Electrolyte engineering and material modification for graphite‐based lithium‐ion batteries operated at low temperature

Graphite offers several advantages as an anode material, including its low cost, high theoretical capacity, extended lifespan, and low Li+‐intercalation potential. However, the performance of graphite‐based lithium‐ion batteries (LIBs) is limited at low temperatures due to several critical challenges, such as the decreased ionic conductivity of liquid electrolyte, sluggish Li+ desolvation process, poor Li+ diffusivity across the interphase layer and bulk graphite materials. Various approaches have therefore been explored to address these challenges. On the basis of graphite anode and corresponding LIBs, this review herein offers a comprehensive analysis of the latest advances in electrolyte engineering and electrode modification. First, electrolyte engineering is discussed in detail, highlighting the design of new electrolyte formula with broad liquid temperature range, optimized solvation structure, and well‐performed inorganic‐rich solid electrolyte interface. The advances in material modification have been then depicted with the view of improving the solid bulk diffusion rate to show general strategies with excellent performance at low temperatures. Finally, the corresponding challenges and opportunities have also been outlined to shed light on viable strategies for developing efficient and reliable graphite anode and graphite‐based LIBs under low‐temperature scenarios.


| INTRODUCTION
Lithium-ion batteries (LIBs) have been widely used in consumer markets and electric vehicles owing to their light weight, high energy density, and long cycle life. As an important component, graphite is a popular anode owing to its relatively economical cost, considerable theoretical capacity (372 mAh g −1 ), good electronic and ionic conductivity, and outstanding extended lifespan, which makes it an ideal choice for high-performance LIBs. [1][2][3][4][5][6] Moreover, the graphite anode shows a significant advantage of low Li + -intercalation potential, typically around 0.1 V versus Li + /Li, which enables the battery to achieve a relatively high output voltage and high energy density. However, the low lithiation potential of graphite really acts as the double-edge sword because severe potential drop at low temperature would inevitably cause capacity loss and even lithium deposition, leaving the graphite-based LIBs suffering from energy decay and even being nonrechargeable. [7][8][9] This can be clearly illustrated in Figure 1, where lithium plating will take place on the graphite surface due to the lower nucleation barrier and the large polarization at low-temperature conditions, instead of the intercalation of Li + into layered graphite. [10][11][12][13][14][15][16][17] Consequently, the plated lithium metal while charging a graphite-based battery at low temperatures would like to react with electrolytes and form lithium dendrites or "dead lithium," which may penetrate the separator and trigger an internal short circuit with safety issues. [18,19] It is a widely held belief that graphite anode takes responsibility for the poor performance of LIBs at low temperatures. [1][2][3] Therefore, much attention and great efforts have been paid to improving the low-temperature operation of graphite anode so as to fabricate graphitebased batteries with high reversible capacity and prolonged cycle life under low-temperature conditions.
To gain a deeper understanding, it is primarily essential to comprehend the process of charge/discharge in the graphite anode, which involves the following six crucial steps: (1) Li + migration in the bulk electrolyte; (2) the desolvation process of Li + ; (3) Li + diffusion through the solid electrolyte interface (SEI) layer of graphite; (4) the diffusion of Li + across the solid-solid interphase between the graphite anode and graphite SEI; (5) Li + diffusion within the graphite anode; and (6) Li + insertion into graphite anode. [20] On the basis of this mechanism, it is widely recognized that the low-temperature performance of graphite-based LIBs is significantly hindered by several critical challenges, including (1) the restricted liquid temperature range and inferior ionic conductivity of electrolytes; (2) sluggish Li + desolvation process and Li + transferring across SEI; (3) poor Li + diffusivity in solid graphite active materials; and (4) the risk of Li plating on the graphite surface, which is likely to cause safety issues. [21][22][23][24][25] To address these issues, researchers are exploring various approaches such as modifying the electrode materials, developing new electrolyte formula, optimizing the electrode structure, and designing robust SEI layer. [26,27] These efforts have shown promising results in improving the low-temperature performance of graphite-based batteries. [14,28,29] Nevertheless, there lacks a comprehensive review of these strategies and corresponding analysis to show the feasibility of these approaches to improve the performance of graphite anode and corresponding LIBs at low temperatures.
This review provides an overview of latest researches on graphite-based LIBs operated at low temperatures in terms of two main aspects, that is, electrolyte engineering and material modification ( Figure 2). As the crucial media to support ions shuttle, electrolyte engineering is first discussed to clarify the formulations, solvation structure, interfacial chemistry, and their influence on low-temperature performance, which outputs feasible strategies for low-temperature electrolyte design. Given the limited diffusion of Li + in the layered graphite at low F I G U R E 1 Schematics of lithium plating and its consequences. [10] Copyright 2020, Royal Society of Chemistry.
F I G U R E 2 Schematic of promising strategies for highperformance low-temperature graphite-based batteries. temperature, the review further outlines the advancements of material modification that focus on improving the solid bulk diffusion rate and classifies the general strategies with excellent performance at low temperatures. With the comprehensive understanding, this review aims to provide valuable guidelines and strategies for extraordinary graphite-based batteries under lowtemperature scenarios to meet the urgent need of temperature-adapted LIBs.

| ELECTROLYTE ENGINEERING
The severe capacity degradation at subzero temperature arises primarily from the electrolyte, which suffers from the poor Li + conductivity, sluggish Li + desolvation process from the solvation sheath, or the failure to form a stable and inorganic-rich SEI. [30][31][32][33] Therefore, it is an essential prerequisite to tailor electrolyte formulations that can well meet the specific demands of these conditions. [34][35][36][37] As an indispensable component of the electrolyte, solvents can determine the main characteristics of electrolyte in operation temperature range, salt dissociation degree and ionic conductivity, solvation affinity with Li + and corresponding desolvation energy barrier, as well as the electrochemical potential window. Commercialized graphite-based LIBs generally take carbonate-based electrolyte, where ethylene carbonate (EC) has long been recognized as an essential component, thanks to its high dielectric permittivity and the ability to form a protective film. [38][39][40] However, its high melting point (36.4°C) and strong binding energy with Li + greatly hinder the transportation of ions in bulk electrolyte and the desolvation process of solvated-Li + at low temperature. For this reason, the greatest concern is to find suitable electrolyte formulations with comprehensive consideration of the conductivity, viscosity, liquid range, and solvation ability ( Figure 3). [41,42] Electrolyte engineering has therefore been proposed to overcome the hurdles of ions movement in the bulk electrolyte and the graphite/electrolyte interphase. [43] Tremendous efforts are made by iteratively adjusting the ratios of salts and solvents, incorporating cosolvents and additives, and exploring novel electrolyte systems, which will be discussed in the following sections.
2.1 | Improving the bulk ionic conductivity with low melting point solvents Electrolytes are predominantly composed of solvents, which play a crucial role in facilitating the conduction of electric charge in batteries. [44] Nevertheless, decreasing temperature would generally cause a rise in solvent viscosity, resulting in the partial or even total solidification of the electrolyte and impeding its practical application. [45] The selection of solvents possessing optimal physicochemical properties can potentially surmount the constraints imposed by harsh conditions and guarantee the proficient functioning of batteries in severe temperatures. Accordingly, the crucial study for low-temperature electrolyte is to develop customized solvent systems that exhibit exceptional characteristics, such as superior ionic conductivity and low melting point. Figure 4A presents a comparative analysis of several carbonate solvents based on their melting points, viscosities (η), and dielectric constants (ɛ). [46] It can be detected that linear carbonates exhibit lower melting point and viscosity than cyclic carbonate, elucidating the commonly used carbonate mixture of EC with methyl carbonate (EMC) or diethyl carbonate (DEC). For instance,~65% of the roomtemperature capacity can be retained in the electrolyte system of 1.0 M LiPF 6 in EC/DEC/DMC/EMC (1/1/1/3, v/v), which is much better than in the baseline electrolyte of 1.0 M LiPF 6 in EC/DMC/EMC (5/3/2, v/v,~15% capacity retention) under −40°C at 1/4 C. [47] Moreover, increasing the amount of EMC, leading to the decreased amount of high melting point EC, would have a positive effect on the Li + -intercalation process and potentially prevent lithium plating at low temperatures. [48] Apart from linear carbonates, linear carboxylate solvents have also been used as cosolvents for low-temperature electrolytes owing to the low melting point, thin viscosity, and acceptable permittivity ( Figure 4B). Benefitting from these features, elevating the content of linear carboxylate solvents can effectively decrease the viscosity and expand the lower limited temperature of the electrolyte. For instance, when the proportion of methyl butyrate (MB) in the electrolyte system of 1.0 M LiPF 6 in EC/EMC/MB (1/1/8) surpasses 75%, more than 80% of the room-temperature capacity can be achieved for graphite|LiMn 2 O 4 pouch cell under −60°C. [49] Specifically, the molecular structures of commonly used linear carboxylate solvents are summarized in Figure 4C, where a shorter chain length generally exhibits smaller viscosity and thus effectively lowers the viscous resistance that drags the ions movement in bulk electrolyte. However, the long-chain ones are favorable to generate an interphase with higher quality, thus after a balance, applying ethyl butyrate (EB) prefers to operate more effective with high capacity for Li|graphite half cell at −20°C. Smart et al. [50] have investigated the effects of various linear carboxylate esters and compared the electrochemical performance of graphite-based LIBs at low temperature, which shows that electrolyte modification with appropriate carboxylate solvents would be viable to improve the ionic conductivity at low temperature. [41] As discussed above, the indispensable solvent EC limits the liquid temperature range of commonly used electrolytes F I G U R E 4 Melting points, viscosity, and dielectric constant of different (A) carbonates and (B) carboxylates. [48] Copyright 2021, Wiley-VCH. (C) The corresponding molecular structure of carboxylates. [48]  owing to its high melting point of 36.4°C. To circumvent the root shortcoming, its immediate sibling, propylene carbonate (PC), has been proposed as its convenient substitute due to a much lower melting point of −48.8°C. [51][52][53] Unfortunately, early research has revealed that the cointercalation of PC would inevitably occur and cause the exfoliation of graphite, leaving the failure of graphite in PC-based electrolyte. [54] With a deep understanding of the electrolyte, it is found that tuning the solvation sheath can mitigate this problem by introducing cosolvent. Therefore, a new cosolvent, N-methylpyrrolidone (NMP, donor number [DN] = 27.3), was introduced as a high-donicity cosolvent to pair with PC, which can decrease the presence of PC in the solvation sheath of Li + . [55] Therefore, it is amazing to find that the cointercalation of PC disappears and the excellent reversibility of Li + intercalation can be observed. As shown in Figure 5A, Li + is mainly surrounded by NMP solvents, where the reduction process of PC is intervened and the gas evolution process is retarded. As a result, this NMP/PCbased electrolyte formula outperforms the commercial PC-based one in graphite-based full cell under the low temperature of −30°C. Similarly, Qin et al. developed a promising PC-based electrolyte with the assistance of fluorobenzene (FB) cosolvent to achieve the lowtemperature adaptation. [56] The intermolecular interaction of FB and PC weakens Li + -PC combination in Figure 5B, yielding an improved desolvation process on graphite surface and facilitating the generation of a thin SEI over a wide-temperature range. Moreover, the usage of DEC as a low coordination-number cosolvent will allow more anions into the solvation shell, greatly improving the compatibility of PC with graphite anode. [57] As a result, a pouch cell with graphite anode and PC-DEC electrolyte demonstrates excellent electrochemical performance over a wide temperature range in Figure 5C. These findings uncover new opportunities from molecular-scale design for developing electrolytes based on PC to profoundly extend the service temperature of the graphite-based LIBs. Therefore, by introducing cosolvents like NMP or DEC to suppress the presence of PC in the solvation structure, or employing the nonsolvating solvent FB to weaken the interaction of PC-Li + , the disadvantages of PC-based electrolyte can be overcome to achieve improved intermolecular interactions with the graphite surface, leading to better desolvation and reduced risk of cointercalation. These modifications have been shown to result in improved reversibility of Li + intercalation and longer cycle life, which are critical for achieving high-performance and long-term stability in LIBs. Therefore, it is a promising avenue for future research and development of high-performance LIBs for a wide range of applications.

| Facilitating the desolvation process with weakly solvating electrolyte (WSE)
The sluggish desolvation progress of Li + is widely recognized as the primary bottleneck hindering the F I G U R E 5 (A) The solvation structure of the PC/NMP electrolyte system and the corresponding electrolyte simulation of the optimized geometry of Li + (NMP) 2 (PC) 2 PF 6 − . [55] Copyright 2022, American Chemical Society. (B) Illustrations for the interfacial behaviors of the PC/FB-based electrolytes and the solvation structure. [56] Copyright 2022, Wiley-VCH. (C) The illustration of the solvation structure in all-climate PC-based electrolyte and the corresponding discharge curves of the pouch cells. [57] Copyright 2022, American Chemical Society. DEC, diethyl carbonate; FB, fluorobenzene; NMP, N-methylpyrrolidone; PC, propylene carbonate; SEI, solid electrolyte interface.
Li + -intercalation of graphite at subzero temperature. The development of WSE has emerged as a highly promising strategy to circumvent this obstacle. [58] By reducing the energy barrier associated with Li + , WSE can effectively promote the desolvation process of Li + from solvation structure before its intercalation into graphite. [59][60][61][62] Nevertheless, the weak solvation affinity with Li + also leads to the loss of ionic conductivity owing to insufficient dissociation ability of lithium salt. Therefore, it is of great importance to identify viable WSE with optimized physicochemical properties that well balance the desolvation energy and ionic conductivity at low temperature.
Among the recently developed WSE, fluorinated electrolytes stand out owing to their unique features, including tamed solvation energy with Li + ions, enhanced anti-oxidation stability, tendency to generate F-containing interphase film, and so on. [63,64] All these can be attributed to the introduction of F atoms with high electron affinity and therefore attract electrons from the surrounding atoms to realize the weak interaction with Li + ions. Extreme conditions can be represented with fluorinated ethers (FE), which are widely applied as the diluent in concentrated electrolytes owing to their rather weak solvation ability and little dissolution ability for lithium salt. [63,64] Researchers have tried to control the fluorination degree and fluorination sites to adjust acceptable fluorinated solvents. Meanwhile, introducing proper cosolvents is also feasible to improve the ionic conductivity at low temperature. As a typical example, ethyl trifluoroacetate (ETFA), identified as a weakly solvated solvent, has been found to facilitate the desolvation process as shown in Figure 6A. Together with a film-forming cosolvent (fluoroethylene carbonate [FEC], 30% in vol), the electrolyte with precise manipulation confers the graphite anode with remarkable performance even at the ultralow temperature of −60°C. [65] To deeply verify the importance of the WSE in low-temperature operation, Lei et al. also formulated an electrolyte, where acetonitrile (AN) with a low DN was chosen as the solvent and FB as the cosolvent. [66] The detailed Li + -solvent interactions of Li + (AN) 2 (FSI − ) were further assessed via quantum chemistry simulations as shown in Figure 6B. The poor solvation ability of FB facilitates the detachment of Li + from the solvation sheath, thus achieving ultrarapid Li + diffusion via SEI and exhibiting the powerful low-temperature performance of graphite anode. Moreover, some novel solvents with weak binding energy have also been investigated to favor the electrochemical performance of graphite-based LIBs. Yao et al. took the advantage of the weak solvation power of 1,4-dioxane solvent, where the electrolyte promotes the desolvation process of Li + and facilitates the coordination of anions with Li + , thus generating an anion-derived interfacial chemistry under low salt concentration ( Figure 6C). [67] Nowadays, a fivemembered heterocyclic compound, isoxazole (IZ), has emerged as a highly promising candidate with a remarkable combination of advantageous properties, including superior ionic conductivity and weak binding energy with Li + . [68][69][70] IZ exhibits a dispersed distribution of charge owing to the existence of a conjugated structure, which is capable of delivering a reversible capacity that is up to twice as high as that of commercial EC-based electrolytes at −30°C as illustrated in Figure 6D, confirming the feasibility of novel solvent with lower binding energy in chilly temperature operation. Thus, the exploration of WSE offers a new insight to overcome the sluggish Li + desolvation process and serves as an emerging principle for coming studies on precise electrolyte engineering. This strategy may ultimately pave the way for the development of innovative energy storage technologies that can thrive in challenging conditions.

| Promoting the ions movement across SEI layer with an inorganic-rich interphase
The study of interphase chemistry represents a fascinating and formidable area of research in the field of LIBs and exerts a profound influence on electrochemical behavior. [71,72] As the interface between the electrode and the electrolyte, SEI plays a crucial role in dictating Li + migration kinetics and ensuring reversible intercalation/deintercalation behavior. [73] The structure and chemical composition of the SEI are key determinants of its properties, making it a complex and challenging system to understand and manipulate. [74] Despite these challenges, advances in interphase engineering offer promising avenues for improving the performance and longevity of LIBs. [75,76] As shown in Figure 7A, a thin and initial SEI is formed when graphite anode is lithiated in the first cycle, and after significant aging via following cycling, an efficient, passivating compact SEI is established consisting of abundant inorganics like Li 2 O, Li 2 CO 3 , which plays a critical role in maintaining fast Li + transport and preventing further degradation. [77] However, if the SEI is predominantly composed of organic alkyl carbonates with few inorganic compounds, an extended and uneven SEI is likely to be formed due to incomplete passivation, which is the principal consumer of Li + . An SEI layer with inferior ionic conductivity will hinder the transport of Li + , particularly at high current rates and low temperatures, resulting in a progressive rise in interface impedance that undermines the performance of LIBs. Recently, the inorganic content in SEI, such as LiF, Li 3 N, Li 2 O, or Li 2 CO 3 , is becoming a popular research field thanks to their good electrical insulation and high ionic conductivity, contributing to fast Li + migration kinetics in the interphase. [78] For example, the application of additives can help modify the components of the SEI layer to achieve an inorganic-rich SEI. And recent advances in the emerging interfacial chemistry of graphite anodes are exhibited in Figure 7B, where adjusting the electrolyte composition to promote the formation of more contact ion pairs (CIPs) and aggregates (AGGs) in the solvation sheath has emerged as the most widely adopted and effective strategy to achieve the F I G U R E 6 (A) Schematic illustration of Li + diffusion process from electrolyte to electrode, with an energy-consuming desolvation step and the discharge curves of graphite(Gr)|LiFePO 4 (LFP) under different low temperatures after charging at +25°C. [65] Copyright 2022, Wiley-VCH. (B) Schematic illustrations of the solvation structures and the process of Li + -intercalation graphite layer in AN-FB-based electrolyte and the corresponding Li + -solvation structure. [66] Copyright 2022, Elsevier Ltd. (C) The solvation structures of different electrolytes and the ranking of solvating power of solvents from high to low. [67] Copyright 2020, Wiley-VCH. (D) The electrostatic potential (ESP) map of IZ and EC, binding energy of Li + -solvent, and the corresponding reversible capacity of EC-based and IZ-based electrolytes obtained under various temperatures at the current rate of 0.1 C. [68] Copyright 2023, Wiley-VCH. DME, dimethoxyethane; DX, dioxane; EC, ethylene carbonate; IZ, isoxazole; LSCE, localized superconcentrated electrolyte; SCE, superconcentrated electrolyte; SEI, solid electrolyte interface. desired interfacial chemistry, which can generate an anion-derived inorganic-rich SEI layer. [79,80] Generally, the adoption of additives is the most convenient and economical method to improve the performance of commercial LIBs and is highly accepted by the industry. [81] The application of additives to enhance low-temperature performance of LIBs has been the subject of intensive investigation in recent years. One typical illustration is that lithium difluorobis(oxalato)phosphate (LiDFOB) was applied as a typical additive to suppress the decomposition of solvents and then endowed the SEI with more LiF and Li 2 CO 3 ( Figure 8A). [82] Incorporating inorganic components into the SEI has been shown to enhance its ionic conductivity and decrease the impedance. Thanks to the presence of LiF, with a high Young's modulus, the interphase becomes more resilient and stable, making it particularly suitable for using in low-temperature conditions. [82] Similar findings were also discovered for lithium difluorophosphate (LiPO 2 F 2 ) additive, which suppresses the continual decomposition of solvent and accelerates the transfer of Li + . [83] Furthermore, the addition of 2 vol% N-N-dimethyltrifluoroacetamide (DTA) to commercial EC-based electrolytes has been shown to endow the SEI with more LiF content and decrease the amount of organic components, which results in reduced resistance of graphite-based half cells, particularly in challenging conditions. [84] When DTA (an electron pair donor) is combined with PF 5 (an electron pair acceptor), a weak composite is supposed to be formed, reducing the reactivity of PF 5 and decreasing the likelihood of gas expansion occurring. Additionally, DTA provides F-groups which can react with Li + to form LiF, leading to an appropriate amount of LiF and ultimately improving the stability of the SEI film ( Figure 8B). [72] In addition, a novel SEI-forming electrolyte additive, known as fluorosulfonyl isocyanate (FI), was used with graphite anodes due to the higher reduction potential (above 2.8 V vs. Li + /Li), before the reduction of EC solvent as shown in Figure 8C. [85] The electrochemical polymerization of the -NCO group in FI leads to polyamide-like SEI components and a higher content of inorganic species, which helps yield a compact and stable SEI and facilitates the charge-transfer process of the graphite anode at subzero temperature. [86,87] Additionally, researchers synthesized an innovative additive, erythritol bis(carbonate) (EBC), with a double-EC structure but a lower lowest unoccupied molecular orbital (LUMO) energy than EC ( Figure 8D), which prefers to reduce at a higher potential. [73] The tailored electrolyte with 2% EBC additive shows off excellent film-forming characteristics and expresses great stability under subzero temperature of −20°C. Guo et al. reported that the addition of 0.5 wt% dimethyl sulfite (DMS) in carbonate electrolyte endows the graphite|LiNi 0.5 Co 0.2 Mn 0.3 O 2 pouch cell with superior low-temperature performance. [88] The decomposition product of DMS presents weak combination with Li + and constructs a more stable and highly conductive SEI film, suppressing the undesired reduction of EC solvent. Therefore, the rational design of a stable SEI film via the adoption of additives has emerged as a cost-effective and highly effective strategy for tailoring electrolyte formulations to break through the bottleneck of the state-of-the-art battery systems. This strategy is particularly relevant for graphite-based LIBs, where low-temperature performance has been identified as a key challenge. By adding suitable additives to the electrolyte, it is possible to modify the SEI film and enhance its stability, ion conductivity, and mechanical strength. [89] These improvements can effectively mitigate the undesirable reactions that occur under harsh operating conditions.
In addition to utilizing additives to regulate the components of the SEI layer, high-concentration electrolyte (HCE) is also adopted to incorporate more anions into the Li + solvation sheath. However, due to the high viscosity, inferior wettability, and low ionic conductivity, its potential for use with graphite anodes under subzerotemperature conditions is severely prevented. To overcome these limitations, a new concept, localized high-concentration electrolyte (LHCE), has been proposed by introducing a diluent to shield the Li + electrostatic F I G U R E 7 (A) Schematic of SEI formation on the carbon negative electrode. [77] Copyright 2019, American Chemical Society. (B) Recent advances in the emerging interfacial chemistry of graphite anodes. [79] Copyright 2020, Royal Society of Chemistry. SEI, solid electrolyte interface.
attraction. [90][91][92] This approach aims to generate a local high-concentration environment surrounding the electrode surface, which allows for more efficient ion transport and a better SEI formation. Moreover, the diluent enhances the wettability and fluidity of the electrolyte, thus improving its low-temperature performance. To deeply clarify the important role LHCE plays in determining the performance of LIBs, Jiang et al. designed an LHCE with 1.5 M LiFSI as the lithium salt, dimethoxyethane (DME) as the solvating solvent and bis(2,2,2-trifluoroethyl) ether (BTFE) as the diluent to optimize the low-temperature performance of graphite ( Figure 9A). [93] BTFE has a low tendency to coordinate with Li + , which effectively dilutes the concentration of the solvent while increasing the concentration of the lithium salt, promoting anion-derived interfacial chemistry. Furthermore, the preferential decomposition of anions results in a robust and uniform SEI composed of inorganic species, which significantly suppresses the cointercalation of ether solvents and enhances Li + transport kinetics. As a result, the graphite|Li with the designed LHCE can yield a considerable capacity of 90 mAh g −1 at 0.1 C under −20°C, demonstrating the impressive low-temperature performance. Besides, they also discovered similar findings in another report, where methyl acetate (MA) and FE were applied as the main solvent and diluent, respectively. [94] Through anion decomposition, the FSI − -derived SEI exhibits lower interfacial impedance and avoids lithium plating, endowing graphite-based pouch cells all-round advantages under −60°C. The MA-based electrolyte exhibits more anions in the first solvation sheath than the commonly used EC-based electrolyte, with fundamentally different film-forming mechanisms. Moreover, Nan et al. reported F I G U R E 8 (A) Mechanism of the SEI derived by LiDFBOP or EC and the corresponding discharge curves of the graphite/Li cells with/without LiDFBOP at −20°C. [82] Copyright 2021, American Chemical Society. (B) The reaction occurring on the surface of the SEI in the electrolyte with and without DTA. [84] Copyright 2020, ESG. (C) HOMO and LUMO orbitals of DMC, EC and FI molecules. [85] Copyright 2019, Elsevier. (D) HOMO and LUMO orbitals of EC, EBC and VC molecules; and the comparison of discharge capacities at −20°C from 0.1 C to 1 C. [73] Copyright 2022, Elsevier. DFBOP, difluorobis(oxalato)phosphate; DMC, dimethyl carbonate; DTA, N-N-dimethyltrifluoroacetamide; EBC, erythritol bis(carbonate); EC, ethylene carbonate; FI, fluorosulfonyl isocyanate; HOMO, highest occupied molecular orbital; LiDFBOP, lithium difluorobis(oxalato)phosphate; LUMO, least unoccupied molecular orbital; SEI, solid electrolyte interface.
another exemplary electrolyte, where the low melting point (−53°C) and small dipole moment (μ = 0.89 D) solvent of EMC solvent suggest the poor ion-dipole combination with Li + and provides improved Li + transport kinetics for desolvation process at low temperatures. [95] Together with the extrinsic help of 1,1,2,2tetrafluoro-3-(1,1,2,2-tetrafluoroethoxy) propane (TTE) diluent, the presence of the TTE-rich and LiFSI (EMC) 1.6 -rich domains can be observed and the solvation structure is dominated by a high ratio of CIPs and AGGs ( Figure 9B), promoting the formation of anion-derived SEI. The unique solvation structure of LHCE can efficiently facilitate the preferential decomposition of anions, generating a highly conductive SEI with considerable content of inorganic species. Similar findings have also been reported by Xu et al. recently, who validated an electrolyte design strategy based on methyl difluoroacetate solvent, exhibiting weak interactions with Li + in Figure 9C. [42] Together with the diluents (TTE), the designed electrolyte promotes the formation of a LiF-rich SEI and endows the graphite-based battery an amazing low-temperature performance of 54% capacity retention at −60°C. Therefore, the development of LHCE opens up a brand new avenue to extend the application of HCE to graphite-based LIBs operating in extreme environments while other practical properties are well maintained.
To further develop liquid electrolytes with fast Li + transport kinetics and enhanced interface stability, a relatively new electrolyte, high-entropy electrolyte (HEE), has entered our field of vision nowadays. [97][98][99][100] HEEs are generally composed of multiple cations and anions, typically five or more, which are mixed in equal  [96] As a result, combining graphite anode with LiNi 0.8 -Co 0.1 Mn 0.1 O 2 cathode can achieve the cycling stability for over 1000 cycles while maintaining a capacity retention of over 90%, where the improved performance can be determined by the unique solvent structure of the HEE. Additionally, with a higher solvent structuring entropy, HEEs can increase the freedom and disorderliness of solutes in solution, thereby slowing down their tendency to form ordered crystals during crystallization. [96] This might endow HEEs with low melting point and high ionic conductivity, potentially improving the lowtemperature performance of the battery. As an area of active research, the HEE electrolyte strategy brings comprehensive favored properties and is likely to become increasingly important in the future development of high-performance electrolyte.

| MATERIAL MODIFICATION
Significant efforts have been made to develop advanced electrolytes to improve electrochemical performance, however, there are still hurdles in achieving satisfactory capacity retention at subzero temperatures, which can be ascribed to the inherent structure of layered graphite. Moreover, a hazardous phenomenon, Li plating, which can lead to the formation of dendrites, dead lithium, and even short circuits is also related to the morphology of the graphite anode and the corresponding SEI on its surface. In particular, the Li deposition can damage the integrity of the SEI, leading to a decline in battery performance and increased safety risks. [2,3] Additionally, the specific surface area of the graphite has a great influence in preventing Li plating and the formation of the SEI. The specific surface area of graphite is related to the shape and surface structure of graphite particles. For instance, artificial graphite has a unique microporous structure with increased specific surface area, which provides a more active site for Li + insertion and results in high-rate performance. In contrast, the excessive specific surface areas of graphite can cause larger capacity loss in the first cycle and increased internal resistance due to the addition of more binders. Therefore, either too large or too small a specific surface area of graphite particles is not conducive to the reversible insertion and extraction of Li + . Only an appropriate specific surface area can maximize the reversible intercalation of Li + . [20,[101][102][103] Generally, the anisotropic nature of Li + diffusion requires a higher energy barrier for Li + intercalation into the basal surface of graphite anodes compared with the edge surface. [20] However, traversing through the edge surface entails that Li + must navigate through a longer diffusion pathway, impeding the kinetics of Li + diffusion. [101][102][103] Additionally, as a resilient form of carbon, graphite exhibits a narrow interlayer spacing of 0.335 nm, which may not be adequately wide to facilitate swift and reversible Li + intercalation/deintercalation. As a consequence, the inherent limitations of graphite, such as long Li + diffusion pathways and sluggish Li + diffusion kinetics within the interlayers, will impede its use in low-temperature scenarios, necessitating the incorporation of anode material a crucial factor during battery fabrication. [104]

| Creating more Li + -intercalation sites
To solve the intrinsic drawbacks of graphite anode mentioned above, graphite structure modification is urgently required. In this way, designing holes on the graphite surface and combining with conductive materials is likely to provide shorter diffusion pathway for Li + , which also favors to the low-temperature performance. Fine-tuning the pore structure of disordered carbon materials offers a powerful and effective solution for enhancing lithium storage kinetics. By meticulously controlling the pore size, shape, and distribution, the diffusion of Li + can be significantly improved, leading to enhanced electrochemical performance. [105] This approach allows the creation of materials with tailored properties that can be optimized for specific applications, such as high-energy-density batteries. Moreover, the ability to control the pore structure offers an advantage in terms of scalability, making this approach a promising avenue for the large-scale production of highperformance energy storage materials. [106] A prime illustration is the combination of porous graphite nanosheets with carbon nanotubes (CNTs) as shown in Figure 10A, where the through-holes serve to reduce the distance of Li + migration and provide additional Li +intercalation sites, while the CNTs prevent the restacking of the two-dimensional graphite sheets. [45,107,108] As a result, together with the low-temperature electrolyte (0.75 M LiTFSI in 1,3-dioxane), the graphite-based battery retains 90% of capacity retention after 500 cycles under 4 C and room temperature and delivers the excellent low-temperature capacity of 300 mAh g −1 at 0.1 C and −20°C. This strategy optimizes the performance of the materials by synergistically combining the advantages of each component, leading to a significant improvement in capacity and cycling stability.

| Expanding the interlayer spacing
The fast diffusion of Li + is also hindered by the small interlayer spacing of graphite, especially under low temperatures. Thus, expanding the interlayer spacing is necessary to effectively improve the sluggish kinetics of Li + insertion, which can significantly enhance the performance of the electrode material, enabling it to maintain a high energy storage capacity and stable cycling performance even under demanding conditions. [111] Zhao et al. [109] and Kim et al. [112] prepared expanded mesocarbon microbeads (EMCMB) by oxidizing pristine mesocarbon microbeads (MCMB) and subsequent heat treatment ( Figure 10B). Following the modification, the expanded interlayer distance reserves more sites for Li + diffusion and reduces the activation energy required for Li + migration in bulk graphite. As a result, EMCMB exhibits an extraordinary capacity of 100 mAh g −1 at −40°C, while less than 50 mAh g −1 at −30°C can be delivered for pristine MCMB. In another contribution, synthesizing multilayer crystalline graphene (GRAL) to enhance surface area also outperforms unmodified graphite anode under low temperature in Figure 10C. [110] This effect significantly enhances the transport of Li + , leading to improved electrochemical performance of the material. Consequently, the electrode material exhibits enhanced performance under challenging operating conditions.

| Surface coating technology
Coating is another effective way to modify graphite and can be achieved through the use of carbon-based materials or inorganic compounds. [113][114][115][116][117] This approach offers a versatile mean of improving the performance of graphitebased electrode materials, allowing for the creation of materials with enhanced energy storage capacity and improved cycling stability. Moreover, the ability to tailor the properties of the coating material enables satisfactory optimization of the electrode performance, making it a promising avenue for the development of highperformance energy storage materials. As demonstrated in Figure 11A, Cai et al. employed a nanoscale turbostratic carbon layer on the graphite surface, where the coating layer possesses a larger interlamellar spacing and provides more Li + -intercalation sites and Li + diffusion motion interlayer from the basal plane of graphite sheet. [118] The larger interlamellar spacing and isotropic nature of the F I G U R E 10 (A) Schematic illustration of Li + insertion process for natural graphite and PGN/CNT as well as the preparation process. [106] Copyright 2019, Elsevier. (B) SEM images of MCMB and EMCMB and the corresponding Li + insertion/extraction curves of MCMB at 0°C, −10°C, and −30°C, and EMCMB at −40°C under the current density of 0.2 C. [109] Copyright 2012, Elsevier. (C) Schematic representation of the steps to produce GRAL and the influence of the temperature on the voltage profiles of GRAL and graphite. [110] Copyright 2015, Elsevier. CNT, carbon nanotube; EMCMB, expanded mesocarbon microbead; GRAL, multilayer crystalline graphene; MCMB, mesocarbon microbead; PGN, porous graphite nanosheet; SEM, scanning electron microscope. turbulent layer carbon can eliminate the anisotropic nature of the material and provide more ordered/ disordered mixed channels for the rapid migration of Li + ions, resulting in reduced polarization and improved electrochemical performance. The ability to facilitate rapid Li + migration is crucial for achieving high specific capacity and high-rate performance at low temperature. [119] Therefore, the turbostratic carbon-coated graphite (G@TC) is prone to benefit the low-temperature performance, where a retarded lithium plating and a higher reversible capacity can be observed at G@TC under 0°C compared with the raw graphite (G). Coating the anode surface with a layer of ceramic Al 2 O 3 has been applied to investigate the aging at low temperatures. [120] The inorganic Al 2 O 3 layer exhibits high thermal stability and minimal shrinkage, which prevents further short circuits and thermal runaway. In Figure 11B, the lithium metal deposit is mostly found underneath the intact coating layer on the electrode surface, indicating that the Al 2 O 3 coating effectively restricts the deterioration in safety behavior and prevents additional short circuits caused by dendrite pierce. It is revealed that, even with a thick lithium metal layer on the anode surface, the Al 2 O 3 coating remains a significant factor in maintaining safety. As the cell ages, the Al 2 O 3 coating may lift, but the lithium metal deposit is still located beneath the mostly intact coating layer on the electrode surface, which makes coated Al 2 O 3 with promising safety feature for large-scale cells and can increase safety during mechanical abuse. [120] Moreover, the Al 2 O 3 layer on its surface could also induce the enhanced electrolyte wettability on the graphite in Figure 11C. [121] The enhanced wetting ability allows for a more effective contact between the electrolyte and the graphite electrode, promoting the transport of Li + and enhancing the energy storage capacity of the system. In addition, studies have demonstrated that the application of an Al 2 O 3 coating on graphite can effectively mitigate the accumulation of electrolyte decomposition products. This, in turn, improves the overall performance of LIBs at wide temperatures. [122]

| Synthesizing metal-graphite composite
The combination of metal and graphite has been reported to improve the low-temperature performance thanks to the synergistic effect between the metal and graphite. Metal-graphite composites typically exhibit better conductivity, which can enhance the power density and energy density of batteries, maintaining high performance even at low temperatures. Additionally, metal-graphite composites can improve the mechanical strength and stability of materials, thereby enhancing the cycle life of batteries. In Figure 12A, it was reported that synthesizing an Sn/C composite with nano-Sn embedded in expanded graphite (Sn/EG) can enhance low-temperature performance. [123] The uniform distribution of nano-Sn particles in the F I G U R E 11 (A) The morphology of G and G@TC, the scheme of Li + diffusion into graphite anode, and the corresponding charge/ discharge curves at the rate of 0.1 C at 0°C. [118] Copyright 2020, Wiley-VCH. (B) Lithium metal deposition mechanism at second charge and anode of the examined cell. [120] Copyright 2017, Elsevier. (C) Schematic illustration of Al 2 O 3 coated on a graphite surface and the corresponding rate capabilities. [121] Copyright 2019, Elsevier. G, graphite; G@TC, turbostratic carbon-coated graphite; SEI, solid electrolyte interface.
YIN and DONG | 581 interlayers of expanded graphite forms a tightly stacked layered structure, which can improve the lithiation kinetics of Li + . This leads to the improvement of the electrochemical performance of the Sn/EG composite as an electrode material, including increased specific capacity and improved cycling stability. Moreover, the catalytic effect of metals can promote the fast desolvation process, which is important for the low-temperature performance of graphite-based electrodes. Thus, adding a small amount of nanometric metal particles to the graphite can be a promising approach to improve the electrochemical performance of the electrode at low temperatures in Figure 12B. [124] Analogously, Marinaro et al. [125] and Mancini et al. [126] verified the significant role of nanometal playing to promote the graphite subzero intercalation performance via applying nanocopper on Super-P carbon (Cu/Super-P) as conductive additive. In this case, the bulk conductivity of graphite and charge-transfer process can be greatly improved due to the catalytic role of Cu in Figure 12C. The metal particles embedded in the graphite matrix can serve as active sites for Li + adsorption, reducing the energy barrier for Li + insertion into the graphite. Additionally, the catalytic effect of the metal can accelerate the desolvation process, leading to fast Li + diffusion and improved low-temperature performance of the battery. The uniform distribution of nanometric metal particles in the graphite interlayers can further improve the lithiation kinetics of Li + . These findings suggest that metal-graphite composites hold great promise for developing LIBs with excellent low-temperature performance.
Besides, it is important to note that despite the potential benefits of the above modification methods, there are still some limitations and drawbacks that need to be carefully considered. For instance, creating too many Li + -intercalation sites or excessively expanding interlayer spacing can lead to a decrease in structural stability and mechanical strength of the material, as well as increase the volume expansion during the charge/ discharge process. Additionally, the coating material may undergo delamination during the charge/discharge process, which can reduce cycling stability. Moreover, the synthesis of metal-graphite composites can be complicated and expensive. Therefore, to achieve the most effective graphite modification, it is necessary to carefully evaluate and consider the advantages and limitations of each method, which allows us to choose the most suitable method or combination of methods to obtain the best performance. In this way, we can F I G U R E 12 (A) Schematic illustration of the synthesis process of Sn/EG. [123] Copyright 2016, Elsevier. (B) Intercalation profiles of Sn-graphite composite recorded at the C/5 rate in the temperature range 20°C to −30°C. [124] Copyright 2010, Elsevier. (C) Intercalation profiles of the Cu-modified electrode/unmodified electrode and the comparison of the capacities under different temperatures of them. [125] Copyright 2013, Elsevier. CVD, chemical vapor deposition; EG, expanded graphite; GO, graphite oxide. optimize the graphite structure and enhance its electrochemical properties, while minimizing any potential negative impacts or drawbacks that may arise from the modification process.

| CONCLUSION AND OUTLOOK
Addressing the challenges graphite anode and graphitebased LIBs facing is crucial to improve the lowtemperature performance and advance practical applications. Herein, we provided an overview of the current status of graphite-based LIBs operating under low temperatures, focusing on the promising strategies to promote Li + diffusion kinetics and mainly highlighting electrolyte engineering and material modification.
The preferred strategy is to optimize the electrolyte formula, which determines the ions movement in the bulk electrolytes as well as the properties of the interphase film involving the desolvation process and subsequent ions diffusion across the SEI layer. The utilization of solvents with a broad liquid range and low viscosity has been shown to enhance the mobility of ions in solvents. Furthermore, the exploration of weakly solvated solvents has been found to be effective in promoting the desolvation process of Li + ions. Certainly, suitable additives for film formation and designing LHCE can play a pivotal role in the generation of an inorganicrich SEI, which can significantly enhance the electrochemical performance of the system at subzero temperatures. The choice of additives can be tailored to optimize the SEI formation process by controlling the composition and properties of the SEI. Moreover, the establishment of LHCE helps form more CIPs or AGGs in the first solvation sheath, which can significantly improve the electrochemical performance of the system by promoting the formation of a robust anion-derived SEI. Through these approaches, it is possible to optimize the electrolyte to simultaneously support fast-charging performance and ensure the safety and stability of the graphite-based LIBs at low temperatures. The other viable strategy to address low-temperature performance issues in graphite-based LIBs is material modification. This can include in, coating with carbon or inorganic components, or even incorporating nanometal to catalyze the process to enhance the fast Li + transport kinetics and improve the overall efficiency of the system.
After decades of research dedicated to lowtemperature graphite-based LIBs, significant progress has been made. However, there are still scientific and engineering challenges that need to be addressed, leaving ample room for future development. We envision several critical frontiers that need to be explored, including: (1) The development of graphite materials with fast kinetics is crucial for meeting the demands of LIBs under low temperatures in the future. As the layered structure facilitates ion access and electronic transport, more promising methods are expected to be explored to increase the interlayer spacing and optimize Li + diffusion of graphite. In addition to improve the ion diffusion coefficient, it is also necessary to reduce the size of graphite particles to shorten ions and electrons conduction pathway and achieve outstanding performance. Additionally, current synthesis processes for fast-charging graphite anode, such as chemical vapor deposition and wetchemical synthesis, still have some limitations such as the expensive nature, ecological concerns, and time-intensive process, rendering them difficult to be produced on an industrial scale. Hence, it is imperative to devise synthesis methods that are economical and amenable to large-scale production. (2) In LIB configurations, the performance of the batteries is dominated by Li + conductivity, chargetransfer resistance, and the graphite interfacial resistance, which is considered as the primary factor responsible for the sluggish kinetics observed at low temperatures. Therefore, developing advanced lowtemperature electrolyte is a highly effective approach for enhancing the electrochemical performance of graphite-based LIBs. It is crucial to explore new electrolyte formula with superior ionic conductivity and novel electrolyte components to promise a smooth process for Li + diffusion and help establish an anion-derived inorganic-rich SEI that can effectively suppress Li plating and dendrite growth under severe conditions, such as lower than −70°C. (3) As previously elucidated, the intricate interplay between SEI formation and solvation structure manipulation lies at the heart of many fundamental challenges in the field of battery electrochemistry. Understanding and controlling these complex processes at the atomic level represents a critical step towards unlocking the full potential of advanced energy storage systems. Efficiently tuning both SEI formation and solvation structure simultaneously is crucial for advancing the performance of nextgeneration energy storage technologies. To this end, a promising strategy involves the use of SEI-forming agents that also function as key solvent components in the electrolyte formulation. By adopting this synergistic approach, we can realize unprecedented levels of control over the intricate interplay between solvation structure and SEI formation, thereby unlocking new avenues for enhancing battery performance.
YIN and DONG (4) It should be noted that the electrochemical behavior of graphite in a full-cell configuration is distinctive from that in a half-cell setup. Discrepancies in the fabrication and electrochemical protocols employed for coin cells versus pouch cells can give rise to marked differences in the SEI layer that develops on the graphite electrode surface. Thus, bridging the gap between academic research and industrial applications is an indispensable part we should take into consideration.
Through the judicious optimization of electrolyte composition and tailored design of graphite anode, the operational temperature range of graphite-based LIBs can be significantly expanded. Notably, certain batteries have been developed to meet the demanding operational requirements of low-temperature environments. As technology continues to evolve, we anticipate that the targeted development of low-temperature graphite-based LIBs will offer promising opportunities for niche market applications.