Significant impact of 2D graphene nanosheets on large volume change tin-based anodes in lithium-ion batteries: A review

Highlights

• The significant impact of 2D graphene nanosheets on Tin-based anodes has been reviewed.

• Various Sn/graphene based binary, ternary composites have been summarized.

• The preparation methods, properties and mechanism of Sn/graphene composites have been concluded.

• The challenges and perspectives of this field have been reviewed as well.

Abstract

Sn-based materials have attracted much attention as anodes in lithium ion batteries (LIBs) due to their low cost, high theoretical capacities, and high energy density. However, their practical applications are limited by the poor cyclability originating from the huge volume changes. Graphene nanosheets (GNSs), a novel two-dimensional carbon sheet with one atom thickness and one of the thinnest materials, significantly address the challenges of Sn-based anodes as excellent buffering materials, showing great research interests in LIBs. In this review, various nanocomposites of GNSs/Sn-based anodes are summarized in detail, including binary and ternary composites. The significant impact of 2D GNSs on the volume change of Sn-based anodes during cycling is discussed, along with with their preparation methods, properties and enhanced LIB performance.

Introduction

As one of the most important energy-storage devices, lithium-ion batteries (LIBs) have attracted a lot of attention in both academic research and commercial applications, such as cell phones, laptops and digital cameras [1], [2], [3], [4], [5]. Despite many advantages over other competing electrochemical power sources, the use of LIBs exhibit some limitations when used in electric vehicles, hybrid electric vehicles and plug-in hybrid electric vehicles, since they require high energy density, good cycling performance and high rate capability [6], [7], [8]. As a result, the search for new anode and cathode materials providing enhanced electrochemical properties while remaining economical is always an ongoing study [9]. Graphite, as the commercialized anode material, has its own advantages, such as low and flat working potential, long cycle life and low cost [10]. On the other hand, graphite limits the lithium storage performance in terms of energy and power densities due to the low theoretical capacity (LiC₆, 372 mAh g⁻¹) and low Li-ion transport rate (10⁻¹²–10⁻¹⁴ cm² s⁻¹), respectively [11], [12].

Of the alternative anode materials for LIBs, silicon shows great promise with a theoretical capacity of 4200 mAh g⁻¹. However, tin exhibits higher electrical conductivity and is less brittle than silicon. In addition, the preparation process of silicon anodes can be quite complex with poor controllability, inevitably increasing cost. Tin anodes shows higher volumetric capacity of about 2000 mAh cm⁻³ and gravitational capacity of 990 mAh g⁻¹ than the commercial graphite as well as other transition metal oxide anodes (such as NiO, Co₃O₄, and Fe₂O₃, etc). Derivatives such as Sn-based alloys, tin oxides, tin sulfides, and stannates, are also attractive anode materials for LIBs. A significant advantage of Sn-based anodes over the commercial graphite is that they avoid solvent intercalation, and improve safety performance of LIBs. Moreover, the Sn based anodes exhibit lower potential hysteresis than transition metal oxides. Therefore, Sn based anodes have attracted tremendous interest as promising anode materials for LIBs. [13], [14], [15], [16], [17], [18], [19], [20]. However, the practical application of Sn-based anodes is hampered by its poor cyclability, resulting from large volume changes of 259%, which results in electrical disconnection from conductive agents (e.g. carbon black) and the current collectors during the charge/discharge process [21], [22]. The high potential hysteresis in the charge/discharge curves is an important concern for LIB applications. For example, very high potential hysteresis more than 1.0 V of the transition metal oxides (like NiO, Co₃O₄, Fe₂O₃, etc) seriously prevents their commercialization as anodes in LIBs. The characteristic Li–Sn alloying/de-alloying reactions are shown as shoulders between 0.4 V and 0.8 V, which are characterized by an acceptable potential hysteresis, enabling the commercialization possibility of these types of anodes. The hysteresis is partly related to the electrical conductivity of the active materials. The active materials with higher electrical conductivity exhibit the improved potential hysteresis, which originates from decreased electrochemical polarization. For instance, the LiFePO₄ cathodes combined with conductive materials such as carbon showed lower hysteresis than pristine LiFePO₄ [23]. In the case of anode side, it was reported that the metal oxide/carbon nanocomposites can also effectively improve the hysteresis problem [24], [25], [26]. Several approaches have been proposed to address the challenges of the large volume change anodes. One effective method is to assemble the active particles into different nanostructures [27], [28], such as nanowires [29], [30], nanospheres [31], [32]， nanotubes [33], [34], [35], nanocubes [36], [37] and hollow nanostructures [15], [38], [39], [40], [41]. The second method involves doping with an inactive element to reduce the volume change [42], [43]. The third method is to make nanocomposites with an inactive/active matrix and stable components [40], [44], [45], [46], [47], [48], [49]. In the nanocomposites, the stable matrices act as a mechanical buffering zone against the large volume change of the active materials [44], [50]. Effective matrices provide the conducting backbone for the active materials, and the soft structures are beneficial to buffering the internal stress of electrodes suffering from large volume change. Therefore, Sn-based nanocomposites with matrices have been demonstrated to effectively improve anode battery performance.

Graphene is a single layer of sp² carbon atoms arranged hexagonally and has generated enormous excitement for various potential applications because of its fascinating properties [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], such as a high intrinsic carrier mobility (200 000 cm² V⁻¹ s⁻¹), excellent thermal conductivity (5000 W m⁻¹ K⁻¹), high optical transmittance (∼97.7%), high theoretical specific surface area (2630 m² g⁻¹), and superior mechanical strength [59].

As one of the new carbon materials, the Li-storage properties of graphene have been widely studied by numerous researchers [63], [64], [65], [66], [67], [68]. P.C. Lian et al. and A. Shanmugharaj's group synthesized graphene nanosheets (GNSs) from graphite powder through oxidation followed by rapid thermal expansion in nitrogen atmosphere [69], [70]. The different electrochemical properties of GNSs, expanded graphite and graphite have been compared as well [71]. The research results show that the reversible capacities of GNSs are almost double those of expanded graphite electrode and three times those of graphite electrode. D.Y. Zhao's group [72] prepared novel mesoporous graphene nanosheets (MGNs) with an excellent reversible capacity of 1040 mAh g⁻¹ at 100 mA g⁻¹ in the second cycle, and 833 mAh g⁻¹ after 60 cycles. The atomically flat graphene layers with mesopores provide high contact surface area for lithium ion adsorption and intercalation, while the open mesopores in the orthogonal direction on the nanosheets offer efficient transport pathways for ion diffusion toward the deep portions of the stacked graphene layers thus delivering excellent lithium ion storage capacity and cycling performance [72]. Several researchers reported that the greatly improved capacities in disordered GNSs are mainly ascribed to additional reversible storage sites such as edges and other defects [58], [73], [74], [75]. However, it should be noted that high surface area of graphene increased the contact area between the anode and the electrolyte, which results in more SEI formation in the first discharge [76]. For example, GNSs could show a high discharge capacity of 1233 mAh g⁻¹, but its high surface area resulted in a low Coulombic efficiency of 54% with irreversible capacity of 672 mAh g⁻¹. In our previous studies [77], three types of GNSs with varying size, edge sites, defects and number of layers have been successfully created. It was demonstrated that controlled morphologies and microstructures of GNSs have important effects on the cyclic performance and rate capability of LIBs. Meanwhile, the specific functional groups can further enhance the reversible capacity of reduced graphene [78]. The capacity at potentials greater than 1.5 V is predominantly attributed to phenol groups, while the capacity of the redox couple at 1.1 V results from cyclic edge ether groups.

Nitrogen (N) or boron (B) as a dopant and/or N/B-containing functional groups was employed to enhance the GNSs performance [79], [80], [81], [82], [83], [84], [85], [86], [87], [88]. For doped GNSs, the intense growth of the disorder-induced D band indicates that Li⁺ intercalation/de-intercalation into graphene sheets upon cycling brings about a more noticeable change in the degree of long-range ordering in the hexagonal lattice than that of GNSs [88]. H.M. Cheng et al. [89] reported an electrode made with heteroatom (N, B)-doped chemically derived GNSs that exhibited extremely high rate performance and large capacity. The doped GNSs showed a high reversible capacity of >1040 mAh g⁻¹. In another study, X.L. Feng's group [89], [90] developed a facile approach to synthesize free-standing fully fluorinated GNSs, which have high capacity, high rate capability and stable cycling performance. The effects of the electron-deficient N-doped GNSs on lithium storage were investigated via first-principles calculations [91], and the enhanced reversible capacity was attributed to the N-doped system. It was concluded that the unique 2D structure, disordered surface morphology, heteroatom defects, better electrode/electrolyte wettability, increased inter-sheet distance, and improved electrical conductivity of the doped graphene are beneficial to Li-storage performances and thus show significant advantages for LIB anodes in comparison to those of pristine chemically derived graphene and other carbonaceous materials [89], [92].

The complex and interesting folded structures of GNSs have attracted much attention, since the folding of a structure can change its form and functionality, which may induce new and distinct properties in GNSs [93], [94], [95]. Thus, various flexible graphene papers (GPs) have been designed. Vacuum filtration is the mostly widely used approach for the development of GPs [96] and the flexible structure of GPs makes them potential anodes for free standing batteries. In our previous research, we studied in detail the influence of GP thickness on its electrochemical performance [97]. We concluded that the capacity decline with the increase of GP thicknesses is associated with the dense restacking of GNSs and a large aspect ratio of GPs. The effective Li⁺ diffusion distance in GPs is mainly controlled by the thicknesses. The diffusion proceeds mainly in the in-plane direction, while cross-plane diffusion is restrained. As such, the effective contact of GNSs with electrolyte is limited and the efficiency of carbon utilization is very low in the thick GPs. However, the unique feature of graphene will be lost when the GNS are assembled into GPs, in which they are closely stacked. Thus, access of electrolyte to the GNS is restrained and consequently, a barrier for Li diffusion into the GNS is established, leading to lower specific capacities of GPs [97], [98]. In addition, the multi-layer graphene/single-walled carbon nanotube flexible free-standing film electrodes were prepared by a time-efficient microwave autoclave method [99]. The flexible film exhibited good charge capacity of about 303 mAh g⁻¹ after 50 cycles. The presence of carbon nanotubes in the electrodes, which forms a unique sandwich structure with GNSs, can provide efficient conductive pathways to improve the ionic conductivity, further enhancing the electrochemical properties of GPs. Thus, for the development of flexible GP-based anodes, pure GPs will not be adopted;the trend has been for researchers to develop complex films with Sn or Si nanomaterials to enhance the properties of GPs.

GNSs have great research interests as buffering materials for large volume change Sn-based anodes in LIBs, owing to their remarkable features: (i) the high flexibility of graphene could be an excellent supporting matrix or coating layer. During the charge/discharge processes, the volume expansion and particle aggregation would be relieved by the buffering GNSs. (ii) The rich functional groups at the GNS surface can serve as an appealing 2D substrate for the anisotropic growth of various Sn-based materials [100]. (iii) The high intrinsic surface area and outstanding electrical conductivity of GNSs provide an ideal platform for the storage and transportation of lithium ions and electrons [100]. As a result, a lot of effort has been devoted to introducing GNSs into Sn-based anode materials, and a large number of papers have been published based on this idea [101], [102], [103], [104], [105], [106], [107], [108]. Sn-based anode materials for LIBs have many merits, such as low cost, high theoretical capacities, and high energy density. However, volume expansion/contraction limits further improvements to the anode performance. The flexibility and high conductivity of graphene makes it an excellent matrix material. The introduction of graphene into Sn-based anode materials addresses the problems of large volume changes, resulting in high reversible capacity, rate capability and structural stability of the composites. The importance strongly motivates us to review the significant effects of graphene as excellent matrix on large volume change Sn-based materials as potential anodes for LIBs. It is important to review the recent achievements and development of GNSs as an excellent matrix to address the challenges of large volume change Sn-based anodes in LIBs. To date, few review papers have focused on this topic in detail. In this review article, we mainly focus on how a matrix of 2D GNSs mitigates the problems of Sn-based anodes originating from large volume change upon cycling, including their preparation methods, properties and enhanced LIB performance. The important effects of GNS doping are also reviewed.