Recent Progress on Graphene-Based Nanocomposites for Electrochemical Sodium-Ion Storage

In advancing battery technologies, primary attention is paid to developing and optimizing low-cost electrode materials capable of fast reversible ion insertion and extraction with good cycling ability. Sodium-ion batteries stand out due to their inexpensive price and comparable operating principle to lithium-ion batteries. To achieve this target, various graphene-based nanocomposites fabricate strategies have been proposed to help realize the nanostructured electrode for high electrochemical performance sodium-ion batteries. In this review, the graphene-based nanocomposites were introduced according to the following main categories: graphene surface modification and doping, three-dimensional structured graphene, graphene coated on the surface of active materials, and the intercalation layer stacked graphene. Through one or more of the above strategies, graphene is compounded with active substances to prepare the nanocomposite electrode, which is applied as the anode or cathode to sodium-ion batteries. The recent research progress of graphene-based nanocomposites for SIBs is also summarized in this study based on the above categories, especially for nanocomposite fabricate methods, the structural characteristics of electrodes as well as the influence of graphene on the performance of the SIBs. In addition, the relevant mechanism is also within the scope of this discussion, such as synergistic effect of graphene with active substances, the insertion/deintercalation process of sodium ions in different kinds of nanocomposites, and electrochemical reaction mechanism in the energy storage. At the end of this study, a series of strategies are summarized to address the challenges of graphene-based nanocomposites and several critical research prospects of SIBs that provide insights for future investigations.


Introduction
Lithium-ion batteries (LIBs) are a kind of high-specific energy secondary battery that are important and widely used in commercial applications nowadays. However, due to the limitation of lithium resources, the cost of lithium-ion batteries has greatly increased recently, which will become a serious constraint for its future massive applications [1]. Therefore, sodium ion batteries (SIBs), have many advantages such as rich resources, lowcost, high-energy conversion efficiency, no over discharge characteristics, compatible with aluminum foil that make it have great application potential in new energy field [2,3]. In addition to partially replacing the highly polluting lead-acid batteries, the most significant potential application of sodium-ion batteries can also focus on balancing the peak and valley of electricity from the wind or solar power, as shown in Figure 1 [4]. People started the research on SIBs as early as the 1970s and it has witnessed a spurt of growth with the development of material science in the recent year [5]. The working principle of SIBs is similar to that of LIBs [6,7]. Its internal structure mainly contains four parts: Positive, negative, electrolyte, separator. Normally, both anode and cathode electrodes are composed The working principle of sodium-ion battery is schematically shown in Figure 2. In essence, sodium-ion battery is an electrode with a concentration difference in ion concentration. In the process of charging the battery, the sodium ions are detached from the positive electrode, pass through the electrolyte, the separator, and transferred to the negative electrode material, the same time the electrons generated by the formation of sodium ions are passed from the outer circuit to the negative electrode, forming a closed circuit. During the discharge of the battery, the sodium ions are detached from the negative electrode and transferred to the positive through the electrolyte and separator, the electrons are also re-transferred from the negative electrode to the positive electrode of the battery through the external circuit, forming a closed loop road, to achieve charge conservation of positive and negative electrodes [8]. However, compared with lithium ion, sodium ion has a stronger solvation with the size of 1.02 Å (0.76 Å for the lithium ion) and the electrochemical potential of sodium-ion (−2.71 V, Na + /Na) is higher than that of lithium ion (−3.04 V, Li + /Li). Therefore, the electrode material of sodium-ion batteries will produce more remarkable volume changes in the charge and discharge process, resulting in more severe problems, including collapse and crushing of electrode structure. At the same time, from the perspective of dynamics, the insertion and extraction speed of larger sodium ions in electrode materials is slower than that of lithium ions, which makes sodium-ion batteries have more significant challenges in cycle life and rate capability performance [9,10]. electrodes are composed of active material powders, adhesives and conductive additives in proportion and separated using a porous separator usually made of glass fibers or porous polymers that allows ions to pass through but prevents electrons that achieve short circuit prevention [6]. The working principle of sodium-ion battery is schematically shown in Figure 2. In essence, sodium-ion battery is an electrode with a concentration difference in ion concentration. In the process of charging the battery, the sodium ions are detached from the positive electrode, pass through the electrolyte, the separator, and transferred to the negative electrode material, the same time the electrons generated by the formation of sodium ions are passed from the outer circuit to the negative electrode, forming a closed circuit. During the discharge of the battery, the sodium ions are detached from the negative electrode and transferred to the positive through the electrolyte and separator, the electrons are also re-transferred from the negative electrode to the positive electrode of the battery through the external circuit, forming a closed loop road, to achieve charge conservation of positive and negative electrodes [8]. However, compared with lithium ion, sodium ion has a stronger solvation with the size of 1.02 Å (0.76 Å for the lithium ion) and the electrochemical potential of sodium-ion (−2.71 V, Na + /Na) is higher than that of lithium ion (−3.04 V, Li + /Li). Therefore, the electrode material of sodium-ion batteries will produce more remarkable volume changes in the charge and discharge process, resulting in more severe problems, including collapse and crushing of electrode structure. At the same time, from the perspective of dynamics, the insertion and extraction speed of larger sodium ions in electrode materials is slower than that of lithium ions, which makes sodium-ion batteries have more significant challenges in cycle life and rate capability performance [9,10]. and good compatibility with other materials, graphene and its derived structures, such as graphene aerogel, stacked graphene layers and graphene foam have been used as the electrode for SIBs [47]. However, the pristine graphene has certain limitations in energy storage and insertion/deintercalation of sodium ions process because it does not provide sufficient active sites and the ideal amount of functional groups for enough interaction with sodium ions [31,48,49].
To solve the above problems, compatible the above-introduced nanoscale active materials with graphene is a feasible way to simultaneously stimulate the storage of energy properties of both active and carbon materials, thus ensuring fast ion/electron transfer and utilization rate of the active substance with excellent conductivity. More importantly, graphene functions as physical barrier that further buffers volume expansion and contraction of nanomaterials during the sodiation/desodiation. In addition, the synergistic effect confinements between graphene and nanoscale active materials can efficiently suppress the aggregation of both the energy storage nanomaterials and graphene during the process of material synthesis and energy storage process [18]. For instance, Ma, J.Y. et al., constructed the nanocomposite of TiO 2 /MoS 2 nanoflake linking with reduced graphene oxide (TiO 2 /MoS 2 /RGO). Compared with MoS 2 (200 mA h g −1 at 5 A g −1 ) and TiO 2 /MoS 2 (250 mA h g −1 at 5 A g −1 ), the nanocomposite showed a high specific capacity of 310 mA h g −1 at a current density of 2 A g −1 and ultrahigh structural integrity after 350 cycles [50]. The synergistic effect among RGO, MoS 2 and TiO 2 provides effective ion diffusion and electron transfer spaces and promotes insertion/deintercalation of Na ions that lead to the high performance of SIBs [50].
Therefore, to achieve the practical applications of sodium ion storage, various graphenebased nanocomposites fabricate stargates have been explored and can be mainly divided into the following categories according to the characteristics of the nanocomposites' structure: graphene surface modification and doping [51,52], three-dimensional structured graphene [53], graphene coated on the surface of active materials [54], the intercalation stacked two-dimensional graphene [55,56], as shown in Figure 3. Through one or more of the above strategies, graphene is compounded with active substances to prepare the nanocomposite electrode, which is applied as the anode or cathode for SIBs. In this study, the progress of graphene-based nanocomposites for SIBs is reviewed, especially for nanocomposites fabricate methods, the structural characteristics of nanocomposites as well as the influence of graphene on the performance of the SIBs based on the above categories. In addition, the relevant mechanism is also within the scope of discussion, such as the Na + storage process in different kinds of nanocomposites, the synergistic effect of graphene and active substances, electrochemical characteristics in the energy storage process. At the end of the overview, a series of strategies are summarized to address the challenges of SIBs and several possible research directions on graphene-based nanocomposites that provide insights for future investigations. The electrochemical performance of graphene based MoSe2 SIBs were assessed fr 0.01 to 3.0 V. As shown in Figure 5a, MoSe2@5%rGO as a representative for cy voltammetry (CV) tests to investigate the electrochemical reversibility and redox proc for SIBs. During the initial CV process, five prominent peaks can be observed, amo which the peaks at 1.88, 1.35, and 0.80 V corresponding to the Na + ions insertion into layer of MoSe2 to produce NaxMoSe2, which is further transformed into Mo and Na [66]. The peaks at 1.07 and 0.46 V correspond to solid state interphase (SEI) film format and electrolyte decomposition [66]. The anodic peaks from the first to third cycle and cathodic peaks from the second to third cycle remain at almost the same voltag The electrochemical performance of graphene based MoSe 2 SIBs were assessed from 0.01 to 3.0 V. As shown in Figure 5a, MoSe 2 @5%rGO as a representative for cyclic voltammetry (CV) tests to investigate the electrochemical reversibility and redox process for SIBs. During the initial CV process, five prominent peaks can be observed, among which the peaks at 1.88, 1.35, and 0.80 V corresponding to the Na + ions insertion into the layer of MoSe 2 to produce Na x MoSe 2 , which is further transformed into Mo and Na 2 Se [66]. The peaks at 1.07 and 0.46 V correspond to solid state interphase (SEI) film formation Nanomaterials 2022, 12, 2837 6 of 33 and electrolyte decomposition [66]. The anodic peaks from the first to third cycle and the cathodic peaks from the second to third cycle remain at almost the same voltages, indicating excellent chemical and structural stability of the lamellar composites [66]. Figure 5b reflects the galvanostatic charge/discharge (GCD) curves of MoSe 2 @5% GO at 100 mA·g −1 in which the plateaus match well with the CV results and the GCD curves are well overlapping after the first GCD test that manifests the reversible electrochemical properties. The first discharge/charge specific capacity of MoSe 2 @5%rGO can be achieved to 712.7/458.3 mAh g −1 with the Coulombic efficiency of 64.3%, which is better than the specific capacity of MoSe 2 (622.3/338.1 mAh g −1 ), MoSe 2 @1%rGO (798.9/495.8 mAh g −1 ) and MoSe 2 @10%rGO (716.9/355.2 mAh·g −1 ). It also indicates that the composite materials possess highly reversible specific capacities and a good cyclic stability. Figure 5c depicts that the MoSe 2 @5%rGO achieved the largest retention value of 383.6 mAh·g −1 after 50 cycles, which is better than the cycling stability of other rGO content. Compared with the other rGO mass content radio in Figure 5d, under the current density of 200, 500, 1000, 2000, and 5000 mA g −1 , the MoSe 2 @5%rGO shows the best rate capability with the capacities of 447.4, 390.0, 348.5, 306.9, and 251.3 mAh g −1 , respectively. In addition, after the above GCD rate capability tests, the MoSe 2 @5%rGO maintained an excellent reversible capacity of 435.3 mAh g −1 when the current density came back to 100 mA g −1 . Therefore, MoSe 2 @5%rGO exhibits the optimal Na-ion storage performances by taking advantage of great kinetics and alleviating volume expansion effect from graphene and high capacity from MoSe 2 [65].   [56]. The nanocomposites prepared through assisted method have some advantages, such as low cost, Fe2O3 comb graphene and high preparation efficiency, which makes it possible to prod on large scale [56]. Figure 6a,b clearly show that Fe2O3 nanoparticles with a d 30-80 nm is uniformly distributed on the interface of SNRGO that prevent t  [56]. The nanocomposites prepared through the microwaveassisted method have some advantages, such as low cost, Fe 2 O 3 combined closely with graphene and high preparation efficiency, which makes it possible to produce the electrode on large scale [56]. Figure 6a,b clearly show that Fe 2 O 3 nanoparticles with a diameter of around 30-80 nm is uniformly distributed on the interface of SNRGO that prevent the agglomeration of Fe 2 O 3 nanoparticles and further reducing the volume change of Fe 2 O 3 during the Na + insertion/extraction [67]. Consistent with SEM images, the TEM images of Fe 2 O 3 /SNRGO nanocomposite in Figure 6c,d confirmed that Fe 2 O 3 nanoparticles are deposited on the surface of SNRGO nanosheets tightly, which should be the main reason result in excellent cycle performance of SIBs.
microwave-assisted method [56]. The nanocomposites prepared through the mic assisted method have some advantages, such as low cost, Fe2O3 combined clos graphene and high preparation efficiency, which makes it possible to produce the on large scale [56]. Figure 6a,b clearly show that Fe2O3 nanoparticles with a diameter o 30-80 nm is uniformly distributed on the interface of SNRGO that prevent the agglo of Fe2O3 nanoparticles and further reducing the volume change of Fe2O3 during insertion/extraction [67]. Consistent with SEM images, the TEM images of Fe2O3 nanocomposite in Figure 6c,d confirmed that Fe2O3 nanoparticles are deposited on th of SNRGO nanosheets tightly, which should be the main reason result in excell performance of SIBs.  The electrochemical behavior, including CV, GCD, rate capability performance and cycling stability of Fe 2 O 3 /SNRGO electrode was investigated through a half cell configuration. The CV curves of Fe 2 O 3 /SNRGO were carried out with a voltage window of 0.01-3 V under the scanning rate of 0.1 mV s −1 in the initial four cycles (Figure 7a). The two irreversible reduction peaks of 0.5 V and 1.1 V at the CV curves in the first negative scan of Fe 2 O 3 /SNRGO correspond to the irreversible formation of the SEI layer [68], meanwhile the reduction peak at about 0.01 V attributed to Fe 3+ →Fe 0 [69]. In the following three CV scans, the two reduction peaks have moved to 0.63 V and 0.92 V, and in the oxidation direction, two peaks can be found at 0.66 V and 1.4 V that correspond to the reversible redox reaction of Fe 3+ ↔Fe 2+ and Fe 2+ ↔ Fe 0 [70,71]. Moreover, the CV curves are well overlapping after the first CV cycle that manifests the reversible electrochemical behavior of the Fe 2 O 3 /SNRGO electrode. The typical GCD curves of 1st, 10th, 20th, 30th, 40th and 50th of Fe 2 O 3 /SNRGO based half-cell can be found in Figure 7b. In the first GCD curve, the as-prepared sample manifests a large irreversible capacity with the Coulombic efficiency of only 29.0%, which can be explained as the formation of the SEI layer due to the irreversible insertion of Na + into the lattice [72]. It is worth noting that no obvious discharge platform has been found in the GCD curve, which is consistent with the voltage trend of the other typical Fe 2 O 3 based reports [67,72,73].
curves of 1st, 10th, 20th, 30th, 40th and 50th of Fe2O3/SNRGO based half-cell can be found in Figure 7b. In the first GCD curve, the as-prepared sample manifests a large irreversible capacity with the Coulombic efficiency of only 29.0%, which can be explained as the formation of the SEI layer due to the irreversible insertion of Na + into the lattice [72]. It is worth noting that no obvious discharge platform has been found in the GCD curve, which is consistent with the voltage trend of the other typical Fe2O3 based reports [67,72,73]. Reproduced with permission [56]. Copyright 2020, Elsevier.

The Three-Dimensional Structured Graphene for SIBs
Recently, three-dimensional structured graphene as the current collector [57] compatible with the energy storage active materials [74] were used as SIBs electrode that demonstrates excellent electrochemical performance due to the synergistic effect and large surface area. The traditional process is common and uses two-dimensional materials such as copper foil and aluminum foil as the substrate, on which the active substances are coated for SIBs electrodes. Thus, in order to bind the active substances tightly on the twodimensional foil surface, the active substance is first mixed with carbon material and binder to make slurry, and then coated on the foil [75]. Therefore, it is necessary to build a three-dimensional structure on which the active material can be firmly deposited by physical or chemical methods without the usage of binder to form the electrodes for rapid ions and electrons transportation. Due to the plasticity of graphene, the 3D structured architecture of graphene has been designed to reduce the restacking of graphene sheets during the electrode fabrication and electrochemical process and keep the large contact area with the electrolyte for enhancement of Na + storage capacity with excellent cycling stability [76]. In addition, additive-free composite not only can enhance conductivity, simplify the preparation process, reduce the cost [77], but also can meet the requirements of large specific surface area for energy storage, higher capacity, and mechanical strength [78]. Reproduced with permission [56]. Copyright 2020, Elsevier.

The Three-Dimensional Structured Graphene for SIBs
Recently, three-dimensional structured graphene as the current collector [57] compatible with the energy storage active materials [74] were used as SIBs electrode that demonstrates excellent electrochemical performance due to the synergistic effect and large surface area. The traditional process is common and uses two-dimensional materials such as copper foil and aluminum foil as the substrate, on which the active substances are coated for SIBs electrodes. Thus, in order to bind the active substances tightly on the two-dimensional foil surface, the active substance is first mixed with carbon material and binder to make slurry, and then coated on the foil [75]. Therefore, it is necessary to build a three-dimensional structure on which the active material can be firmly deposited by physical or chemical methods without the usage of binder to form the electrodes for rapid ions and electrons transportation. Due to the plasticity of graphene, the 3D structured architecture of graphene has been designed to reduce the restacking of graphene sheets during the electrode fabrication and electrochemical process and keep the large contact area with the electrolyte for enhancement of Na + storage capacity with excellent cycling stability [76]. In addition, additive-free composite not only can enhance conductivity, simplify the preparation process, reduce the cost [77], but also can meet the requirements of large specific surface area for energy storage, higher capacity, and mechanical strength [78].
Therefore, an additive-free composite of the energy storage substances with a large specific surface of three-dimensional graphene current collectors for sodium-ion storage is an ideal method to realize the next generation of high-performance SIBs with higher capacity and mechanical intensity [57]. Figure 8a depicted the fabrication procedure of hierarchical periodic thickness adjustable microlattice with micrometer pores and submillimeter filaments/channels carried out by Yan, J. et al., through a simple 3D printing method using reduced graphene oxide/carbon nanotube (rGO/CNT) aerogel [47]. As shown in Figure 8b-d, the 3D rGO/CNT microlattice aerogel has a strong electrical conductivity and abundant active nucleation sites for Na plating/stripping with a reduced current density, resulting in uniform sodium deposition to overcome the issues of dendrite. Moreover, the ion transportation rate also greatly improved due to the ordered microchannels that existed in the artificial periodic microlattice that facilitate the electrolyte permeation into the microlattice and promote the contact between active substance and electrolyte. Intrinsically, the CNTs were stable scaffolds by the rGO that improved the surface kinetics and overall conductivity synergistically, further improve the distribution of energy storage substances uniformity as well as the overall performance of the composite electrodes. microchannels that existed in the artificial periodic microlattice that facilitate permeation into the microlattice and promote the contact between active electrolyte. Intrinsically, the CNTs were stable scaffolds by the rGO that impro kinetics and overall conductivity synergistically, further improve the distribu storage substances uniformity as well as the overall performance of the compo To evaluate the practical performance of rGO/CNTs microlattice aerog metal were electrodeposition on the surface of rGO/CNT as the anode (Na@r Na3V2(PO4)3 mixed with carbon on rGO (NVP@C-rGO) as the cathode were as full sodium ion battery (Figure 9a). Figure 9b depicted that under the current mA g −1 , the NVP@C-rGO cathode achieved a charge/discharge capacity of 98.8 The performance of SIBs was also evaluated within the potential window of 2.0 charge/discharge capacities of the full battery are calculated to be 101.3 mA h g at the 7th cycle (Figure 9c,d). After the active process of the first five cycles at cycling performance was carried out at 100 mA g −1 and delivered good cyclic s 100 cycles. The long-term cycling stability of the Na@rGO/CNT//NVP demonstrate that the 3D printed rGO/CNT microlattice has a great potenti application. To evaluate the practical performance of rGO/CNTs microlattice aerogel for SIBs, Na metal were electrodeposition on the surface of rGO/CNT as the anode (Na@rGO/CNT) and Na 3 V 2 (PO 4 ) 3 mixed with carbon on rGO (NVP@C-rGO) as the cathode were assembled into a full sodium ion battery (Figure 9a). Figure 9b depicted that under the current density of 100 mA g −1 , the NVP@C-rGO cathode achieved a charge/discharge capacity of 98.8/91.1 mA h g −1 . The performance of SIBs was also evaluated within the potential window of 2.0-3.8 V and the charge/discharge capacities of the full battery are calculated to be 101.3 mA h g −1 /93.1 mA h g −1 at the 7th cycle (Figure 9c,d). After the active process of the first five cycles at 20 mA g −1 , the cycling performance was carried out at 100 mA g −1 and delivered good cyclic stability during 100 cycles. The long-term cycling stability of the Na@rGO/CNT//NVP@C-rGO SIBs demonstrate that the 3D printed rGO/CNT microlattice has a great potential for practical application.
Wu, D.J. et al., reported the multi-layer nanostructured graphene modified 3D macroporous electrically conductive network (n-MECN) as the sodium ion battery substrate with superior electrochemical performance [24,79,80]. As described in the previous report and illustrated in Figure 10, the multilayer nanographene was prepared on the surface and inter-channel of MECN by solvothermal carburization method, from which the ordered porous-rich electrode substrate of n-MECN was realized [81]. In addition, then the selfbranching MnCo 2 O 4 was hydrothermally in situ growth on the surface and inter-channel of n-MECE at 180 • C to fabricate the MnCo 2 O 4 @n-MECN electrode. The special ordered structure with micro-channel of n-MECN not only acts as the skeleton in the electrochemical energy storage process and facilitate the transfer of electrolyte but is also compatible with the microelectronic process for the micro energy storage devices. Wu, D.J. et al., reported the multi-layer nanostructured graphene modified macroporous electrically conductive network (n-MECN) as the sodium ion bat substrate with superior electrochemical performance [24,79,80]. As described in previous report and illustrated in Figure 10, the multilayer nanographene was prepa on the surface and inter-channel of MECN by solvothermal carburization method, f which the ordered porous-rich electrode substrate of n-MECN was realized [81] addition, then the self-branching MnCo2O4 was hydrothermally in situ growth on surface and inter-channel of n-MECE at 180 °C to fabricate the MnCo2O4@n-ME electrode. The special ordered structure with micro-channel of n-MECN not only act the skeleton in the electrochemical energy storage process and facilitate the transfe electrolyte but is also compatible with the microelectronic process for the micro ene storage devices.   Wu, D.J. et al., reported the multi-layer nanostructured graphene modified 3 macroporous electrically conductive network (n-MECN) as the sodium ion batter substrate with superior electrochemical performance [24,79,80]. As described in th previous report and illustrated in Figure 10, the multilayer nanographene was prepare on the surface and inter-channel of MECN by solvothermal carburization method, fro which the ordered porous-rich electrode substrate of n-MECN was realized [81]. addition, then the self-branching MnCo2O4 was hydrothermally in situ growth on th surface and inter-channel of n-MECE at 180 °C to fabricate the MnCo2O4@n-MEC electrode. The special ordered structure with micro-channel of n-MECN not only acts the skeleton in the electrochemical energy storage process and facilitate the transfer electrolyte but is also compatible with the microelectronic process for the micro energ storage devices.  electronic conductivity substrate. The 3D interconnected structure of n-MECN inherited from MECN improved the electronic conductivity and the specific surface area to facilitate electrochemical reactions. The self-branching MnCo2O4 nano-architecture is incorporated into the surface (Figure 11c,d) and cross-sectional (Figure 11e,f) of n-MECN with a diameter around 50 nm, indicating that the MnCo2O4/n-MECN was successfully prepared. To evaluate the practical application of the materials and structure design strategy, the MnCo2O4@n-MECN served as the anode in SIBs. In the cathodic process of the first CV curves cycle shown in Figure 12a, the two reduction peaks can be found at 0.25 V and 0.87 V, which are related to the irreversible formation of the SEI layer and the process of Mn 2+ to Mn, Co 3+ to Co 2+ , Co 2+ to Co [82]. The two broad peaks related to metallic Mn to Mn 2+ and Co to Co 2+ , Co 2+ to Co 3+ are anodic oxidation processes at 0.84 V and 1.59 V [83][84][85]. After the first CV cycles, the CV curves are well overlapping and the moves of the cathodic peak to 0.87 V related to the formation of the NaxMnCo2O4 phase, demonstrating reversible electrochemical behavior of MnCo2O4@n-MECN electrode [86]. Figure 12b shows the 1st, 2nd, 3rd, and 50th GCD curves of the sodium ion battery at 50 mA g −1 . A noticeable discharging plateau at ~0.45 V can be found in the GCD curves corresponding to the reduction of MnCo2O4 nanoarrays which is in accordance with the morphologies of CV curves [87]. The charging/discharging specific capacity calculate from the 1st GCD curve of MnCo2O4@n-MECN reaches 614.7/1120.0 mAh g −1 with the Coulombic efficiency of 75.1% [88,89]. Figure 12c reflected that MnCo2O4@n-MECN based SIBs show the specific capacities of 541.2, 200.5, 150.1, 110.0, and 70.5 mAh g −1 at the current densities of 50, 100, 200, 400, and 800 mA g −1 , respectively, and the reversible capacity is maintained when the current density back to 50 mA·g −1 . As shown in Figure 12d, the SIBs reflected a good cycling ability of 100 cycles under different current densities after 1st cycle as well as the morphology and structure of MnCo2O4@n-MECN in SIBs are maintained after 100 cycles. Figure 12e depicted that the b value of MnCo2O4@n-MECN based SIBs is 0.81 at the cathodic peak indicating that the energy storage mechanism of the SIBs involves both faradic and non-faradic processes and the capacitive-controlled process makes the main contribution to the energy storage characteristics of MnCo2O4@n-MECN. With the increase of the scan rates the contribution of capacitive-controlled gradually increased, when scan rates from 0.2 mv s −1 to 1.0 mv s −1 , the capacitive contribution improved from To evaluate the practical application of the materials and structure design strategy, the MnCo 2 O 4 @n-MECN served as the anode in SIBs. In the cathodic process of the first CV curves cycle shown in Figure 12a, the two reduction peaks can be found at 0.25 V and 0.87 V, which are related to the irreversible formation of the SEI layer and the process of Mn 2+ to Mn, Co 3+ to Co 2+ , Co 2+ to Co [82]. The two broad peaks related to metallic Mn to Mn 2+ and Co to Co 2+ , Co 2+ to Co 3+ are anodic oxidation processes at 0.84 V and 1.59 V [83][84][85]. After the first CV cycles, the CV curves are well overlapping and the moves of the cathodic peak to 0.87 V related to the formation of the Na x MnCo 2 O 4 phase, demonstrating reversible electrochemical behavior of MnCo 2 O 4 @n-MECN electrode [86]. Figure 12b shows the 1st, 2nd, 3rd, and 50th GCD curves of the sodium ion battery at 50 mA g −1 . A noticeable discharging plateau at~0.45 V can be found in the GCD curves corresponding to the reduction of MnCo 2 O 4 nanoarrays which is in accordance with the morphologies of CV curves [87]. The charging/discharging specific capacity calculate from the 1st GCD curve of MnCo 2 O 4 @n-MECN reaches 614.7/1120.0 mAh g −1 with the Coulombic efficiency of 75.1% [88,89]. Figure 12c reflected that MnCo 2 O 4 @n-MECN based SIBs show the specific capacities of 541.2, 200.5, 150.1, 110.0, and 70.5 mAh g −1 at the current densities of 50, 100, 200, 400, and 800 mA g −1 , respectively, and the reversible capacity is maintained when the current density back to 50 mA·g −1 . As shown in Figure 12d, the SIBs reflected a good cycling ability of 100 cycles under different current densities after 1st cycle as well as the morphology and structure of MnCo 2 O 4 @n-MECN in SIBs are maintained after 100 cycles. Figure 12e depicted that the b value of MnCo 2 O 4 @n-MECN based SIBs is 0.81 at the cathodic peak indicating that the energy storage mechanism of the SIBs involves both faradic and non-faradic processes and the capacitive-controlled process makes the main contribution to the energy storage characteristics of MnCo 2 O 4 @n-MECN. With the increase of the scan rates the contribution of capacitive-controlled gradually increased, when scan rates from 0.2 mv s −1 to 1.0 mv s −1 , the capacitive contribution improved from 52% to 73%, and when the scanning speed is greater than 0.8 mv s −1 , the capacitance contribution rate hovers around 70% (Figure 12f). This study fully demonstrates that the three-dimensional structured graphene exhibits excellent electrochemical properties, optimizing ion and electron transport, and improves the utilization of active material so as to increase the rate capability ability of the SIBs [85][86][87].
anomaterials 2022, 12, x FOR PEER REVIEW 12 of 32 52% to 73%, and when the scanning speed is greater than 0.8 mv s −1 , the capacitance contribution rate hovers around 70% (Figure 12f). This study fully demonstrates that the three-dimensional structured graphene exhibits excellent electrochemical properties, optimizing ion and electron transport, and improves the utilization of active material so as to increase the rate capability ability of the SIBs [85][86][87]. Relationship between log scanning rates and log cathodic peak currents; (f) Normalized contribution ratio at different scan rates. Reproduced with permission [79]. Copyright 2020, Elsevier.
Jo, M. S. et al. described an electrode of graphene-based nanofibers entangled with bamboo-shaped N-CNTs containing CoSe2 nanocrystals at each node (N-CNT/rGO/CoSe2 NF) through the heat treatment process of electrospinning nanofibers [90]. Figure 13 describes the detailed formation mechanism and fabrication steps of this nanostructured electrode. An electrospinning process ( Figure 13(a①)) was used to fabricate the precursor nanofibers that were composed of graphene oxide (GO), Co salt, poly (vinyl alcohol) (PVA), and PS nanobeads (40 nm). Next, the highly integrated structure of mesopores rGO nanosheets mixed with carbon-covered Co nanometals structure (Figure 13(a②)) were formed through the initial step of the first heat treatment of the precursor nanofibers with DCDA at 400 °C under N2 atmosphere. This step initially produced Co nanometals surrounding with carbon formed by PVA decomposition that prevents the aggregation of Co nanometals. Meanwhile, during the initial step of the first heat treatment step, the uniform mesopores between the GO sheets were formed due to the decomposition of PS nanobeads distributed in the GO matrix and the GO matrix was reduced to rGO at the same time. The uniform-sized Co nanoparticles were uniformly formed in the structured mesopores when the reducing gas penetrates into the depth of the structure from the mesopores. (e) Relationship between log scanning rates and log cathodic peak currents; (f) Normalized contribution ratio at different scan rates. Reproduced with permission [79]. Copyright 2020, Elsevier.
Jo, M.S. et al. described an electrode of graphene-based nanofibers entangled with bamboo-shaped N-CNTs containing CoSe 2 nanocrystals at each node (N-CNT/rGO/CoSe 2 NF) through the heat treatment process of electrospinning nanofibers [90]. Figure 13 describes the detailed formation mechanism and fabrication steps of this nanostructured electrode. An electrospinning process ( Figure 13(a 1 )) was used to fabricate the precursor nanofibers that were composed of graphene oxide (GO), Co salt, poly (vinyl alcohol) (PVA), and PS nanobeads (40 nm). Next, the highly integrated structure of mesopores rGO nanosheets mixed with carbon-covered Co nanometals structure ( Figure 13(a 2 )) were formed through the initial step of the first heat treatment of the precursor nanofibers with DCDA at 400 • C under N 2 atmosphere. This step initially produced Co nanometals surrounding with carbon formed by PVA decomposition that prevents the aggregation of Co nanometals. Meanwhile, during the initial step of the first heat treatment step, the uniform mesopores between the GO sheets were formed due to the decomposition of PS nanobeads distributed in the GO matrix and the GO matrix was reduced to rGO at the same time. The uniform-sized Co nanoparticles were uniformly formed in the structured mesopores when the reducing gas penetrates into the depth of the structure from the mesopores.
As shown in Figure 13(a 2 ), the further step of the first heat treatment was carried out at 800 • C under the CH x and NH 3 gases atmosphere and the bamboo-like N-CNTs were grown in the inner and outer parts of the rGO matrixed nanofibers by using the metallic Co nanocrystals in the mesopores as catalysts via catalytic chemical vapor deposition [91]. Interestingly, N-CNTs formed by aggregation of the C and N atoms from CH x and NH 3 gases that adsorbed and dissolved on the surface of the Co nanoparticles and the Co nanoparticles move forward by the force arising from the growth of CNTs, so that Co nanoparticles were uniformly distributed at the nodes of bamboo-like N-CNTs. Since H 2 Se gas cannot penetrate the C layer of CNTs, it is difficult to convert the CNTs coated Co nanoparticles into CoSe 2 directly by high temperature gas phase method, as shown in Figure 13b. Therefore, part of the CNT walls surrounding the Co nanoparticles were thermally etched during the second heat treatment step at 300 • C for 30 min in an air atmosphere ( Figure 13(a 3 )), although Co metal partially oxidized into Co 3 O 4 during this step, that was ignorable and inconsequential. Finally, in the third heat treatment step, Co and Co 3 O 4 nanoparticles inside the N-CNTs were selenization at 350 • C under H 2 Se gas atmosphere to form the CoSe 2 nanoparticles (Figure 13(a 4 )). Via the above steps, the electrode of graphene-based nanofibers entangled with bamboo-shaped N-CNTs containing CoSe 2 nanocrystals at each node (N-CNT/rGO/CoSe 2 NF) was successfully obtained. As shown in Figure 13b, selenization of Co nanocrystals captured in the walls of carbon nanotubes is very difficult because H 2 Se gas used for selenization cannot penetrate the thick carbon nanotube walls. Therefore, the gas diffusion process and CO phase change can be achieved by introducing the defect part into the CNT wall during the intermediate heat treatment to achieve the goal of successfully generating CoSe 2 . As shown in Figure 13(a②), the further step of the first heat treatment was carried out at 800 °C under the CHx and NH3 gases atmosphere and the bamboo-like N-CNTs were grown in the inner and outer parts of the rGO matrixed nanofibers by using the metallic Co nanocrystals in the mesopores as catalysts via catalytic chemical vapor deposition [91]. Interestingly, N-CNTs formed by aggregation of the C and N atoms from CHx and NH3 gases that adsorbed and dissolved on the surface of the Co nanoparticles and the Co nanoparticles move forward by the force arising from the growth of CNTs, so that Co nanoparticles were uniformly distributed at the nodes of bamboo-like N-CNTs. Since H2Se gas cannot penetrate the C layer of CNTs, it is difficult to convert the CNTs coated Co nanoparticles into CoSe2 directly by high temperature gas phase method, as shown in Figure 13b. Therefore, part of the CNT walls surrounding the Co nanoparticles were thermally etched during the second heat treatment step at 300 °C for 30 min in an As shown in Figure 14a-c, the FE-SEM image of N-CNT/rGO/CoSe 2 NF was wellmaintained throughout, even after multi-step fabrication. From the HR-TEM image of Figure 14d,e, the authors found that the diameters of the nodes in the bamboo-like N-CNTs were around 5-10 nm. The lattice spacings of 0.22 nm match well with the (111) crystal planes of CoSe 2 in N-CNTs (Figure 14e) [92] and the specific peaks of the XRD pattern of N-CNT/rGO/CoSe 2 NF match well with the PDF#53-0449 of CoSe 2 (Figure 14f) that indicating the complete selenization of Co and Co 3 O 4 into CoSe 2 [92,93]. Figure 14a-c, the FE-SEM image of N-CNT/rGO/CoSe2 NF was wellmaintained throughout, even after multi-step fabrication. From the HR-TEM image of Figure 14d,e, the authors found that the diameters of the nodes in the bamboo-like N-CNTs were around 5-10 nm. The lattice spacings of 0.22 nm match well with the (111) crystal planes of CoSe2 in N-CNTs (Figure 14e) [92] and the specific peaks of the XRD pattern of N-CNT/rGO/CoSe2 NF match well with the PDF#53-0449 of CoSe2 (Figure 14f) that indicating the complete selenization of Co and Co3O4 into CoSe2 [92,93]. In Figure 15, the electrochemical performance of the two electrodes of N-CNT/rGO/CoSe2 NF and CoSe2-filled NF as the anode for SIBs were compared using carbonate-based electrolytes. At a scanning rate of 0.1 mV s −1 , the initial five CV cycles of the two SIBs were carried out under the potential window of 0.001-3.0 V (Figure 14a,b). For the above two samples in the first cycle, the peak of ≈0.78 V in discharging process owing to the formation of metallic Co nanocrystals and Na2Se from the conversion reaction between the Na + and CoSe2 as well as the anodic peaks at 1.84 and 1.96 V owing to the restoration of metallic Co and Na2Se into CoSe2 nanocrystals in the charging process [90,92,94]. Moreover, after the first CV cycle, the insertion of Na + into CoSe2 corresponded to the cathodic peaks at 1.41 and 1.11 V, and due to the formation of ultrafine CoSe2 nanocrystals, the peak shift to higher potential from the second cycle [90,92,94]. Moreover, the conversion reaction with further Na + insertion was manifest at the cathodic peak of 0.7 V in the consequence CV cycles [90,94]. Figure 15 c,d show the GCD curves of the N-CNT/rGO/CoSe2 NF and bare CoSe2-filled NF based SIBs at a current density of 0.5 A g −1 for the 1st and 2nd cycles. Typically, anode materials with crystalline materials exhibit flat plateaus and ultrafine nanocrystals or amorphous phases exhibit sloped charge/discharge profiles [95,96]. In the first discharge curve, the N-CNT/rGO/CoSe2 NF manifested a comparatively short plateau at ≈0.89 V with a sloped charge/discharge profile due to the ultrafine crystallite size of the CoSe2. While the bare CoSe2-filled NF showed a lower and flatter plateau compared with CNT/rGO/CoSe2 NF composite electrode due to the crystalline materials of CoSe2 in CoSe2-filled NF. Bare CoSe2-filled NF shows the increase of resistance in the electrolyte due to the polarization from slow Na + ion diffusion, while In Figure 15, the electrochemical performance of the two electrodes of N-CNT/rGO/CoSe 2 NF and CoSe 2 -filled NF as the anode for SIBs were compared using carbonate-based electrolytes. At a scanning rate of 0.1 mV s −1 , the initial five CV cycles of the two SIBs were carried out under the potential window of 0.001-3.0 V (Figure 14a,b). For the above two samples in the first cycle, the peak of ≈0.78 V in discharging process owing to the formation of metallic Co nanocrystals and Na 2 Se from the conversion reaction between the Na + and CoSe 2 as well as the anodic peaks at 1.84 and 1.96 V owing to the restoration of metallic Co and Na 2 Se into CoSe 2 nanocrystals in the charging process [90,92,94]. Moreover, after the first CV cycle, the insertion of Na + into CoSe 2 corresponded to the cathodic peaks at 1.41 and 1.11 V, and due to the formation of ultrafine CoSe 2 nanocrystals, the peak shift to higher potential from the second cycle [90,92,94]. Moreover, the conversion reaction with further Na + insertion was manifest at the cathodic peak of 0.7 V in the consequence CV cycles [90,94]. Figure 15c,d show the GCD curves of the N-CNT/rGO/CoSe 2 NF and bare CoSe 2 -filled NF based SIBs at a current density of 0.5 A g −1 for the 1st and 2nd cycles. Typically, anode materials with crystalline materials exhibit flat plateaus and ultrafine nanocrystals or amorphous phases exhibit sloped charge/discharge profiles [95,96]. In the first discharge curve, the N-CNT/rGO/CoSe 2 NF manifested a comparatively short plateau at ≈0.89 V with a sloped charge/discharge profile due to the ultrafine crystallite size of the CoSe 2 . While the bare CoSe 2 -filled NF showed a lower and flatter plateau compared with CNT/rGO/CoSe 2 NF composite electrode due to the crystalline materials of CoSe 2 in CoSe 2 -filled NF. Bare CoSe 2 -filled NF shows the increase of resistance in the electrolyte due to the polarization from slow Na + ion diffusion, while the nanostructure of N-CNT/rGO/CoSe 2 NF has an increased Na + ion diffusion rate results a more easily Na + insertion/desertion reactions [97,98]. The initial discharge capacities of the N-CNT/rGO/CoSe 2 NF was 645 mA h g −1 with Coulombic efficiencies of 72.4%, while the bare CoSe 2 -filled NF was 531 mA h g −1 with the Coulombic efficiencies of 83.5%, although the active materials in N-CNT/rGO/CoSe 2 NF were smaller than the bare one. the nanostructure of N-CNT/rGO/CoSe2 NF has an increased Na + ion diffusion rate a more easily Na + insertion/desertion reactions [97,98]. The initial discharge capa the N-CNT/rGO/CoSe2 NF was 645 mA h g −1 with Coulombic efficiencies of 72.4% the bare CoSe2-filled NF was 531 mA h g −1 with the Coulombic efficiencies of although the active materials in N-CNT/rGO/CoSe2 NF were smaller than the bar In order to investigate the long performance of the SIBs based on N-CNT/rGO/C and bare CoSe2-filled NF, the cycling performance was carried out at a current densi A g −1 (Figure 15e). The bare CoSe2 based SIBs exhibited a significate capacity fading t 34 mA h g −1 after the 100th cycle because of volume change and structural destruction during repeated insertion/extractions of Na + ion. In contrast, the N-CNT/rGO/CoSe2 N SIBs showed the discharge capacities of 512 and 400 mA h g −1 at 2nd and 100th indicating the constant cycling ability up to 100th cycles. Due to the involvement of gr the vertical interconnection between graphene sheets and N-CNTs embedded wit nanoparticles both restrained the aggregation and buffered the volume expansion o nanoparticles during the cycling. To investigate the rate capacity of the nanocomp current densities of 0.5, 1, 2, 5, 10, and 20 A g −1 , the N-CNT/rGO/CoSe2 NF achie discharge capacities of 516, 470, 420, 341, 318, and 215 mA h g −1 , respectively (Figu Furthermore, the discharge capacity was maintained at 512 mA h g −1 when the current back to 0.5 A g −1 after 60 cycles. In contrast, as the increased of current density from 0. A g −1 , the discharge capacity of the bare CoSe2-filled NF decreased from 388 mA h g −1 t zero. In order to investigate the long performance of the SIBs based on N-CNT/rGO/CoSe 2 NF and bare CoSe 2 -filled NF, the cycling performance was carried out at a current density of 0.5 A g −1 (Figure 15e). The bare CoSe 2 based SIBs exhibited a significate capacity fading that only 34 mA h g −1 after the 100th cycle because of volume change and structural destruction of CoSe 2 during repeated insertion/extractions of Na + ion. In contrast, the N-CNT/rGO/CoSe 2 NF based SIBs showed the discharge capacities of 512 and 400 mA h g −1 at 2nd and 100th cycles, indicating the constant cycling ability up to 100th cycles. Due to the involvement of graphene, the vertical interconnection between graphene sheets and N-CNTs embedded with CoSe 2 nanoparticles both restrained the aggregation and buffered the volume expansion of CoSe 2 nanoparticles during the cycling. To investigate the rate capacity of the nanocomposite, at current densities of 0.5, 1, 2, 5, 10, and 20 A g −1 , the N-CNT/rGO/CoSe 2 NF achieved the discharge capacities of 516, 470, 420, 341, 318, and 215 mA h g −1 , respectively (Figure 15f). Furthermore, the discharge capacity was maintained at 512 mA h g −1 when the current density back to 0.5 A g −1 after 60 cycles. In contrast, as the increased of current density from 0.5 to 20.0 A g −1 , the discharge capacity of the bare CoSe 2 -filled NF decreased from 388 mA h g −1 to almost zero.

Graphene Coated on the Surface of Active Materials
In recent years, nanomaterials uniformly encapsulated by graphene as the Na-ion battery electrodes with shortened diffusion lengths of electrons/sodium ions have been widely studied and the graphene network formed the shell enables a highly conductive matrix. The graphene shell also provides enough physical space for the volume expansion of coated nanoparticles and at the same time constrains the nanoparticles within the conductive graphene shell for stable working conditions [99]. As expected, the fabricated nanocomposites manifest outstanding electrochemical performance with both remarkable rate ability and long cycling ability. Additionally, the carbonaceous conductive graphene materials provide the pathways for electron transport that facilitate the reaction kinetics, increasing the Coulombic efficiency and stabilizing the electrode-electrolyte interfaces [58,100,101].
Tan, X.Y. et al., reported a nanocomposite based on Sn 4 P 3 nanoparticles coated with porous graphene tubes (GT) as the Sn 4 P 3 /GT for the anode of SIBs [58]. The graphene tube is composed of two layers of graphene, the inner layer is nitrogen doped hydrophilic graphene tube and the outer layer is un-doped hydrophobic graphene tubes that provide the accommodation for the volume expansion of the coated Sn 4 P 3 nanoparticles. N-doping not only increases the hydrophilicity contributes to the directional growth of nanomaterials and improves the electron transfer kinetic performance, but also the synergistic effect of N and graphene keeps the electrode structure stable and further improves the electrochemical performance. As illustrate in Figure 16, the graphene tubes are firstly dispersed in the aqueous solution containing tin precursor and the aqueous solution can penetrate into the hydrophilic layer of graphene tube easily. After that, the SnO 2 nanoparticles prepared in the inside of the graphene tubes by selective growth of hydrothermal method and then a phosphidation process was used to forming the Sn 4 P 3 /GT.

Graphene Coated on the Surface of Active Materials
In recent years, nanomaterials uniformly encapsulated by graphene as the Na-ion battery electrodes with shortened diffusion lengths of electrons/sodium ions have been widely studied and the graphene network formed the shell enables a highly conductiv matrix. The graphene shell also provides enough physical space for the volume expansion of coated nanoparticles and at the same time constrains the nanoparticles within th conductive graphene shell for stable working conditions [99]. As expected, the fabricated nanocomposites manifest outstanding electrochemical performance with both remarkabl rate ability and long cycling ability. Additionally, the carbonaceous conductive graphen materials provide the pathways for electron transport that facilitate the reaction kinetics increasing the Coulombic efficiency and stabilizing the electrode-electrolyte interface [58,100,101].
Tan, X.Y. et al., reported a nanocomposite based on Sn4P3 nanoparticles coated with porous graphene tubes (GT) as the Sn4P3/GT for the anode of SIBs [58]. The graphene tub is composed of two layers of graphene, the inner layer is nitrogen doped hydrophili graphene tube and the outer layer is un-doped hydrophobic graphene tubes that provid the accommodation for the volume expansion of the coated Sn4P3 nanoparticles. N-doping not only increases the hydrophilicity contributes to the directional growth o nanomaterials and improves the electron transfer kinetic performance, but also th synergistic effect of N and graphene keeps the electrode structure stable and furthe improves the electrochemical performance. As illustrate in Figure 16, the graphene tube are firstly dispersed in the aqueous solution containing tin precursor and the aqueou solution can penetrate into the hydrophilic layer of graphene tube easily. After that, th SnO2 nanoparticles prepared in the inside of the graphene tubes by selective growth o hydrothermal method and then a phosphidation process was used to forming th Sn4P3/GT. Figure 16. The fabrication steps the nanocomposite of Sn4P3/GT. The schematic illustration show that the SnO2 nanoparticles were firstly prepared in the inside of the graphene tube by selectiv growth of hydrothermal method and then a phosphidation process was used to form Sn4P3/GT; th following three diagrams depict the sodiation/desodiation process of Sn4P3. Reproduced with permission [58]. Copyright 2021, Wiley-VCH. Figure 17a depicted that in the voltage window of 0.01-2 V (versus Na/Na + ) at 0.5 A g −1 , the electrode of Sn4P3/GT shows a first GCD curve of charge/discharge capacities o 1002 mA h g −1 /722 mA h g −1 with a Coulombic efficiency of 72.2%. The formation of SE film is the main reason for the irreversible capacity during the first GCD cycle [102,103] To realize the real application of the Sn4P3 based nanocomposite electrode, optimization of electrolyte and further exploring SEI composition should be carried out to reduce th Figure 16. The fabrication steps the nanocomposite of Sn 4 P 3 /GT. The schematic illustration shows that the SnO 2 nanoparticles were firstly prepared in the inside of the graphene tube by selective growth of hydrothermal method and then a phosphidation process was used to form Sn 4 P 3 /GT; the following three diagrams depict the sodiation/desodiation process of Sn 4 P 3 . Reproduced with permission [58]. Copyright 2021, Wiley-VCH. Figure 17a depicted that in the voltage window of 0.01-2 V (versus Na/Na + ) at 0.5 A g −1 , the electrode of Sn 4 P 3 /GT shows a first GCD curve of charge/discharge capacities of 1002 mA h g −1 /722 mA h g −1 with a Coulombic efficiency of 72.2%. The formation of SEI film is the main reason for the irreversible capacity during the first GCD cycle [102,103]. To realize the real application of the Sn 4 P 3 based nanocomposite electrode, optimization of electrolyte and further exploring SEI composition should be carried out to reduce the initial irreversible loss for a higher Coulombic efficiency beyond 90% [104]. A capacity of 713 mA h g −1 has been realized after 500 cycles with well overlapping GCD curves, indicating the reversible electrochemical behavior of the Sn 4 P 3 /GT electrode. initial irreversible loss for a higher Coulombic efficiency beyond 90% [104]. A capacity of 713 mA h g −1 has been realized after 500 cycles with well overlapping GCD curves, indicating the reversible electrochemical behavior of the Sn4P3/GT electrode. Figure 17. (a) GCD curves of Sn4P3/GT for the 1st, 2nd, 5th, 50th, 250th, and 500th cycles with the voltage window of 0.01-2 V versus Na/Na + at 0.5 A g −1 ; the comparison of electrochemical properties between the electrode of Sn4P3/GT and bare Sn4P3: (b) GCD cycling tests at 0.5 A g −1 , (c) rate capability under different current densities for 500 cycles, (d) Nyquist plot, Inset of (d) shows the relationship between ω -1/2 (where ω is the angular frequency in the low-frequency region, ω = 2πf) and the real part of the impedance spectra (Zre), (e) Utilization radio of the active material at different GCD current densities. Reproduced with permission [58]. Copyright 2021, Wiley-VCH. Figure 17c depicted that the Sn4P3/GT electrode showed a good rate performance with the discharge capacities of 821, 722, 636, 547, 435, 386 and 326 mA g h −1 at current densities of 0.2, 0.5, 1, 2, 5, 10 and 20 A g −1 , respectively. Additionally, the discharge capacity is maintained of 808 mA h g −1 when the current density is back to 0.5 A g −1 that indicating the recoverability of the electrode. In contrast, the discharge capacity decreased from ~650 mA h g −1 to almost zero of the bare Sn4P3 as the current density increased from 0.2 to 20.0 A g −1 . The electrochemical impedance spectroscopy (EIS) was employed to further explore the performance of Sn4P3/GT and bare Sn4P3 electrode with the frequency from 0.01 Hz to 10 MHz and a perturbation amplitude of 10 mV. From the Nyquist plot of Figure 17d, the Sn4P3/GT electrode shows a much smaller diameter of semicircle compare with bare Sn4P3 electrode which indicate that the Sn4P3/GT composite electrode has a lower interfacial resistance as well as better kinetics and higher ion-diffusion rate during the process of sodiation/desodiation [105,106]. The active material utilization rate versus charge-discharge current density of the Sn4P3/GT and bare Sn4P3 electrode were investigated. As shown in Figure 17e, the utilization rate of the active material in Sn4P3/GT decreases with the current density, which is 88%, 77%, and 67% at 0.5 A g −1 , 1 A g −1 and 2 A g −1 , respectively. In contrast, the bare Sn4P3 electrode shows the active material utilization rate of 79%, 28%, and 3% at 0.5 A g −1 , 1 A g −1 and 2 A g −1 , respectively. All of Figure 17. (a) GCD curves of Sn 4 P 3 /GT for the 1st, 2nd, 5th, 50th, 250th, and 500th cycles with the voltage window of 0.01-2 V versus Na/Na + at 0.5 A g −1 ; the comparison of electrochemical properties between the electrode of Sn 4 P 3 /GT and bare Sn 4 P 3 : (b) GCD cycling tests at 0.5 A g −1 , (c) rate capability under different current densities for 500 cycles, (d) Nyquist plot, Inset of (d) shows the relationship between ω -1/2 (where ω is the angular frequency in the low-frequency region, ω = 2πf) and the real part of the impedance spectra (Z re ), (e) Utilization radio of the active material at different GCD current densities. Reproduced with permission [58]. Copyright 2021, Wiley-VCH. Figure 17c depicted that the Sn 4 P 3 /GT electrode showed a good rate performance with the discharge capacities of 821, 722, 636, 547, 435, 386 and 326 mA g h −1 at current densities of 0.2, 0.5, 1, 2, 5, 10 and 20 A g −1 , respectively. Additionally, the discharge capacity is maintained of 808 mA h g −1 when the current density is back to 0.5 A g −1 that indicating the recoverability of the electrode. In contrast, the discharge capacity decreased from~650 mA h g −1 to almost zero of the bare Sn 4 P 3 as the current density increased from 0.2 to 20.0 A g −1 . The electrochemical impedance spectroscopy (EIS) was employed to further explore the performance of Sn 4 P 3 /GT and bare Sn 4 P 3 electrode with the frequency from 0.01 Hz to 10 MHz and a perturbation amplitude of 10 mV. From the Nyquist plot of Figure 17d, the Sn 4 P 3 /GT electrode shows a much smaller diameter of semicircle compare with bare Sn 4 P 3 electrode which indicate that the Sn 4 P 3 /GT composite electrode has a lower interfacial resistance as well as better kinetics and higher ion-diffusion rate during the process of sodiation/desodiation [105,106]. The active material utilization rate versus charge-discharge current density of the Sn 4 P 3 /GT and bare Sn 4 P 3 electrode were investigated. As shown in Figure 17e, the utilization rate of the active material in Sn 4 P 3 /GT decreases with the current density, which is 88%, 77%, and 67% at 0.5 A g −1 , 1 A g −1 and 2 A g −1 , respectively. In contrast, the bare Sn 4 P 3 electrode shows the active material utilization rate of 79%, 28%, and 3% at 0.5 A g −1 , 1 A g −1 and 2 A g −1 , respectively. All of the above electrochemical properties indicate that the overall performance of Sn 4 P 3 /GT was significantly improved by the introduction of graphene tubes on the surface of Sn 4 P 3 nanoparticles.
As shown in the outline in Figure 18a, the graphene-coated Na 2 . 4 Fe 1 . 8 (SO 4 ) 3 (denoted as NFS) nanograins (NFS@rGO microspheres) was successfully fabricated by Fang Y.J. et al., the 3D graphene microsphere network via a spray-drying route [99]. From the SEM images of Figure 18b-d, the Na 2 . 4 Fe 1 . 8 (SO 4 ) 3 nanoparticles are homogeneously distributed between layers of 3D spherical graphene and the porous in 3D graphene network are well preserved. The overall electrochemical performance of the nanocomposite of NFS@rGO was improved because of the synergistic effect between surface coated 3D graphene and Na 2 . 4 Fe 1 . 8 (SO 4 ) 3 nanograins [99]. Nanomaterials 2022, 12, x FOR PEER REVIEW 18 of 32 the above electrochemical properties indicate that the overall performance of Sn4P3/GT was significantly improved by the introduction of graphene tubes on the surface of Sn4P3 nanoparticles. As shown in the outline in Figure 18a, the graphene-coated Na2.4Fe1.8(SO4)3 (denoted as NFS) nanograins (NFS@rGO microspheres) was successfully fabricated by Fang Y.J. et al., the 3D graphene microsphere network via a spray-drying route [99]. From the SEM images of Figure 18b-d, the Na2.4Fe1.8 (SO4)3 nanoparticles are homogeneously distributed between layers of 3D spherical graphene and the porous in 3D graphene network are well preserved. The overall electrochemical performance of the nanocomposite of NFS@rGO was improved because of the synergistic effect between surface coated 3D graphene and Na2.4Fe1.8(SO4)3 nanograins [99]. To evaluate the sodium-storage behaviors of the materials and structure design strategy, the NFS@rGO microspheres were investigated by assembling 2016-type coin cells. As shown in Figure 19a, during the initial CV cycle, two pairs of distinct peaks can be found, while in the subsequent cycles, three pairs of peaks appeared, and the above phenomenon matches well with the previous study [107][108][109]. Figure 19b reflects the 1st to 4th cycles of the GCD curves of NFS@rGO microspheres at 0.05 C and the 1st charge curve shows a different shape compared to other curves, which agrees well with the CV in Figure 19a and previous research [107,108]. The reversible specific capacity of NFS@rGO electrode reaches 100 mAh g −1 with the average working potential of 3.75 V, obtaining an energy density of 375 Wh kg −1 for the corresponding full cell. Similar to CV curves, the GCD curves are almost overlapped after the initial cycle that manifests the electrode's reversible electrochemical behavior in subsequent cycles. Furthermore, it may be due to the formation of the cathode electrolyte interface layer and the irreversible decomposition of electrolyte, the Coulombic efficiency of the NFS@rGO has a certain reduction in the subsequent few cycles compared with the initial Coulombic efficiency of 89% [110]. To evaluate the sodium-storage behaviors of the materials and structure design strategy, the NFS@rGO microspheres were investigated by assembling 2016-type coin cells. As shown in Figure 19a, during the initial CV cycle, two pairs of distinct peaks can be found, while in the subsequent cycles, three pairs of peaks appeared, and the above phenomenon matches well with the previous study [107][108][109]. Figure 19b reflects the 1st to 4th cycles of the GCD curves of NFS@rGO microspheres at 0.05 C and the 1st charge curve shows a different shape compared to other curves, which agrees well with the CV in Figure 19a and previous research [107,108]. The reversible specific capacity of NFS@rGO electrode reaches 100 mAh g −1 with the average working potential of 3.75 V, obtaining an energy density of 375 Wh kg −1 for the corresponding full cell. Similar to CV curves, the GCD curves are almost overlapped after the initial cycle that manifests the electrode's reversible electrochemical behavior in subsequent cycles. Furthermore, it may be due to the formation of the cathode electrolyte interface layer and the irreversible decomposition of electrolyte, the Coulombic efficiency of the NFS@rGO has a certain reduction in the subsequent few cycles compared with the initial Coulombic efficiency of 89% [110].   [107,[111][112][113]. Additionally, the discharge plateaus of th NFS@rGO dropped with the increase of current rate, while the working potentials are sti maintained above 3 V at a current rate even up to 30 C, better than most of the relative Fe based cathodes, such as FePO4 [114], NaFePO4 [115], NaFePO4F [116], and Na2FeP2O7 [117].

Intercalation Layer Stacked Graphene
As a promising carbon material, graphene has a series of advantages, such as hig conductivity, large surface area and exceptional mechanical strength [118]. Therefore, kind of intercalation layer stacked graphene embedded in active materials betwee graphene layer was designed to achieve a shortened electron/Na + transport distance wit improved reaction kinetics [59]. The graphene layer act as the conducting pathway fo electrons and mechanical support for nanomaterials that are confirmed to have a stron affinity to active materials, thereby mitigating the fragmentation of substances during th reaction [119]. The graphene layer and the active materials formed a sandwich electrod have a mechanically robust "nest" between the graphene layer to induce uniform deposition of active materials, therefore, inducing a strong surface capacitive contributio with high-rate capability [119]. Additionally, in the traditional electrode preparin process, polymer binders of insulation were used to adhesion the active nanomaterials o   [107,[111][112][113]. Additionally, the discharge plateaus of the NFS@rGO dropped with the increase of current rate, while the working potentials are still maintained above 3 V at a current rate even up to 30 C, better than most of the relative Fe-based cathodes, such as FePO 4 [114], NaFePO 4 [115], NaFePO 4 F [116], and Na 2 FeP 2 O 7 [117].

Intercalation Layer Stacked Graphene
As a promising carbon material, graphene has a series of advantages, such as high conductivity, large surface area and exceptional mechanical strength [118]. Therefore, a kind of intercalation layer stacked graphene embedded in active materials between graphene layer was designed to achieve a shortened electron/Na + transport distance with improved reaction kinetics [59]. The graphene layer act as the conducting pathway for electrons and mechanical support for nanomaterials that are confirmed to have a strong affinity to active materials, thereby mitigating the fragmentation of substances during the reaction [119]. The graphene layer and the active materials formed a sandwich electrode have a mechanically robust "nest" between the graphene layer to induce uniform deposition of active materials, therefore, inducing a strong surface capacitive contribution with high-rate capability [119]. Additionally, in the traditional electrode preparing process, polymer binders of insulation were used to adhesion the active nanomaterials on the surface metal Cu or Al current collector, which limited the rate capability, flexibility and energy density of the active materials [3,120]. Furthermore, the interfacial bonding of the traditional electrode between active materials and metal current collectors have many limitations during the repeated sodiation/desodiation process that will cause damage to the electrode under the long-term volume expansion [118,121,122]. Therefore, constructing binder-free and freestanding electrode with layer stacked graphene sheets will obtain the mechanical flexibility, high lightweight and electrical conductivity electrodes that will effectively improve the overall performance of SIBs.
Liu, M.K. et al., designed and developed a 3D architecture by CNFs directly penetrating into graphene sheets (CNFIG) as the electrically conductive template and then the MoS 2 nanoflakes were in situ deposited on the surface of the CNFIG aerogel as the MoS 2 @CNFIG hybrid for the anode of SIBs (Figure 20a) [118]. A versatile interfacial method was first carried out to depose the MoS 2 on the surface of CNFIG and then high-temperature treatment of the product of the first step to obtain the MoS 2 @CNFIG anode. As shown in Figure 20d-f, the layered MoS 2 nanoflakes were uniformly grown on both graphene layers and the CNF bridges that increasing the loading amount of the active substance. The vertically aligned channels in the inter of MoS 2 @CNFIG nanocomposite accelerate the penetration of electrolyte and facilitate the rapid transfer of sodium ions (Figure 20b). Since CNFIG is composed of carbon materials that provide an efficient path for rapid electron transport, facilitates the insertion/extraction ability of sodium ions. and energy density of the active materials [3,120]. Furthermore, the interfacial bonding of the traditional electrode between active materials and metal current collectors have many limitations during the repeated sodiation/desodiation process that will cause damage to the electrode under the long-term volume expansion [118,121,122]. Therefore, constructing binder-free and freestanding electrode with layer stacked graphene sheets will obtain the mechanical flexibility, high lightweight and electrical conductivity electrodes that will effectively improve the overall performance of SIBs. Liu, M.K. et al., designed and developed a 3D architecture by CNFs directly penetrating into graphene sheets (CNFIG) as the electrically conductive template and then the MoS2 nanoflakes were in situ deposited on the surface of the CNFIG aerogel as the MoS2@CNFIG hybrid for the anode of SIBs (Figure 20a) [118]. A versatile interfacial method was first carried out to depose the MoS2 on the surface of CNFIG and then hightemperature treatment of the product of the first step to obtain the MoS2@CNFIG anode. As shown in Figure 20d-f, the layered MoS2 nanoflakes were uniformly grown on both graphene layers and the CNF bridges that increasing the loading amount of the active substance. The vertically aligned channels in the inter of MoS2@CNFIG nanocomposite accelerate the penetration of electrolyte and facilitate the rapid transfer of sodium ions (Figure 20b). Since CNFIG is composed of carbon materials that provide an efficient path for rapid electron transport, facilitates the insertion/extraction ability of sodium ions. The SEM image and corresponding EDX elemental mappings (C, Na, Mo, and S) of MoS2@CNFIG anode after 1000 cycles were applied to investigate the structural and element integrity of the MoS2@CNFIG anode. From the SEM image of Figure 21a, the porous of the MoS2@CNFIG were preserved, so that the ionic diffusion path and the The SEM image and corresponding EDX elemental mappings (C, Na, Mo, and S) of MoS 2 @CNFIG anode after 1000 cycles were applied to investigate the structural and element integrity of the MoS 2 @CNFIG anode. From the SEM image of Figure 21a, the porous of the MoS 2 @CNFIG were preserved, so that the ionic diffusion path and the volume expansion space for MoS 2 nanoflakes can be guaranteed during the long-time cycling. All of the above properties undoubtedly contribute to the quick sodiation/desodiation ability and rapid transfer of sodium ions. The EDX elemental mappings of C, Na, Mo, and S elements can be detected in MoS 2 @CNFIG anode after 1000 cycles proving that the composition of the anode was well maintained, and some sodium ions were adsorbed by the anode during the chemical redox reactions. Figure 21c shows three assembled full cells based on MoS 2 @CNFIG anode in parallel powering the 12 LEDs which demonstrate the potential applications of the MoS 2 @CNFIG nanocomposite. volume expansion space for MoS2 nanoflakes can be guaranteed during the long-time cycling. All of the above properties undoubtedly contribute to the quick sodiation/desodiation ability and rapid transfer of sodium ions. The EDX elemental mappings of C, Na, Mo, and S elements can be detected in MoS2@CNFIG anode after 1000 cycles proving that the composition of the anode was well maintained, and some sodium ions were adsorbed by the anode during the chemical redox reactions. Figure 21c shows three assembled full cells based on MoS2@CNFIG anode in parallel powering the 12 LEDs which demonstrate the potential applications of the MoS2@CNFIG nanocomposite. To evaluate the rate ability and long-term cycling ability (Figure 21d), the MoS2@CNFIG anode was cycled firstly under the current density of 1 A g −1 for 300 cycles and then 10 A g −1 for 400 cycles with high Coulombic efficiencies approaching ~100%. The MoS2@CNFIG anode provides a capacity of 322 mA h g −1 after 300 cycles at 10 A g −1 and To evaluate the rate ability and long-term cycling ability (Figure 21d), the MoS 2 @CNFIG anode was cycled firstly under the current density of 1 A g −1 for 300 cycles and then 10 A g −1 for 400 cycles with high Coulombic efficiencies approaching~100%. The MoS 2 @CNFIG anode provides a capacity of 322 mA h g −1 after 300 cycles at 10 A g −1 and 303 mAh g −1 in the 700th cycle. Additionally, the specific capacity was maintained at 421 mA h g −1 when the current density back to 1 A g −1 and can continue the cycle of 300 times, which indicates the recovery ability of the electrode. The excellent rate ability and cycling ability of MoS 2 @CNFIG nanocomposite may be ascribed to its structural stability, excellent conductive of CNFIG and the high utilization efficiency of MoS 2 nanoflakes with large specific surface area. The Nyquist plots of the full cell were investigated after being cycled 2 and 1000 cycles and presented in Figure 21e. The EIS spectra of the MoS 2 @CNFIG anode were obtained after being cycled two and 1000 times. The R ct value reflects the diameter of the semicircle in Figure 21e represents the charge-transfer resistance. The MoS 2 @CNFIG hybrid at 2nd cycle shows a smaller R ct value of 100 Ω compared with the R ct value of 162.3 Ω at the 1000th, indicating that the initial electrode has a lower interfacial resistance as well as better kinetics and higher ion-diffusion rate during the sodiation/desodiation process. Furthermore, the slope of the low-frequency region does not decrease much with the increase of GCD cycles, implying that the electrochemical performance of the electrode has not change significantly. From the TEM images in Figure 21f, the MoS 2 @CNFIG also shows clear lattice fringes with flake-like MoS 2 structures (inset of Figure 21f) after long-time cycles. Finally, the author attributes the excellent electrochemical storage ability of the SIBs to the 3D nested structures of MoS 2 @CNFIG producing nanoreservoirs between adjacent MoS 2 nanoflakes [121], the vertically aligned channels in CNFIG [122] and the MoS 2 anchored tightly on the surface of CNFIG with good electrical contact (Figure 21g).
Yang, Z.X. et al., reported the hierarchically structured graphene/MoS 2 composites hybrid sandwich structure (3DG-MoS 2 ) synthesized by solution evaporation followed the hydrothermal reaction method [30]. At the synthesis stage of the sample, NH 2 CSNH 2 to the system, under hydrothermal conditions, releases H 2 S gas through the reaction of NH 2 CSNH 2 + 2H 2 O → 2NH 3 + H 2 S +CO 2 . After the reaction with H 2 S, the molybdate in G/MCu nanocomposite was converted into MoS 2 nanosheets [123]. In order to improve the sodium storage ability, the 3DG-MoS 2 nanocomposite was fabricated by embedding the Cu 2 O nanoparticles on the surface of graphene sheets (G/MCu) and then assembling MoS 2 nanosheets further on the surface of graphene sheets by hydrothermal reaction (Figure 22a) [30,124,125]. Figure 22b indicated the FE-SEM image of as-prepared G/MCu, from which we can find that the Cu 2 O nanoparticles have been embedded into the graphene layer with the formation of the rough surface. The cross-section FE-SEM images of G/MCu further indicated that Cu 2 O as the tiny particles distributed uniformly among the layers of graphene. After the hydrothermal reaction with thiourea, the molybdate in G/MCu nanocomposite was converted into MoS 2 nanosheets that formed the sandwich layer structure of 3DG-MoS 2 , as shown in Figure 22d,e. The particular architecture of layer structured graphene accelerated the penetration of electrolyte (NaClO 4 (1 mol/L) in DEC: EC (1:1 v/v) solution) and improved the specific surface of active material that facilitates the overall performance of SIBs.
As shown in Figure 23, the author further investigated the electrochemical performance of the 3DG-MoS 2 electrode under different conditions. Na storage ability of the 3DG-MoS 2 electrode was initially evaluated through CV curves of the 3DG-MoS 2 electrode at the voltage range of 0.0 to 3.0 V versus Na/Na + under the scan rate of 0.5 mVs −1 . As shown in Figure 23A, a wide peak of 0.8 V can be observed in the first cycle that may be related to the insertion of sodium ions into MoS 2 and the irreversible formation of SEI layer [126,127]. Furthermore, the CV curves are well overlapping after the initial cycle, demonstrating excellent reversible electrochemical behavior of 3DG-MoS 2 electrode. Figure 23B reflects the 1st, 2nd, 5th, 10th, 50th and 100th cycles of the GCD curves of 3DG-MoS 2 electrode at a current density of 0.5 A g −1 and the 1st charge curve shows a different shape compare with other curves, which match well with the CV in Figure 23A. The different shapes of the first GCD curve may represent the capacity loss due to the irreversible formation of the SEI layer.
graphene layer with the formation of the rough surface. The cross-section FE-SEM images of G/MCu further indicated that Cu2O as the tiny particles distributed uniformly among the layers of graphene. After the hydrothermal reaction with thiourea, the molybdate in G/MCu nanocomposite was converted into MoS2 nanosheets that formed the sandwich layer structure of 3DG-MoS2, as shown in Figure 22d,e. The particular architecture of laye structured graphene accelerated the penetration of electrolyte (NaClO4 (1 mol/L) in DEC EC (1:1 v/v) solution) and improved the specific surface of active material that facilitates the overall performance of SIBs.   As shown in Figure 23, the author further investigated the electrochemical performance of the 3DG-MoS2 electrode under different conditions. Na storage ability of the 3DG-MoS2 electrode was initially evaluated through CV curves of the 3DG-MoS2 electrode at the voltage range of 0.0 to 3.0 V versus Na/Na + under the scan rate of 0.5 mVs −1 . As shown in Figure 23A, a wide peak of 0.8 V can be observed in the first cycle that may be related to the insertion of sodium ions into MoS2 and the irreversible formation of SEI layer [126,127]. Furthermore, the CV curves are well overlapping after the initial cycle, demonstrating excellent reversible electrochemical behavior of 3DG-MoS2 electrode. Figure 23B reflects the 1st, 2nd, 5th, 10th, 50th and 100th cycles of the GCD curves of 3DG-MoS2 electrode at a current density of 0.5 A g −1 and the 1st charge curve shows a different shape compare with other curves, which match well with the CV in Figure 23A. The different shapes of the first GCD curve may represent the capacity loss due to the irreversible formation of the SEI layer. To evaluate the long-term cycling ability ( Figure 23C), the 3DG-MoS2 electrode provides a specific capacity of 524 mAh g −1 under the current density of 0.5 A g −1 after 100 To evaluate the long-term cycling ability ( Figure 23C), the 3DG-MoS 2 electrode provides a specific capacity of 524 mAh g −1 under the current density of 0.5 A g −1 after 100 cycles (the slight increase of capacity with cycling may be due to activation process), which higher than that of the G/MoS 2 and MoS 2 electrodes of 320 and 284 mAh g −1 , respectively. In the author's opinion, the improvement of the cycling ability of the 3DG-MoS 2 electrode is mainly ascribed to the structure of the strong G/MoS 2 thin layers existing in 3DG/MoS 2 electrode, which prevents the damage caused by the volume expansion of active material. In order to further investigate the overall performance of 3DG-MoS 2 electrode, further 300 long-term cycling ability and Coulomb efficiency evaluations of the 3DG-MoS 2 electrode were carried out at a current density of 1 A g −1 in Figure 23D. Obviously, the 3DG-MoS 2 electrode still provides a capacity up to 479 mA h g −1 even after 300 cycles and exhibits the coulombic efficiency of~100% in the whole 300 cycles except for the first one.

Discussion and Prospects
Although the graphene-based nanocomposites have been well developed in the field of new energy, there are still several key problems that should be solved to achieve high performance sodium-ion batteries for practical applications. Firstly, although a relatively comprehensive structural system of SIBs has been developed based on graphene, it is necessary to carry out the research based on more positive materials with high energy density and power density, as well as negative materials with small volume change in the process of circulation [128]. Furthermore, further improving the cycle stability of the batteries are not only an important strategy to improve the performance of sodium ion batteries, but also the critical point to realize the large-scale application of sodium ion batteries for commercial application [129]. In addition, at present, the cost of graphene-based electrodes is relatively high, and the synthesis methods of composite electrode materials are relatively complicated. More efficient strategies for combining graphene with active substances should be developed and the theoretical model of SIBs materials should be established in accordance with the development of graphene-based composite electrodes [130]. With the progress of the above technologies, sodium ion batteries can first replace the lead-acid batteries with high pollution and low energy density, and then play a greater role in more aspects. From a macro and pragmatic perspective, more research efforts of SIBs should be focused on the following four points as the prospects of this review: (1) In-depth knowledge of Na+ storage mechanism At present, the mechanism of sodium storage is still inconclusive and needs further study, especially on low pressure platforms, ultrafine pore and closed pore filling [131]. Further studies using in/ex situ techniques [132], as well as theoretical calculations and simulations [133], are needed to better understand the role of pore size distribution, openings and closures, surface functional groups, structural defects, and pseudo graphitic domains in the sodium/desulphurization process. Developing a deeper understanding of interaction mechanism between graphene and active materials is helpful to design more novel graphene-based nanomaterials with special structures and high capacity for SIBs [134].
(2) Reasonable structure and process design of electrodes In order to further improve the performance of the anodes, methods such as physical and chemical activation can be used. Such as: graded pore structure, microwave activation [135], H 2 reduction [136] and high current pretreatment [137], heteroatom doping [138], metallic NPs (or allowable materials) synthesis of nano-composite materials [139]. Some efforts should be exerted on investigating the effects of morphology, porosity, size, and defects of graphene-based electrodes to achieve better electrochemical performance [134]. Of course, in addition to some anode materials introduced in the article, it is also important to explore more efficient cathode materials and composite method of graphene based electrodes, such as NaxMO 2 (M = Co, Mn, Ni) [140][141][142] [145], Na 3 V 2 (PO 4 ) 3 (NVP [146,147], Na 2 MP 2 O 7 (M = Fe, Co, Mn) [148][149][150] and NaFe(SO 4 ) 2 [151].
(3) Environmental, economic, and scaling-up aspects By implementing low-energy processes and avoiding the use of expensive and/or hazardous chemicals, in other words, at relatively mild temperatures, SIBs can be promoted closer to commercialization by using cheaper and sustainable sources (even waste streams) such as heteroatom doping [152]. The achievement of high-quality, mass-produced, and low-cost graphene through facile methods is hard but urgent. In addition, the modification of the graphene with higher utilization efficiency should not be ignored [131]. The initial coulombic efficiency (ICE) of SIBs for both anode and cathode materials, is a key parameter for high performance SIBs, and the point is to increase the transport rate of the Na + ions. Therefore, developing SIBs with high ICE and rate performance becomes vital to boost the commercialization of SIBs [153]. Improving initial coulombic efficiency (ICE) is due to some decomposition of electrolyte and also to the capture of irreversible sodium in pore, functional group and interlayer space. Optimizing the formulation of electrolyte and binder [141], HC engineering defects, surface area and functional groups [154], and a thorough understanding of the formation and function of SEI are essential to promoting a rational and effective electrode design [155].

(4) Full cell fabrication and commercialization of SIBs
Full-SIB devices assembled from graphene-based anodes or cathodes show promising prospects in practical applications and related performance [156][157][158][159][160][161][162][163][164][165][166]. Full batteries with graphene-based anodes have lower average discharge voltage when they have large capacity [156][157][158][159][160]. In contrast, a full battery consisting of graphene-based cathodes yields higher voltage, but its capacity and cycle capacity are unsatisfactory [165,166]. In future studies, it is worth considering the balance between voltage and capacity of all SIB based on graphene. Anodes and cathodes with high capacity, stability and suitable voltage can all achieve high performance SIBs. Further research on the application of excellent electrochemical properties in semi-batteries to full batteries and hybrid capacitors is very popular. In the case where the cathode does not provide unlimited sodium ions, it is necessary to study the anode in the full battery to evaluate its actual performance. In addition, electrolytes and additives are particularly related to the compatibility of anode and cathode materials. Building flexible SIBs is a trend of graphene-based materials [131].

Conclusions
This review systematically introduces the combination of graphene with active substances as the nanocomposite electrodes for sodium ion electrochemical energy storage. To achieve this target, four categories of graphene-based nanocomposites fabricate stargates have been introduced in this article. Based on those categories and the research results of recent years, this review focus on the fabricate methods and structural characteristics of the graphene-based nanocomposites and the influence of graphene and morphology on the performance of the SIBs. With the development of nanotechnology as well as the excellent conductivity and plasticity of graphene, nanocomposite fabricates methods needed to be extended and expanded to stimulate the development and application of electrochemical energy storage in a benign direction. The discussion on research frontiers and applications in the field of graphene-based SIBs provide evidence that the application of graphene in advanced electrode materials is still a research hotspot, and more ingenious structure design and energy storage mechanism analysis will be the future development on the direction of nanocomposites for SIBs. Although graphene-based materials have been widely studied in SIBs and show good development prospects, their practical applications still face enormous challenges. In the first place, high-quality, large-scale production and low-cost graphene by simple methods are the basis for the production of graphene-based sodium ion batteries and should be further studied. In another, modification of graphene with higher utilization efficiency and better understanding of the mechanism of interaction between graphene and active materials is the basis of designing new electrodes. Studying the relationship of the performance of SIBs between the morphology, porosity, size and defect of graphene-based composite electrode is the key to improve the cycle life, high Coulomb efficiency and high energy efficiency of SIBs, which is the key to the marketization of SIBs. Furthermore, reaction mechanisms and theoretical analysis associated with Na + insertion/depletion need further exploration.

Conflicts of Interest:
The authors declare no conflict of interest.