Structures, performances and applications of green biomass derived carbon in lithium-ion batteries

Carbon for cathode materials

Several cathode materials such as LiCoO₂, spinel LiMn₂O₄, LiNi_0.8CO_0.15Al_0.05O₂, LiNi_0.33CO_0.33Mn_0.33O₂ and olive LiFePO₄ have been widely used. However, the specific capacity of all these materials can only reach 120-200 mAh g^-1[40]. There are also problems for these cathode materials, including transition metal dissolution, low conductivity, poor cycle stability and fast voltage fading^[41-44]. Therefore, researchers tried to solve these problems through surface coating of carbon materials on corresponding active cathode materials^[42,45-47]. In particular, carbon as the coating can not only protect the cathode materials but also improve the electronic conductivity and rate capabilities of LIBs^[48]. Zhang et al. demonstrated that carbon nanotube coatings (CNTs) on LiMn₂O₄ exhibited a better electrochemical performance and longer cycle life than LiMn₂O₄ without a coating^[49]. The coating layer decreased the contact area between the cathode material and electrolyte, which prevented the dissolution of Mn. Moreover, the coating also increased conductivity. Luo et al. stated that LiFePO₄ nanoparticles coated with CNTs exhibited good conductivity due to the formation of a conductive network by CNTs through intimate connections with LiFePO₄ nanoparticles^[50]. LIBs assembled with such cathode materials exhibited an excellent cycling performance with a capacity loss of less than 10% after 450 cycles at 10 C. In addition, graphene is a desirable additive that can be divided into single-, double- and multi-layer types. Regardless of the type, the number of layers, in general, is less than 10. For this reason, we collectively referred to them as graphene. These graphene materials share common features, including large specific surface areas (SSA), high ion and electron carrier mobility, and unique physical and chemical properties. As an additive, these characteristics make graphene highly desirable for improving the performance of anode materials. Yang et al. used reduced graphene oxide as the coating material to improve the electrochemical properties of LiNi_0.8Co_0.1Mn_0.1O₂ (RGO-NCM)^[51]. However, as the number of cycles increases, the graphene falls off the surface of the active material powders, resulting in the failure of the coating. Therefore, in order to make graphene more tightly and evenly wrapped around the positive electrode material and increase the interface conductivity, epoxy-functionalized silane (KH560) was added between the graphene and LiNi_0.8Co_0.1Mn_0.1O₂ particles to build a dual-function coating (RKNCM1), which improved the capacity retention rate, rate performance and high-temperature performance of cathode materials, as illustrated in Figure 2.

Figure 2. (A) Structure schematics of RGO- and RGO-KH560-coated LiNi_0.8Co_0.1Mn_0.1O₂ particles; (B and C) SEM images of RKNCM1 particles; (D and E) SEM images of RGO-NCM particles; (F and G) SEM images of NCM particles; (H) Cycling performance at room temperature of RKNCM1, RGO-NCM and NCM particles; (I) Cycling performance at 55 °C of RKNCM1, RGO-NCM and NCM particles; (J) Rate performance of RKNCM1, RGO-NCM and NCM particles. These figures are quoted with permission from Yang et al.^[51].

Carbon for anode materials

Graphite is one of the most widely used anode materials, and its layered structure ensures the smooth and stable insertion and desertion of lithium ions. It has a wide range of sources and is relatively easy to produce industrially while being low-cost, safe, and environmentally friendly^[52-55]. However, LIBs with graphite-based anodes have inherent shortcomings of low specific capacity, low Coulombic efficiency (CE) and poor rate performance^[56-58]. Although graphite is still being further developed and optimized, its electrochemical performance has almost reached the theoretical limit^[59]. Scientists have developed alternative anode materials of higher theoretical capacity, such as graphene, silicon, germanium, metal oxide, alloy, metal-organic frameworks, and so on, along with changing the structure and size of anode materials to further improve the electrochemical performances of LIBs^[60-66]. However, carbon materials are still often used as additives to protect active anode materials and/or form a conductive framework to increase the electric conductivity.

Some new anode materials have such problems as volume expansion and poor electrical conductivity^[67]. Carbon materials are often combined with these anode materials to improve electrical conductivity and reduce stress inside the electrode by providing a buffer volume, thus enhancing electrochemical characteristics. Lee et al. prepared anode materials by combining ultrafine TiO₂ nanocrystals (TiO₂ NP) with a macroporous three-dimensional graphene network (3D PG) to achieve a faster electron transfer [Figure 3A and B]^[59]. The binder-free 3D PG, on the other hand, has a low sheet resistance and can be made in direct contact with the current collector for ultra-fast electron transfer, providing a good cycle performance of a CE value close to 100% after 10,000 cycles [Figure 3C]. Zhang et al. proposed a hierarchical tubular structures constructed by carbon-coated ultra-thin tin oxide nanoplates (SnO₂@C-HTs) [Figure 3D], combining the advantages of low-dimensional ultra-thin nanoplates, hollow tube structures and nanocarbon coatings^[68]. Anodes of SnO₂@C-HTs exhibited a higher specific capacity and better cycle stability than SnO₂ without ultra-thin carbon coatings [Figure 3E and F].

Figure 3. (A) Structure of TiO₂ NP-PG; (B) Working mechanism of TiO₂ NP-PG; (C) Discharging (circle)/charging (square) specific capacities of TiO₂ NP-PG (blue) and TiO₂ NP (green) at different current densities along with the CE (black). (D) Schematic illustration of the formation and TEM images of SnO₂@C-HTs; (E) Cycling performance of SnO₂@C-HTs and SnO₂ without ultra-thin carbon coatings; (F) Rate capacities of SnO₂@C-HTs and SnO₂ without ultra-thin carbon coatings; (G) Three-dimensional structure of Si pomegranate; (H) SEM images of Si pomegranate; (I) Reversible capacities for 1,000 cycles of the Si pomegranate and other structures tested at C/2. (A-C) are quoted with permission from Lee et al.^[59]; (D-F) from Zhang et al.^[68]; and (G-I) from Liu et al.^[69].

Silicon is seen to be the most promising alternative to graphite as an anode material because its high specific capacity is more than ten times that of graphite. However, the problem of large volume changes of silicon during charging and discharging needs to be solved. Structural design or carbon coating are the most popular methods and can greatly alleviate the problem. Liu et al. designed anode materials of pomegranate structure, using a conductive carbon layer encapsulating a single silicon nanoparticle and then a thicker carbon layer encapsulating the ensemble of these nanoparticles [Figure 3G and H]^[69]. Such design resulted in a high CE, high volumetric capacity (1,270 mAh cm^-3) and stable cycling over 1,000 cycles [Figure 3I].

Porous biomass-derived carbon

Porous structures are complex and changeable but easy to access. Porous biomass-derived carbons (PBCs), in general, have a high SSA, which increases the loading of active substances. The pore size and structure can be adjusted according to target applications and functions, facilitating ion transport and allowing the electrolyte to penetrate into the whole electrode.

In general, a two-step method of hydrothermal carbonization and KOH activation strategy is used to prepare porous carbons. Qiang et al. synthesized a porous form of biomass-derived carbons through a two-step process: hydrothermal carbonization of cypress coats, followed by an activation step using KOH as a porogen. By modulating the KOH to carbon mass ratio during the activation phase, it is possible to regulate the morphology and porous characteristics of resulting carbons^[104]. Recently, a simple, low-cost and green one-step method has been proposed by Gong et al. Such a method gave carbon products abundant microporous structures, a high degree of graphitization and many oxygen-containing groups^[105]. The 3D porous carbon (3DPC) was synthesized from bamboo char using this one-step method. The porous carbon synthesized through a one-step method was found to have more abundant micro/meso pores, higher degree of carbonization and better electrochemical performances than the carbon synthesized using a traditional two-step method.

PBCs can be used in both anodes and cathodes. However, their impacts on the performance of cathodes and anodes are different. In cathodes, porous carbons derived from biomass are introduced to cathodes as a coating, improving electrical conductivity and stability, which generally increases the cycling stability. Noerochim et al. utilized, for example, sucrose as a carbon precursor to fabricate a porous and conductive coating for LiFePO₄ cathodes, aiming to enhance the electrical conductivity and stability of the cathodes when used with an aqueous electrolyte^[106]. Their findings indicated that a coating with a 9 wt% carbon yielded a product of the best performance, achieving an initial specific discharge capacity of 13 mAh g^-1, with only a 2% capacity loss after 100 cycles. Zhang et al. designed a high-rate performance cathode material (FF@3DPC) by embedding FeF₃·0.33H₂O nanoparticles in nori-derived nitrogen-oxygen double-doped 3DPC [Figure 4A and B]^[107]. FF@3DPC had high reversible capacities of 134 and 104 mAh g^-1 at 5C and 20C, respectively [Figure 4C]. After 500 cycles, it maintained a reversible capacity of 101 mAh g^-1 at 5C, as shown in Figure 4D.

Figure 4. Schemes, structure and performance of porous biomass-derived carbons. (A) Synthesis process of 3DPC and FF@3DPC; (B) SEM images of 3DPC; (C) Rate capability of FF@3DPC electrode and bulk FeF₃·0.33H₂O electrode at different current density; (D) Cycling performance of FF@3DPC electrode at 5C. (E) Schematic illustration of the porous structure and working mechanism of HPNDC; (F) SEM images of HPNDC; (G) Rate capabilities of HPNDC at different current densities; (H) Cycling performance of HPNDC at 100 mA g^-1.(A-D) are quoted with permission from Zhang et al.^[107]; and (E-H) from Ou et al.^[108].

PBCs applied to the anode are often doped with other elements to further improve the rate capacity and cycling stability. Ou et al. proposed to use chrysanthemum as a novel carbon source for creating a hierarchical porous nitrogen-doped carbon (HPNDC) anode, as depicted in Figure 4E and F^[108]. This HPNDC demonstrated a superior high-rate performance, as shown in Figure 4G, with a specific capacity of 347 mAh g^-1 at 5 A^[108]. Furthermore, the anode exhibited a commendable electrochemical performance, maintaining a reversible capacity of 975 mAh g^-1 after 100 cycles at 100 mA g^-1, as illustrated in Figure 4H. Xu et al. explored the use of sulfur and nitrogen-doped porous carbon derived from bean shells as an anode material for LIBs^[109]. At 1C, this material sustained a capacity of 262 mAh g^-1 even after 100 cycles. Zheng et al. reported the synthesis of porous nitrogen-doped carbon (PNDC) from duckweed, concluding that the material could support LIBs with a capacity of 1,071 mAh g^-1 after 100 cycles when subjected to a current density of 100 mA g^-1[110]. Yan et al. developed an anode material from bamboo leaves, resulting in a HPNDC^[111]. This material not only showcased an impressive rate capability but also exhibited significant long-term stability. After 500 cycles at a current density of 0.2 A g^-1, the optimized carbon could deliver a discharge capacity of 450 mAh g^-1.

To sum up, natural substances with an intricate network of interconnected channels and pores are the ideal precursors of PBC. PBC is easy to prepare using two- and one-step methods, including the activation method, template method or direct carbonization. Moreover, it has been proven to have high capacity and excellent cycling ability due to its pore structure which provides electroactive sites to contribute additional capacity, fast diffusion channels for ions to enhance kinetics and a buffer zone to decrease interfacial pressure. If the preparation method can be further simplified and more environmentally friendly, PBC may become a viable substitute for graphite.

Fibrous biomass-derived carbon

The potential of fibrous biomass-derived carbon (FBC) materials applied to the anode has been discovered in recent years. These materials are usually from cellulose, silk and chitin, which have complex hierarchical structures composed of micro- and nano-fiber. Therefore, the fibrous carbon of various shapes can form from these FBCs. It is often fabricated by dissolution and regeneration, followed by carbonization. Cellulose is dissolved and separated in the solution, and then regenerated fibers are processed into a new dimension. Finally, FBC is obtained by carbonization. This processing method is simple and low-cost.

FBC used in LIBs provides fast transport channels for ions and electrons as such that the electrochemical performance of LIBs is improved significantly. Zhang et al. transformed bamboo, pine, poplar and paulownia into uniform hollow fibers of micrometer sizes by delignification and carbonization^[112]. FBC had a larger interlayer spacing of about 0.39-0.40 nm which was calculated by Bragg’s law using the results of X-ray diffraction (XRD) characterization. The anode made from the fibrous carbon exhibited a better electrochemical performance than commercial graphite anodes. LIBs had a high reversible capacity, 435 mAh g^-1 at 50 mA g^-1, competitive rate capacity of up to 150 mAh g^-1 at 2 A g^-1 and good cycling stability of 76% capacity retention after 500 cycles. Scientists have discovered that wild fungus is also an ideal source for fabricating FBC. Tang et al. reported synthesis of carbonaceous fibers from Tyromyces fissilis wild fungus [Figure 5A] by controlling the carbonization process^[113]. The carbon produced as such had a larger interlayer spacing of 0.386 nm as compared with 0.335 nm of graphite. It exhibited longer-term cycling stability, as shown in Figure 5B. Its specific reversible capacity was 340 mAh g^-1 at C/10 and 300 mAh g^-1 at a fast 1C charge/discharge rate [Figure 5C]. Seaweed is also an ideal carbon source for the preparation of carbon materials. Lv et al. harnessed the templating ability of the natural structure of alginate molecular chains to produce a high-performance fibrous anode [Figure 5D]^[114]. Fe₂O₃ particles are embedded in the fibers to form a yolk-shell structure through coordination (C@Fe₂O₃-F). This fibrous anode demonstrated excellent cycling performance with no capacity decrease after 200 cycles (1,035 mAh g^-1) and good rate performance [Figure 5E and F].

Figure 5. Schemes, structure and performance of fibrous biomass carbon. (A) Synthesis process of carbonaceous fibers from Tyromyces fissilis wild fungus; (B) Cycling performance of carbonaceous fibers from Tyromyces fissilis wild fungus; (C) Specific discharge capacities of carbonaceous fibers from Tyromyces fissilis wild fungus. These figures are quoted with permission from (D) Synthesis process of C@Fe₂O₃-F; (E) Cycling performance of C@Fe₂O₃-F; (F) Specific discharge capacities of C@Fe₂O₃-F. (A-C) are quoted with permission from Tang et al.^[113]; and (D-F) from Lv et al.^[114].

In summary, natural products with fibrous, filamentary or directional textures are more likely to produce FBC. Though such an approach has requirements for the structure of biomass materials, its fabrication process is relatively easy and environmentally friendly, only using carbonization. In addition, it has delivered good performance for LIBs because fibrous carbon forms a cross-linked conductive network. Such a network increases the overall conductivity and provides a rapid transmission channel for lithium ions, increasing the overall rate performance. Therefore, FBC has a good future of commercial utilization in the anode.

Spherical biomass-derived carbon

Spherical carbon has good structural stability, high packing density and a high surface-to-volume ratio. Therefore, it shows great potential for the use in LIBs. Spherical carbon is often prepared by carbonization of spherical biomass which is referred to as spherical biomass-derived carbon (SBC). Kim et al. synthesized micrometer-diameter SBC particles as the anode of LIBs using coffee oil as raw material^[115]. It was found that the spherical carbon retained its initial structure after 250 cycles. These LIBs had good cycling stability as the capacity remained 290 mAh g^-1 after 200 cycles. Song et al. obtained a robust carbon microsphere negative electrode material (MES600) by carbonizing the maleic anhydride esterified starch [Figure 6A and B]^[116]. Compared with starch-derived carbon without esterification modification (PS600), MES600 had better rate and cycling performance [Figure 6C and D]. To promote lithium-ion transport, carbon is processed into porous carbon spheres. Shi et al. utilized sucrose to prepare carbon microspheres with hierarchical micropores using an impregnation-carbonization method^[117]. The electrode with hierarchical porous carbon microspheres presented good rate capacities (1,141, 650, and 348 mAh g^-1 at the current densities of 0.05, 0.2, and 1 A g^-1, respectively). Chen et al. used corn starch to fabricate microporous carbon spheres through enzymolysis, pre-oxidation, and carbonization^[118]. The microporous carbon spheres showed an exceptional rate capability (150 mAh g^-1) at an ultrafast charge/discharge current density of 20 A g^-1, and as a result, LIBs were fully charged in 27 s. Ming et al. used glucose to prepare porous carbon balls as Fe₃O₄ coating layers (Fe₃O₄@porous-C) to improve the electrochemical performance of Fe₃O₄[Figure 6E and F]. Fe₃O₄@porous-C demonstrated a better rate capacity and cycle stability [Figure 6G]^[119].

Figure 6. Schemes, structure and performances of spherical biomass carbon. (A) Schematic illustration of the route to fabricate MES600 and PS600; (B) SEM image of MES600; (C) Rate performance of MES600 and PS600; (D) Cycling performance of MES600 and PS600; (E) The synthesis route of Fe₃O₄@porous-C; (F) SEM images of glucose-derived carbon microparticles; (G) Comparative cycling performance and rate capability of primary Fe₃O₄ and Fe₃O₄@porous-C. (A-D) are quoted with permission from Song et al.^[116]; and (E-G) from Ming et al.^[119].

In general, natural substances of spherical or regular shapes are more likely to form spherical carbon materials. SBC is usually obtained through carbonization of spherical biomass. The resulting SBC has high structural stability and a large specific area, which can improve cycle life. When combined with the porous structure and fabricated into a porous carbon sphere, the SBC can supply channels to transport ions quickly, promoting the fast charge and discharge performance and improving the rate capacity of LIBs.

Layered biomass-derived carbon

Raw biomass materials often contain fibers, which can be arranged into a layered structure similar to that of graphite anodes in LIBs. Layered biomass-derived carbon (LBC) can be synthesized by hydrothermal treatment, followed by graphitization and KOH activation. This process is similar to the fabrication of PBC. For the application in LIBs, the layered carbon often replaces graphite as the anode. It can provide a larger interlayer spacing to help fast transport of electrons and ions, containing more positions for the insertion and withdrawal of Li⁺ ions.

Zhou et al. synthesized complex and interconnected highly graphitic carbon nanosheets (HGCNS) using wheat stalks as the precursor^[120]. The HGCNS consists of ultrathin nanosheet frameworks with a 0.3362 nm interlayer spacing [Figure 7A]. The LIBs made with HGCNS were reported to have a high reversible capacity of 502 mAh g^-1 at 0.1 C [Figure 7B]. In addition, such LIBs had an excellent rate capability and superior cycling performance, providing 215 mAh g^-1 at 5 C after 2,000 cycles [Figure 7C]. Mondal et al. used eucalyptus tree leaves to prepare interconnected and porous carbon nanosheets which were used in the anode of LIBs^[121]. The anode showed a high specific capacity (819 mAh g^-1 at 100 mA g^-1) and good cycling stability (513 mAh g^-1 after 500 cycles). Gao et al. used oyster shells to prepare two-dimensional (2D) hierarchically porous carbon (HPC) [Figure 7D]^[122]. Sheet-like HPC electrodes displayed an excellent capacity retention at both 0.5 and 1 A/g after 400 cycles [Figure 7E]. The HPC-3-3 exhibited a high reversible capacity of 1,151 mAh g^-1 at 0.1 A g^-1 because of its cross-linked porous carbon network, more than three times higher as compared with pure carbon, along with good rate capacities [Figure 7F].

Figure 7. Schemes, structure and performances of layered biomass carbon. (A) Schematic diagram of lithium-ion transport during charging and discharging in HGCNS nanosheet frameworks; (B) Rate capacities of HGCN and flake graphite electrodes at different rates; (C) Cycling performance of HGCNS electrode at 1 C; (D) Schematic illustration of synthesis of sheet-like HPC; (E) Cycling performance of sheet-like HPC electrode at 1 and 0.5 A g^-1; (F) Rate capacities of sheet-like HPC. (A-C) are quoted with permission from Zhou et al.^[120]; and (D-F) from Gao et al.^[122].

In a word, natural substances of cellulose can be used to synthesize LBC. However, LBC needs to be treated first with a strong alkali and then carbonized. Such a processing method is similar to that used for graphite but the layered morphology can improve the capacity and cycling performance of LIBs. The layered carbon nanosheets can supply sites containing lithium ions and facilitate rapid transport of both electrons and lithium ions, which have the potential to replace the commercialized graphite in the future.

Failure mechanisms

Hard carbon is mostly prepared using polymer precursors, and the precursors of biomass-derived carbons are certain types of polymer substances, such as glucose, cellulose, starch, etc. As a result, the prepared and obtained BCMs are mostly hard carbon materials. Since the precursors of BCMs are polymers, the reactions in the pyrolysis process are complex, including dehydrogenation, condensation, hydrogen transfer and isomerization occurring simultaneously and releasing H₂, CH₄, CO, CO₂ and other gases during the pyrolysis process. During the pyrolysis, the precursors evolve gradually from a 3D amorphous phase to short-range ordered graphite microcrystals in the form of a 2D regular graphitized carbon layer. Due to the existence of molecular cross-linking and covalent C-O-C bonds in biomass materials, it is easier to form rigid cross-linked structures. The ordered regions are interconnected by twisted layers of graphitized carbon, which inhibits the growth and orientation of graphite layers during the carbonization stage, and hinders the directional movement of electrons. As a result, the electrical conductivity of hard carbon materials derived from biomass is lower than that of graphite^[125,126].

Table 1 shows that most biomass materials used in LIB electrodes exhibit low ICE, which is not even mentioned in many studies. Compared to graphite carbon, hard carbon has low crystallinity and poor conductivity. Although its high SSA improves the specific capacity, its ICE and reversible capacity are low as a result of the irreversible decomposition of the electrolytes and the excessive formation of the solid electrolyte interface (SEI) layer, which requires a large amount of lithium [Figure 8A and B]. During cycling, the SEI layer formed on the anode is unstable, causing the thickness and composition of the SEI layer to change constantly. The thick SEI layers increase ohmic resistance and hinder ion diffusion, resulting in a continuous increase in high voltage hysteresis and capacity decay^[127]. In addition, the pyrolyzed hard carbon materials have a small amount of oxygen elements and some residual fragments of terminal aromatics as a result of the precursors being rich in aromatic compounds and oxygen-containing functional groups which inhibit the graphitization of the precursor^[128]. Therefore, the hard carbon exhibits less graphitization and more holes and defects, causing the Li⁺ ions to bond to these sites and produce deposits [Figure 8C]^[129]. In this case, the removal of Li⁺ ions from these locations shifts the potential toward a higher voltage, resulting in a voltage hysteresis [Figure 8D]^[130].

Figure 8. Schemes of lithium storage and failure mechanisms of biomass carbon. (A) Schematics showing bulk lithium storage in the mesopores and micropores; (B) Diagram of an unstable SEI film; (C) Schematic showing lithium storage mechanism of lithium ions in biomass carbon materials; (D) Discharge/charge curves of wheat straw-derived carbon. (A) is quoted with permission from Adams et al.^[125]; (B) from Yao et al.^[130]; (C) from Zhao et al.^[126]; and (D) from Chen et al.^[129].

Despite its great importance, there are few studies on failure mechanisms of BCMs. The analysis of BCM failure mechanisms here employs a method similar to that used for hard carbon because both types of carbon share a similar lithium-ion storage mechanism, involving a combination of intercalation and adsorption. The intercalation/deintercalation of lithium ions occurs at the low voltage, and the adsorption/desorption and possible additional intercalation take place at the higher voltage. Therefore, the failure mechanisms can be analyzed from these two aspects. On the one hand, the principle of intercalation and delamination is similar to that of graphite, but most of the BCMs have a porous structure and hence a larger SSA than that of graphite. As a result, more lithium ions are consumed when the SEI layer is formed, eventually leading to the depletion of lithium ions and the failure of the material. On the other hand, excessive lithium ions adsorbed in the pores after multiple cycles will block the pores, resulting in less lithium ions being able to adsorb and desorb, causing shortening and tilting of the electrochemical platform of biomass-derived carbons, as shown by increasing voltage hysteresis. In addition to these two aspects, there may be other undiscovered causes^[131]. Therefore, future research on BCMs should focus more on analyzing failure mechanisms to provide a scientific basis for designing viable solutions to these problems and further improving the electrochemical performance of BCMs.

Solutions

Based on the two reasons summarized from the above failure analysis, there are two prominent pathways to improving BCMs: enhancing the degree of graphitization and prelithiation, which will boost ICE and reduce irreversible capacity.

Improving the degree of graphitization

Two common methods to improve the degree of graphitization are increasing the carbonization temperature and using catalysts. With the increase in the processing temperature, both the degree of graphitization of BCMs and the degree of long-range order grow significantly. Although the crystallites/domains of graphite are small, they exhibit a considerable short-range order and have evolved into a more graphitized structure in a clearly identifiable domain. This feature is caused by the order enhancement and the increase of crystallinity in the nanoscale domain^[132]. Wang et al. used a hydrothermal method to carbonize water chestnut shells at 900, 1,100, 1,300 and 1,500 °C (HT-X, X: carbonization temperature) [Figure 9A]^[133]. The length of the platform became shorter and the degree of graphitization was found to increase with the temperature of carbonization, indicating the generation of SEI film for samples carbonized at higher temperatures [Figure 9B]. ICE was found to increase accordingly from 53% to 67% [Figure 9C], with less change in irreversible capacity [Figure 9D].

Figure 9. Solutions by improving the degree of graphitization of biomass carbon. (A) TEM and SAED images of HT-1500; (B) Raman spectra of the water chestnut-based hard carbon at different carbonization temperatures; (C) Cycling performance of the water chestnut-based hard carbon at various carbonization temperatures; (D) Rate performance of the water chestnut-based hard carbon at multiple carbonization temperatures; (E) Preparation progress of MDF-Fe carbon; (F) Degree of graphitization of Fe-catalyzed MDF and uncatalyzed MDF carbon; (G) Rate performance of MDF-Fe carbon by pyrolysis between 850 and 2,000 °C. (A-D) are quoted with permission from Wang et al.^[133]; and (E-G) from Gomez-Martin et al.^[135].

In addition, the use of catalysts could reduce the temperature and improve the efficiency of reaction, while enhancing the degree of graphitization. The types of catalysts used in biomass-derived carbon for LIBs are roughly divided into three types: elemental, compound, and alloy catalysts. The catalyst and biomass are mixed to react and form carbides, which decompose at high temperatures to form graphite. The mechanism of the catalysis in the process of graphitization is the catalyst to participate in the C-C bond recombination in the conversion process^[134]. Gomez-Martin et al. synthesized hard carbon anodes (MDF-Fe) by graphitizing the medium-density recycled wood of fiberboard using an iron catalyst at 2,000 °C [Figure 9E]^[135]. Compared with the hard carbon produced without an iron catalyst, the structural order and degree of graphitization of MDF-Fe were higher, with the degree of graphitization reaching 70% at 2,000 °C [Figure 9F]. The MDF-Fe formed as such exhibited a significant increase in the reversible capacity and a doubling of the ICE from 32% to 64%, resulting from the decrease in the amount of non-basal plane areas and SSA, which decreases the formation of SEI layers [Figure 9G].

Prelithiation

Prelithiation refers to the addition of lithium ions prior to battery assembly. It aims to compensate for the loss of active lithium caused by SEI formation and other side reactions. After prelithiation of carbon materials, a complete and stable SEI film will form on the surface, thereby increasing the reversible specific capacity and ICE. Four major approaches for prelithiation include direct contact with lithium metal, Li-rich material additives, electrochemical prelithiation, and chemical prelithiation^[136]. Electrochemical prelithiation is one of the simplest and most common methods, which is a slow process but performs a precise and uniform prelithiation. This strategy not only ensures a more uniform SEI film, but also limits crack formation, which can effectively improve electrochemical performance^[137].

Prelithiation can be performed on both cathodes and anodes. Huang et al. prelithiated waste feather-derived active carbon (prAC) [Figure 10A]^[138]. Li₂C₂O₄ was used as the source of lithium, which broke down into Li₂O nanowires. Compared with the waste feather-derived carbon material without prelithiation (AC) and the carbon material directly added to Li₂O (AC + Li₂O), prAC exhibited a higher energy density of 122 Wh kg^-1 at 220 W kg^-1 and a superior cycling performance [Figure 10B and C]. Drews et al. used spruce wood to synthesize hard carbon materials for LIBs and employed the electrochemical prelithiation method to develop a lithium-ion full cell of enhanced cycle life^[139]. The electrode performance after prelithiation was significantly improved, showing a significant increase in ICE from 63% to 87%. After 150 cycles, the prelithiated full cell retained a capacity of 80% as compared with only 66% for the non-prelithiated full cell. In addition, great efforts have been made to dope graphene with P-type dopants, such as O, P, N, etc., to promote the formation of an artificial SEI film^[140]. These elements are more electronegative than carbon, adsorbing at the stable site of graphene and forming an artificial SEI film. Electrons are transferred from the graphene sheet to P-type dopants, resulting in a depletion of electrons and an increase in the hole concentration on the graphene plane, leading to an increase in the work function of the whole system and a decrease in formation of SEI films. Therefore, proper control of the work function is clearly one of the prelithiation methods, which can reduce the subsequent SEI film formation and improve ICE. Ren et al. doped Cl in hard carbon derived from cotton to obtain anode materials (Li//Cl-HC) that can inhibit the reduction of solvent molecules in the electrolyte and reduce the formation of SEI film [Figure 10D]^[141]. Compared with no Cl doping (Li//HC), the ICE of the prelithiated anode material increased from 65% to 78% [Figure 10E]. The cycle stability was also improved [Figure 10F].

Figure 10. Solutions by prelithiation of biomass carbon. (A) Schematic illustration of the prelithiation mechanism of prAC; (B) Gas chromatogram of the prAC cathode over the first five cycles; (C) Rate performance of the prAC cathode after prelithiation, AC and AC + Li₂O cathode; (D) Cycle performance of the prAC cathode after prelithiation, AC and AC + Li₂O cathodes at a density of 1 A g^-1 for 10,000 cycles. (E) Mechanism of Cl doping prelithiation; (F) Initial charge-discharge curves at 50 mA g^-1 of Li//HC and Li//Cl-HC half-cells; (G) Rate performance of Li//HC and Li//Cl-HC half-cells; (H) Long cycle performance of Li//HC and Li//Cl-HC half-cells. (A-D) are quoted with permission from Huang et al.^[138]; and (E-H) from Ren et al.^[141].