composite (Si@GC/PAC (PhyA-derived amorphous carbons)) with a 3D conductive network structure was created utilizing glucose and phytic acid [257]. In this case, glucose carbonization forms an indeterminate carbon coating that uniformly surrounds the Si nanoparticles, creating a distinct core-shell structure, while the controlled space between the Si nanoparticles and the carbon shell allows the Si nanoparticles to expand during the charge–discharge processes. In addition, the pyrolytic carbon formed by the PhyA’s heat degradation wraps the outer layer, leading to a large surface area and a highly 3D conductive network. The arrangement offers various advantages: the uniform amorphous carbon coating can improve the electrical conductivity and mechanical stability of the Si anode. The carbon skeleton formed by PhyA can also handle the volume variation of the Si anode. Pyrolytic carbon, together with glucose’s carbon coating, has a double limiting impact on volume expansion. The produced anode material, as modified by the two carbon sources, exhibits better structural and property synergy. The suitable doping of the glucose-based carbon matrix with phosphorus elements in the phytate can increase the interaction between Li-ions and the carbon coating. The developed Si@GC/PAC anode for Li-ion batteries has a reversible capacity of 1612 mAh/g at 0.1 A/g and retains 600 mAh/g after 200 cycles. Figure 14d shows the rates of bare Si, Si@GC, and Si@GC/PAC. The cycle stability of bare Si, Si@GC, and Si@GC/PAC was tested at 0.1 A/g, as shown in Figure 14e. After 200 cycles, the reversible specific capacity of the Si@GC/PAC composite remained at 600 mAh/g (56% of its initial capacity). The capacity of the bare Si and Si@GC electrodes was lowered to 652 and 766 mAh/g, respectively, with capacity retention rates of 20% and 27%.

Since α-Fe₂O₃ is abundant, nontoxic and inexpensive, and it has a larger theoretical capacity (1007 mAh/g), it is thought to be one of the most promising anode materials for LIBs with ultra-long cycle performance [206,207]. However, the significant volume expansion during the lithiation process (~96%) and poor electronic conductivity are the key issues that prevent the practical implementation of α-Fe₂O₃ anodes [258]. The hollow α-Fe₂O₃ with spherical morphology has been synthesized using the citric acid-assisted HT synthetic method, and the core-shell α-Fe₂O₃@C may be formed after carbon coating via the annealing process [205]. The results demonstrate that the α-Fe₂O₃@C hollow spheres have a high reversible Li storage capacity as well as good cycling stability, with an initial discharge–charge specific capacity of 1377 mAh/g and 723 mAh/g at 1C, and a discharge specific capacity of 793.2 mAh/g after 600 cycles. Further, better rate performance is achieved: 997.8 mAh/g at 0.1 C, 943.8 mAh/g at 0.2 C, 829.2 mAh/g at 0.5 C, 764.8 mAh/g at 1 C, and 697.6 mAh/g at 2 C, with a capacity of 966.3 mAh/g recoverable when the current density returns to 0.1 C. The hollow structure of α-Fe₂O₃@C not only reduces volume change during cycling, but it also facilitates Li-ion and electron transport and helps to stabilize the SEI layer.

Because of its abundance, low oxidation potential, and competitive electrochemical performance, Mn₃O₄ has been thoroughly explored as one of the most promising anode materials [259]. However, a number of difficulties prevent the use of TMOs as anode materials in LIBs. It has been widely acknowledged that nano-engineering and carbon hybridization are effective methods to overcome or reduce these difficulties [260]. A solvothermal reaction was used to create a self-assembled manganese (Mn)-based metal organic complex (Mn-MOC) with a spherical structure [206]. Thermal annealing was used to transform the Mn-MOC precursor materials to hierarchical porous Mn₃O₄/C nanospheres. As an anode in LIBs, these nanospheres demonstrated high reversible specific capacity (1237 mAh/g at 200 mA/g), good ratability (425 mAh/g at 4 A/g), and extraordinarily extended cycle life (no substantial capacity decline after 3000 cycles at 4 A/g). The higher capacity of the Mn₃O₄/C nanospheres to store Li was facilitated by their unique porous hierarchical structure, which is composed of uniformly distributed Mn₃O₄ nanocrystals with thin carbon shells.

Considering the combined benefits of carbon and TMOs, Yu et al. used spray pyrolysis with cobalt nitrate and glucose as precursors to produce Co₃O₄-doped hollow hierarchical porous carbon spheres [207]. The wall thickness of hollow porous spheres can be altered by varying the cobalt nitrate–glucose ratio. The electrochemical test results show that the product has a discharge capacity of 1297.1 mAh/g after 200 cycles at 0.2 A/g when 6.8 is the molar ratio of glucose to cobalt nitrate. The sample has a discharge specific capacity of 841.0 mAh/g after 300 cycles at 1 A/g. The hollow hierarchical porous structure effectively reduces the path of Li-ion transport, and the appropriate wall thickness can accommodate volume changes during charge and discharge. Figure 14c clearly shows that the sample’s charge and discharge ability is stable, with a coulombic efficiency of >98%. Trukawka et al. accomplished completely reversible conversion in hollow carbon spheres (HCSs) functionalized with copper oxide (Cu₂O) when tested as an anode material for LIBs [208]. HCSs were created using a hard template and glucose as a carbon precursor. Cu₂O was created by a thermally decomposing copper precursor. The uniform diameter of the spherical structure was ~160 nm and the shell thickness was ~10 nm. Cu₂O nanoparticles (~25 nm in diameter) are uniformly dispersed inside HCSs. The carbon structure between Cu₂O nanoparticles buffers the volume change and prevents Cu₂O nanoparticle aggregation. It also provides open channels for electron transport and Li⁺ diffusion during charge and discharge processes. When compared to HCSs, HCSs with Cu₂O nanoparticles give an enhanced high specific capacity of 682 mAh/g at a current density of 50 mA/g and highly stable cycling capabilities even at higher current rates when used as anode material for LIBs.

Nitrogen-doped carbon-coated TiO₂ (TiO₂/N-C) demonstrated a reversible specific capacity of 277 mAh/g after 220 cycles at a current density of 0.6 C as an anode material for LIBs [209]. An acceptable rate capability was also obtained at various current densities. The elevated performance of the TiO₂/N-C was caused by the N-doped carbon and assembly structure, which might facilitate electrical conductivity and buffer volume expansion during the charge and discharge process. Furthermore, small nanoparticles can shorten the Li-ion transportation path, and the gaps formed between connected small nanoparticles can accommodate volume change.

4.2. Porous Carbon-Based Anodes

Due to their high electronic conductivity, large surface area, tunable pore size and structure properties, chemical stability, and low cost, porous carbon materials are widely used in electrochemical energy storage, including LIBs, metal–air batteries, hydrogen-oxygen fuel cells, and supercapacitors (Figure 15a). Furthermore, porous carbons can be easily blended with active materials to create composites via sol–gel [261], ball-milling [262], grinding [263], melt-diffusion [264], sonochemical [265], and wet impregnation [266] methods. Vulcan XC-72 [267] and carbon black (super P) [268] are examples of porous conductive carbons. Industrially produced Vulcan XC-72R is known to have a surface area of ∼250 m²/g and high electric conductivity of ∼2.77 S/cm [269]. Porous carbon allows for volume expansion while also providing rapid Li-ion diffusion and electron transportation paths through the 3D porous structure’s channels (Figure 15b) [270]. In LIB, using solely porous carbon as an electrode results in mainly irreversible capacity loss during the first cycle due to electrolyte decomposition and SEI development on the large surface area.

Template methods [272] and activation processes [273] are the most common methods used to fabricate porous carbons. The template approach, comprising a hard-template process and a soft-template process, has been used to create many types of porous carbon materials. The activation technique is typically used to produce microporous activated carbons.

Adding metal oxides to porous carbon anode material integrates well to enhance electrochemical performance of porous carbon anode materials: V₂O₃ [274], ZnO [275], MoO₂ [276], α-Fe₂O₃ [277], TiO₂ [278,279], SnO₂ [280], V₂O₃ [281], NiO [282,283], and Co₃O₄ [284]. Through the multiple-electron transfer procedure during the conversion reaction, they can address the enormous irreversible capacity issue [270].

Figure 16. SnO₂, CMK-3, and SnO₂/CMK-3 electrodes. (a) Synthesis steps for SnO₂/CMK-3; (b) rate performances at different current densities; (c) cycle performances at 500 mA/g [280]. (d,e) Fe₃O₄@N-HPCNs. (d) Synthetic procedure of Fe₃O₄@N-HPCNs; (e) microporous carbon shell on Fe₃O₄@N-HPCNs [285]. (f,g) Three-dimensional porous FeSe₂/C composite: (f) illustration of synthesis of 3D Porous FeSe₂/C composite; (g) cycling performance and coulombic efficiency of 3D porous FeSe₂/C composite under the current density of 100 mA/g with 0.5 g PVP and the bare FeSe₂ for LIBs [286].

Figure 16. SnO₂, CMK-3, and SnO₂/CMK-3 electrodes. (a) Synthesis steps for SnO₂/CMK-3; (b) rate performances at different current densities; (c) cycle performances at 500 mA/g [280]. (d,e) Fe₃O₄@N-HPCNs. (d) Synthetic procedure of Fe₃O₄@N-HPCNs; (e) microporous carbon shell on Fe₃O₄@N-HPCNs [285]. (f,g) Three-dimensional porous FeSe₂/C composite: (f) illustration of synthesis of 3D Porous FeSe₂/C composite; (g) cycling performance and coulombic efficiency of 3D porous FeSe₂/C composite under the current density of 100 mA/g with 0.5 g PVP and the bare FeSe₂ for LIBs [286].

Li et al. produced SnO₂ nanodots (avg. size of ~5 nm) within the channels of the commercial ordered mesoporous carbon (CMK-3) [280]. The Sn precursor is prefilled under vacuum and easily transformed into SnO₂ by heating at 350 °C in air (Figure 16a). The SnO₂@CMK-3 composite demonstrates a discharge capacity of 898.8 mAh/g at the 150th cycle at 500 mA/g, surpassing that of the bare SnO₂ and CMK-3 electrodes (Figure 14b,c). This elevated electrochemical performance can be attributed to the high conductivity and large pore volume of the CMK-3 carbon matrices, which effectively reduce volume change-induced pulverization and increase electrode conductivity. Three-dimensional Si@TiO₂@C nanohybrids prepared using a three-step synthesis strategy outperformed Si@C and C (derived from polyaniline) electrodes in terms of electrochemical performance [287]. This was due to the existence of a robust TiO₂ shell and a polyaniline-derived porous carbon framework that can withstand volume changes of Si through lithiation and promote the overall electrode’s electric conductivity. The homogeneous dispersion of Si nanoparticles in Si@TiO₂@C allowed for larger Li storage sites. The Si@TiO₂@C hybrid electrode has an initial discharge capacity of 1748.6 mAh/g at 100 mA/g, and a steady reversible capacity of 1112.5 mAh/g at 100 mA/g following 500 cycles, having a coulombic efficiency of 98.8%. The electrode demonstrated a high reversible capacity of 508.9 mAh/g at 500 mA/g during the 500th cycle, with a coulombic efficiency of 99.1%.

Fe₃O₄-based materials are regarded as a viable anode choice due to their nontoxicity, theoretical capacity of 924 mAh/g), and low cost [288,289,290]. However, pure Fe₃O₄ exhibits low conductivity and a substantial volume expansion problem (~200%) during lithiation, leading to poor rate performance and a rapid capacity decline [291]. As previously stated, N-doping is thought to be an effective approach for increasing the Li storage characteristics of carbonaceous materials. A soft-templating method was used to create Fe₃O₄ nanoparticle-embedded and N-doped hierarchically porous carbon nanospheres (Fe₃O₄@N-HPCNs) (Figure 16d) [285]. The nanospheres have a core-shell, hierarchical porous structure with dual mesoporous cores and microporous carbon shells (~30 nm thick, Figure 16e). The Fe₃O₄@N-HPCNs anode has reversible specific capacities: 1240 mAh/g (after 100 cycles at 0.1 A/g) and 581 mAh/g (after 400 cycles at 1 A/g). Even at 10 A/g, the specific capacity remains around 290 mAh/g.

Xu et al. synthesized N-doped double-shelled hollow mesoporous carbon spheres (N-DHMCSs) using a simple soft-template technique involving self-assembly of amphiphilic anionic block copolymer polystyrene-b-poly(acrylic acid) (PS-b-PAA) micelles as the soft template, TEOS as the Si source, and dopamine (DA) as carbon and N sources [292]. After carbonizing the DA precursors and removing Si, the resulting N-DHMCSs had a large specific surface area (>600 m²/g), tunable N doping content (4.02–5.91% in weight ratios), and double-shell hollow structures. The double-shelled carbon structures, hierarchical porous architecture, and appropriate N doping level not only reduce the diffusion channel length and conductivity, but also enhance active sites. When the current density is increased progressively from 0.1 A/g to 5 A/g, the N-DHMCSs (carbonized at 800 °C) had good rate capability. After 100 cycles (0.1 A/g), the same sample had a high discharge specific capacity of 920.3 mAh/g. Pappas et al. studied the performance of carbon nanospheres (CNSs) doped with three heteroatoms (N, P, and S) as LIB anode materials. The CNSs were formed by carbonizing strongly crosslinked organo (phosphazene) nanospheres (OPZs) with a diameter of 300 nm [293]. The OPZs were produced via a single polycondensation process involving hexachlorocyclotriphosphazene (HCCP) and 4,4′-sulphonyldiphenol. The CNS anode material contained micro- and mesopores, had a high specific surface area of 875 m²/g, and demonstrated good stability and coulombic efficiency throughout a lengthy charge–discharge cycle of up to 1000 cycles at a 1C rate, generating approximately 130 mAh/g.

Melamine-formaldehyde resins are another commonly available and inexpensive potential carbon source [294,295]. To increase the graphitic N content, Sui et al. employed melamine-formaldehyde resin as both a crosslinking agent and a N source, and they created N-doped porous Gr materials via freeze-drying and thermally annealing a Gr/melamine-formaldehyde hydrogel [296]. The end product had a discharge capacity of 672 mAh/g when measured at a current density of 100 mA/g. Likewise, Si nanoparticles (Si NPs) were self-incorporated into melamine-formaldehyde resin microspheres using an HT approach. The precursors were then annealed to form hollow N-doped carbon microspheres (hNC MSs) with 10.8 wt% Si NPs [297]. Melamine was immediately polymerized with formaldehyde, giving N and carbon sources. After 360 cycles, the Si@hNC MS LIB anode maintained a reversible capacity of 872 mAh/g (at 200 mA/g). The hNC MSs significantly reduced the Si volumetric variations caused by repetitive charging and discharging, resulting in a high electrochemical stability of the Si anode [297].

A binder-free anode made of disordered carbon matrix, comprising meso- and macropores and nanosized sp²-clusters (avg. size ~15–20 nm) prepared via the virtual cathode deposition technique showed specific capacity of ~1250 mAh/g for the first cycle and capacity of >900 mAh/g at a 0.1 C rate, and >600 mAh/g at a 0.5 A/g rate during cycling [298]. A coulombic efficiency > 99.5% was observed for 500 cycles. Further, the anode yielded areal capacity (~4.5 mAh/cm²) and volumetric (>1400 Ah/L) capacities. The surface area was ~2100 m²/g. The specific carbon matrix microstructure brought about by the process-induced polymorphism, together with its beneficial mechanical and electrical qualities, was the reason for the enhanced Li⁺ intercalation.

Recently, metal selenides have been tested for their Li/Na storage properties because of their relatively high theoretical capacities and appropriate working potentials. These include MoSe₂ [299], FeSe_x [300], Sb₂Se₃ [301], CoSe_x [302], Cu₂Se [303], and NiSe₂ [304], Iron selenide has gained increased attention due to its extensive Fe resources, superior chemical stability, and high theoretical capacity [305]. However, considerable volume changes occur throughout the cycling process, resulting in structural degradation, and the creation of unstable SEI limits its widespread application. Nanocompositing iron selenides with carbon structures is a good option.

A 3D porous carbon framework embedded with FeSe₂ nanoparticles (3D porous FeSe₂/C composite) produced via a simple calcined method, following a selenized method without a template, shows reversible capacities of 798.4 and 455.0 mAh/g for LIBs and sodium-ion batteries, respectively, at 100 mA/g after 100 cycles (Figure 16f,g) [286]. The composite has better cycle stability than bare FeSe₂ nanoparticles. The uniformity of FeSe₂ nanoparticles, the 3D porous structure’s accommodative volume stress upon cycling and shortening electron/ion transport paths, and increased conductivity due to carbon as a buffer matrix are the factors for enhanced electrochemical performance.

5. Summary, Challenges, and Outlook

This review examined the common LIB anode materials, including their conducting mechanisms, morphological characteristics, synthesis techniques, and energy storage capabilities. To improve the energy densities of LIBs, nanocarbon-based hybrids can be synthesized to harness the synergistic properties of both nanocarbons and high Li storage capacity materials. Certainly, such composite anode materials can greatly improve battery performance when compared to a single constituent.

In terms of dimensions, the 1D structure has a low density, strong mechanical strength, and outstanding conductivity, making it suitable for use as a conductive network and inhibiting volume expansion for LIB electrodes. Nevertheless, CNT-based anodes cannot be used as a replacement for graphite-based anodes in LIB anode applications because of their highly irreversible capacity and minor improvements in Li storage capacity and cyclability.

The number of defects, lengths, and diameters of CNTs determine the performance of CNT-based anode materials. The chemical etching and ball-milling treatments have been shown to cause structural modifications and the development of surface functional groups on CNTs. The structural alterations, such as lateral defects on the surface of CNTs and shortening of CNT lengths, can boost Li insertion capability. However, both chemical etching and mechanical ball-milling processes typically result in the degradation of graphitic structure, uneven size reduction with lengthy CNTs remaining, and a considerable loss of raw materials. As a result, new technologies must be devised to produce short CNTs with narrow length and breadth distributions and minimal material loss. The actual nature of CNTs, including the number and types of defects, CNT diameters, and lengths that aid in enhancing the anode performance, must be determined. The application of first-principles calculations to determine the electronic properties of CNT can help us in deciding the maximum intercalation density, information on where the intercalated ions sit, and whether the intercalation modifies the structure of the nanotube. The outcome of such calculations can help us choose the promising candidate material for anodes in battery applications without undergoing labor-intensive experimental etiquette. The simplification of nanocarbon synthesis techniques is a necessity for the production of large-scale, low-cost end products. Fabrication of a free-standing thick CNT paper for LIB anodes is critical since it allows for the design of LIB without a current collector and reduces the weight of the power source. Furthermore, the CNT paper’s ease of handling allows it to be easily shaped into the various shapes required in a wide range of flexible and lightweight electronic devices. CNTs can be coupled with conducting polymers and high-capacity anode materials like transition oxide, Si, and Ge nanoparticles to create functional inks that can be used to print flexible paper-like electrodes using different printing technologies.

The 2D carbon material has strong physiochemistry, a huge surface area, and good electrical conductivity. It can be employed as a conductive layered structure and volume expansion restricting agent in the LIB anode. Gr offers excellent electronic conductivity, mechanical flexibility, and chemical activity. As a result, it can be used as an appropriate 2D support for assembling nanoparticles with diverse shapes, and excellent results have been produced using Gr–metal oxide nanoparticle composites. Furthermore, Gr can inhibit nanoparticle agglomeration during cycling, and nanoparticles can prevent Gr sheets from stacking again. The effect of defects in Gr on Li adsorption requires further investigation. In such efforts, computational tools can be employed. Graphene functionalized with heteroatoms and different functional groups can attract Li-ions and improve their energy storage capacity. Multidimensional structures on highly conductive carbon nanomaterials with abundant electroactive sites and high SSA contribute to LIBs’ excellent capacitance and cycle performance.

Hybrid materials comprised of 3D carbon nanostructure and transition-metal oxides, carbonate, sulfides, Si, Ge, etc., for LIB anodes have particular characteristics: the porous hierarchical hollow designs enable substantially shorter transport channels for electrolyte ions and electrons, resulting in faster electrode kinetics. The hollow carbon sphere frameworks provide increased electrical conductivity and structural integrity, resulting in accelerated performance and long cycling stability. Three-dimensional structures, particularly porous, hollow, and core-shell spheres, can withstand significant volumetric fluctuations during the charge–discharge process but reduce the volumetric capacity of the whole structure. Despite notable advances, the difficulties connected with their synthesis, dimensional control, and large electrode–electrolyte surface area, which produces the high irreversible capacity, have hampered their widespread use in LIBs. Computational and experimental research is needed to understand the interrelationships between the structural and chemical properties of 3D nanocarbon-based anodes and LIB performance. Density functional theory-based first-principles calculations can be used to predict de-lithiated and lithiated phases, charge distribution, phase stability and transition, and Li-ion diffusion kinetics. The self-templating method, together with Ostwald ripening and the Kirkendall effect, offers a novel approach to creating nanohybrids from hollow carbon spheres. The fabrication of porous carbon materials using metal–organic frameworks as templates and biomass as a sustainable carbon precursor is highly promising.

Small-scale production of high-performance anode materials might be simple in research labs. Real commercial-scale synthesis, however, can be quite challenging and needs complex techniques. The HT method under different temperatures and duration followed by calcination (annealing) is a widely employed technique to synthesize nanocarbon hybrids. However, the HT technique needs very stringent reaction conditions. Such strategies demand preliminary investigations to comprehend the suitable HT reaction parameters. The modified Hummers process is the most popular way used to prepare graphene. However, it involves numerous processes, employs hazardous reactants, and produces heavy metal ions and other hazardous materials. Mechanical exfoliation and chemical vapor deposition, on the other hand, have limited yields and are expensive to undertake. As a result, novel strategies are required to produce large-area graphene in a simple, environmentally friendly, and cost-effective manner on a large scale.

The low ICE of intercalation-type active anode materials is due to the creation of the SEI layer and the irreversible nature of Li adsorption. The poor ICE of alloying-type materials is associated with the formation of the SEI layer, the imperfect reversible reaction caused by the coarsening of active materials, and the irreversible breakdown of certain oxides. As a result, SEI formation continues to be a significant concern. The material qualities and composition of the SEI layer therefore require additional investigation.

MXenes and Xenes, two emerging 2D materials, deserve further attention as anodes due to their high conductivity, adjustable layer distance, and highly exposed active sites. Similarly, transition-metal dichalcogenides have a graphene-like layered structure, and a single layer can be made via the stripping process. This arrangement can aid in the transport of Li ions, and the loosely layered structure can also reduce volume expansion during charging and discharge. It is worth noting that most 2D layered materials may be produced via the mechanical stripping method, although this method is inefficient and cannot produce large-area materials.

The cycling characterization must be performed at a high cycle number since Li-ion batteries are intended for long-term use. Microwave-assisted synthesis and modification of nanocarbons are fast and energy-efficient. Since nanocarbons are good microwave absorbers, this property can be used to functionalize various types of nanocarbons.