Lignin derived carbon materials: current status and future trends

Abstract

Developing novel techniques to convert lignin into sustainable chemicals and functional materials is a critical route toward the high-value utilization of lignocellulosic biomass. Lignin-derived carbon materials hold great promise for applications in energy and chemical engineering, catalysis and environmental remediation. In this review, the state-of-art sciences and technologies for controllable synthesis of lignin-derived carbon materials are summarized, pore structure engineering, crystalline engineering, and morphology controlling methodologies are thoroughly outlined and critically discussed. Green chemical engineering with cost-effectiveness and precise carbonization tuning microstructure are future research trends of lignin-derived carbon materials. Future research directions that could be employed to advance lignin-derived carbon materials toward commercial applications are then proposed.

Highlights

✧ Recent developments in lignin-derived carbon materials for energy and environmental applications.

✧ Engineering methodologies of pore structure, crystalline, morphology, and surface chemistry are critically summarized.

✧ New opportunities for lignin-derived carbon materials in commercial applications are proposed.

1 Introduction of lignin

Environmental conservation and efficient utilization of sustainable resources are two of the primary challenges for human civilization. Sustainable lignocellulosic biomasses are the most widely distributed, abundant, and renewable organic carbon resources on the earth. High-value utilization of lignocellulosic biomasses to reduce the usage of non-renewable petroleum-based chemicals is significant for the reduction and neutralization of carbon dioxide (Chen et al. 2020a; RameshKumar et al. 2020; Li et al. 2021c; Liu et al. 2021b). Therefore, it is essential and urgent to develop biomass-derived chemicals and materials for modern industry.

Cellulose, hemicellulose, and lignin are the three main organic components of plant lignocellulosic biomass. Efficient utilization of these three components in biomass is the central scientific obstacle to biomass valorization. Basically, cellulose is commercially utilized in paper-making and biorefinery industries (Fig. 1). Cellulose is usually applied in the manufacturing of papers. The cellulose components in biomass could also be hydrolyzed into glucose which afterward could be converted to fuels or chemicals such as ethanol or 2,5-furandicarboxylic acid (Takkellapati et al. 2018; Huang et al. 2022). At present, scientists have a great interest in utilizing cellulose to produce nano cellulose fibers (NCF) for functional materials (Hu et al. 2013b), while hemicellulose could be converted into lactic acid and furfural. Ethanol could be applied for the production of bio-jet. 2,5-furandicarboxylic acid, lactic acid, and furfural are the precursors for bio-plastics for packaging. Lignin is a by-product produced in the above paper-making and biorefinery industries. Kraft lignin (KL), alkaline lignin (AL), sodium lignosulfonate (SLS) are produced in the paper-making industry, and enzymatic hydrolysis lignin (EHL) is produced in the biorefinery industry. Currently, SLS is used as commercial dispersing agents and surfactants (Li et al. 2017, 2018; Lou et al. 2019; Zhan et al. 2019). KL, AL and EHL are mainly utilized as low-value fuels for boilers for energy recovery, although they could be valuable bio-precursors for various carbonaceous electrode materials. The origins of different lignin derivates are described in our previous review article (Zhang et al. 2021g).

Fig. 1

A roadmap toward the sustainable utilization of lignocellulosic biomasses

Among the three main components of lignocellulosic biomass, lignin is the only aromatic polymer with benzene rings in its building block with a phenylpropane structure. Lignin has many attractive properties, e.g., low price, high carbon content (> 60%), high aromaticity, and abundant oxygen functional groups that offer good tunability in chemical structure, etc. (Isaac et al. 2019; Wang et al. 2022b). The annual production of lignin is approximately 500–3,600 million tons worldwide (Wang et al. 2021c). They are mainly produced in paper-making and biorefinery industries as by-products, as mentioned previously. The primary lignin derivatives are KL/AL and SLS. At present, only a tiny fraction (< 10%) of total lignin produced in the industry has been utilized as chemicals and materials, such as concrete additives, stabilizing agents or dispersants, and surfactants, while most of the lignin is discarded as waste or burnt as low-grade fuels.

Carbon materials have been widely used in energy and environmental applications owing to their relatively high conductivity, chemical and mechanical stability, considerable specific surface area (SSA), and tunable pore structure (Yin et al. 2020a; Zhu et al. 2020). The state-of-art commercial carbon materials are mainly produced from byproducts of petroleum refining, such as petroleum coke and pitch, and petroleum-derived polymers such as polyacrylonitrile (PAN) (Marsh and Rodríguez-Reinoso 2006). Lignocellulosic biomasses are not often utilized for carbon materials due to their heterogeneity and hard chemical tunability. Nevertheless, lignocellulosic biomasses could be transformed into a variety of carbon materials by separation, purification, morphological control, controllable carbonization and activation processes (Liu et al. 2015a; Borghei et al. 2018; Sundriyal et al. 2021). The aromatic structure and chemical advantages of lignin resources, the motivation to reduce the contamination of lignin byproducts, and the substitute of petroleum-based chemicals by lignin have intensified the research interests in designing controllable preparation methodologies of lignin-derived carbon (LDC) materials.

In lignocellulosic biomass, lignin is present in the spaces between cellulose and hemicellulose, playing mainly as a binder in the biomass matrix (Fig. 2a) (Liu et al. 2015b). The structure of lignin is highly branched compared with cellulose and hemicellulose (Fig. 2b) (Zakzeski et al. 2010). Lignin consists of three typical hydroxycinnamyl alcohols, i.e., p-coumaryl, coniferyl, and sinapyl alcohols, corresponding to the three subunits of hydroxyphenyl (H), guaiacyl (G), and syringyl (S) in lignin (Fig. 2c) (Shen et al. 2017). There are approximately 50% aromatic compounds in lignin (Liu et al. 2015b). In a lignin molecule structure, there are various linkages of ether and C–C bonds among the three subunits, typically β-O-4, α-O-4, 4-O-5, β-1, 5–5’, β-β and β-5 bonds (Fig. 2d) (Chakar and Ragauskas 2004). The proportion of C–O–C bonds is the most abundant in lignin, generally existing as either ethers or a part of furan rings (Liu et al. 2015b). The β-O-4 linkage is dominant among these C–O–C linkages in lignin (Zakzeski et al. 2010). Besides, the presence of various oxygen-containing groups on the C_α and C_β of lignin, including methoxy, carbonyl, and hydroxyl groups, etc., further contributes to the complexity of lignin molecular structure (Zakzeski et al. 2010).

Fig. 2

a Structural illustration of lignin components in lignocellulosic biomass. Reproduced with permission from Ref. (Liu et al. 2015b). Copyright © 2015, Roal Society of Chemistry; b Typical branched structure of lignin. Reproduced with permission from Ref. (Zakzeski et al. 2010). Copyright © 2010, American Chemistry Society; c Basic units in lignin structure. Reproduced with permission from Ref. (Shen et al. 2017). Copyright © 2017, Progress in Chemistry; d Typical linkages in a lignin molecule structure. Reproduced with permission from Ref. (Chakar and Ragauskas 2004). Copyright © 2021, Elsevier

Lignin has a branched and random structure, hence is a complex, heterogeneous and three-dimensional polymer. Furthermore, the abundant aromatic sub-structure rich in β-O-4 linkage accounts for the hydrophobic properties of lignin. Therefore, natural lignin is difficult to be dissolved in water at acidic or neutral pH values, but lignin can be dissolved in some alkaline solutions (ammonium, hydroxide or carbonates) or some organic solvents (Li et al. 2017; Lou et al. 2019; Wang et al. 2021b).

Based on the chemical and physical characteristics of lignin, one can design various carbonization strategies to synthesize a variety of carbon materials with tunable morphology, crystalline structure, pores structure, and surface chemistry for various industrial applications in battery, supercapacitor, catalysis and environment (Fig. 3). There have been a couple of review papers on lignin-based electrode materials for energy storage (Espinoza-Acosta et al. 2018; Zhu et al. 2020; Liu et al. 2021b), and carbon materials from lignin precursors (Li et al. 2021b; Qu et al. 2021; Wang et al. 2021c). It is necessary for us to summarize the designing principles of carbonaceous materials from lignin and their structure-performance relationship toward energy and environmental applications.

Fig. 3

Multi-dimensional strategies for the synthesis of LDC materials and their industrial applications

2 Lignin-derived carbon materials (LDC)

2.1 Carbonization

A variety of carbonization techniques, such as thermal carbonization (Leng et al. 2022), hydrothermal carbonization (Hu et al. 2010; Zhou et al. 2016), flash light carbonization (Gengler et al. 2013; Choi et al. 2016; Kim et al. 2019; Song et al. 2021), and laser carbonization etc. (Wynn and Fountain 1997; Vander Wal and Choi 1999; Zhang et al. 2019d, c), have been employed for the production of carbonaceous materials from lignocellulosic biomass (Zhang et al. 2021g). In industrial practice, thermal carbonization is usually employed to obtain carbonaceous materials. In terms of the carbonization of lignin, lignin experiences solid-state carbonization, i.e., lignin does not melt during thermal treatment, resulting from the oxygen-rich backbone of the lignin molecules. As a result, LDCs are amorphous hard carbon materials rich in sp³ hybridized C–C bonds and are hard to be graphitized at high temperatures, which is in sharp contrast to the petroleum-based precursors (petroleum coke and pitch) (Liu et al. 2020). The carbonization of lignin can be divided into three stages (Fig. 4), i.e., dehydration stage (30–200 °C), active pyrolysis stage (200–450 °C) and passive pyrolysis stage (> 450 °C) (Liang et al. 2015; Liu et al. 2016a; Ma et al. 2016; Leng et al. 2022).

Fig. 4

A summary of pyrolytic characteristics in various pyrolysis stages of lignin molecule. Reproduced with permission from Ref. (Leng et al. 2022). Copyright © 2022, Elsevier

Dehydration reaction occurs in the dehydration stage, generally below 200 °C. The active pyrolysis stage is the primary process of eliminating the heteroatoms in lignin. During the active pyrolysis stage in the temperature range of 200–450 °C, unstable β-O-4 linkages are broken (Liu et al. 2016a). The C–C and C-O bonds are further cleaved with increasing temperature (Li et al. 2014). Besides, the functional groups and the branches (methoxyl, methyl, and carboxyl) decompose forming the gas products (CO₂, CO, CH₄, etc.) (Fenner and Lephardt 1981; Liu et al. 2008; Hu et al. 2013a; Li et al. 2014; Zhao et al. 2014).

Due to the massive elimination of heteroatoms, extensive rearrangement reactions occur in solid products together with the release of volatile macromolecular products. The scission and rearrangement of side-chains, the removal of functional groups, and polymerization mainly occur in the passive pyrolysis stage (> 450 °C). The elimination rate of heteroatoms becomes slow, resulting in the formation of small molecule compounds (olefins, alkyl, and aromatic products) accompanied with releasing less amount of gas (CO₂, CO, CH₄, etc.). The slow decomposition and carbonization of lignin residues (Yang et al. 2007; Liang et al. 2015), and the polymerization of volatile matters occur to form coke. The coke prepared at low temperature is generally called amorphous carbon which is apparently different from hard carbon (obtained under carbonization temperature > 1000 °C). The carbon layers in amorphous carbon are randomly stacked to form a turbostratic structure, although small graphene nanodomains (GNDs) are also formed, which could enlarge with increasing temperature. However, these GNDs are highly cross-linked; therefore, lignin is difficult to be graphitized to form artificial graphite (Zhang et al. 2019e). In this review, LDCs refer to the carbonaceous materials derived from lignin under a carbonization temperature above 500 °C. The fewer GNDs in an LDC carbon matrix and the irregular distribution of GNDs endow its amorphous characteristics. From this perspective, LDCs are not suitable carbon sources for preparing graphite and graphene, whereas certain chemical processes could enable the production of graphene-like amorphous carbon nanosheets (Liu et al. 2017) and graphene quantum dots (Si et al. 2018; Shi et al. 2019; Zhang et al. 2019a) from lignin.

2.2 Graphitization

Graphitization produces carbon materials with a high graphitization degree and high conductivity. Graphitic carbon materials could be used as anodes of lithium-ion and potassium-ion batteries. Regardless of the recently-developed novel laser or flash carbonization techniques (Wynn and Fountain 1997; Vander Wal and Choi 1999; Gengler et al. 2013; Secor et al. 2015; Song et al. 2021), high-temperature carbonization (> 2000 °C) is a common and effective strategy to eliminate the disordered structure and improve the graphitization degree of LDCs (Kurban et al. 2010). However, as discussed above, the hyperbranched, oxygen-rich, and complex network of lignin restricts the order rearrangement of the carbon layer in the carbonization process, making it difficult for the production of highly graphitized carbon materials from lignin (Torres-Canas et al. 2020). As a result, high-temperature treatment only results in the formation of hard carbons. The GNDs and amorphous regions are randomly cross-linked in hard carbons (Zhang et al. 2021g). The microporous structure is formed due to the disorderly stacked GNDs. With increasing temperature, the number of graphene stacking layers increases, and the interlayer spacing grows, leading to the increase in the size of the inner micropores (Kim et al. 2017; Fromm et al. 2018). Catalysts such as nickel, iron, or molybdenum catalysis proved to be effective for the graphitization of biomass or non-graphitizable precursors. Gindl-Altmutter et al. demonstrated that magnesium could play a catalytic role in improving the graphitization degree of LDC materials (Gindl-Altmutter et al. 2019). On the other hand, carbonization with some guide agents, such as graphene oxide, pitch and petroleum coke, could result in LDCs with a higher graphitization degree.

3 Morphology control of LDCs

3.1 Carbon fibers

Given the renewable characteristic of biomass, lignin is an ideal precursor for carbon fiber production. However, the spinnability of lignin is poor. Although lignin could be melt-spun in an energy-efficient way compared with wet-spinning, the spinning of various lignin is only possible in a mixed polymer solution (Kim et al. 2015; Liu et al. 2016b). In practice, melt-spinning, wet-spinning and electrospinning are usually used to prepare LDC fibers.

The preparation route for the production of LDC fibers is depicted in Fig. 5a (Fang et al. 2017). In a typical preparation process, lignin is firstly pretreated by purification, refining, additional modification, and blending. The pretreated lignin solution is spun into fibers via extrusion approaches, including melt-spinning, wet-spinning, dry-spinning, gel-spinning, or electrospinning (Lallave et al. 2007; Ruiz-Rosas et al. 2010). Then, the lignin-based fibers are oxidatively thermostabilized under an oxygen-containing atmosphere to form a cross-linked carbonaceous structure at a slow heating rate, which is one of the critical procedures in carbon fiber fabrication (Lallave et al. 2007; García-Mateos et al. 2018). After thermal stabilization, the LDC fibers further experience the carbonization process at temperatures above 600 °C, when the condensed structure is formed and the LDC fibers gain high mechanical strength. Particularly, modification treatments (e.g., activation or graphitization) could be further performed to obtain activated carbon fibers and graphitic LDC fibers with enhanced structural or functional properties (Sagues et al. 2019).

Fig. 5

a Typical preparation processes for LDC fibers. Reproduced with permission from Ref. (Fang et al. 2017). Copyright © 2016, Roal Society of Chemistry; b Methodologies of LDC fibers by electrospinning technique. Reproduced with permission from Ref. (Svinterikos et al. 2020). Copyright © 2020, American Chemistry Society; c A schematic showing the assembly of PEO and lignin. Reproduced with permission from Ref. (Imel et al. 2016). Copyright © 2016, American Chemistry Society; d Images of carbon fibers from different types of lignin. Reproduced with permission from Ref. (Shi et al. 2018). Copyright © 2020, American Chemistry Society

Figure 5b shows the general methodology for the production of LDC fibers using the electrospinning technique (Svinterikos et al. 2020). Generally, electrospun lignin fibers are prepared from a mixed solution of lignin and other polymers, e.g., polyacrylonitrile (PAN), polyvinyl alcohol (PVA), and polyethylene oxide (PEO), with better spinnability. Loscertales et al. prepared LDC fibers by electrospinning pure Alcell lignin without using other polymer binders (Lallave et al. 2007). Alcell lignin or called organosolv lignin is another lignin derivative from Alcell pulping process, which contains less impurities than Kraft lignin. The direct electrospinning of lignin is nevertheless rather difficult. Therefore, it is necessary to add another polymer with high molecular weight (such as PAN, PVA and PEO) into the electrospinning solution. PAN is usually chosen as a binder in the lignin solution due to its easy spinnability.

Lee et al. prepared lignin/PAN-based carbon fibers at mass ratios from 0/100 to 50/50 (lignin/PAN) (Choi et al. 2013), where the PAN used in this work had a molecule weight of 150,000. The obtained fibers were further thermostabilized, treated under 250 °C for 3 h in air and carbonized at 1000 °C for 1 h in an inert atmosphere. The carbon fibers became thinner, and the SSA decreased with increasing lignin content. Wang et al. found that increasing lignin content resulted in a decrease in average diameter and an increase of SSA (Wang et al. 2018). Ferraris et al. used Kraft lignin and PAN with mass ratios from 10/90 to 70/30 to prepare lignin/PAN-based carbon fibers (Perera Jayawickramage et al. 2019), followed by thermal stabilization, carbonization and physical activation at 1000 °C in CO₂ flow. With the highest lignin blending ratio (PAN: lignin = 70:30), the obtained carbon fiber achieved the largest SSA of 2543 m² g⁻¹ and excellent conductivity of 530 S m⁻¹. Beck et al. prepared lignin/PVA-based carbon fibers with a lignin/PVA blend at a mass ratio of 70/30, and obtained carbon fibers with average diameters less than 200 nm, SSA of 583 m² g⁻¹ and average pore sizes around 3.5 nm (Beck et al. 2017). The high SSA was believed to result from the phase separation process during the carbonization of the PVA/PAN blend. PEO is another choice for blending with lignin in the electrospinning process. As illustrated in Fig. 5c, PEO (1–2 wt%) can form hydrogen bonds with the hydroxyl groups of lignin macromolecules, leading to abundant chain entanglements, which would benefit electrospinning (Imel et al. 2016). Shi et al. studied the carbonization behaviors of blends of three lignins and PEO at a mass ratio of 95/5 (Fig. 5d) (Shi et al. 2018). The low-molecular-weight ethanol organosolv lignin could not be thermostabilized in the melting process. The formic acid/acetic acid organosolv lignin-derived carbon fibers showed poor mechanical properties due to the existence of many side chains. Because of the better orientation and fewer side-chains in lignin macromolecules, the alkaline LDC fibers displayed a high graphitization degree and high tensile strength of 15.58 MPa.

3.2 Carbon spheres

Carbon spheres and hollow carbon spheres are promising carbon materials for energy storage and environmental applications. The carbonization of colloidal spheres and spheres derived from spray drying are the common methods to prepare LDC spheres. The self-assembly technique for lignin nanospheres is the key to the preparation of the spherical morphology of lignin. Lignin nanospheres were prepared by solvent exchange (solvent and antisolvent) (Xiong et al. 2020) followed by spray drying or in an aerosol flow reactor (Xiong et al. 2017). Monodispersed lignin colloidal spheres can be prepared by a coupled strategy of solvent extraction, solvent/antisolvent self-assembly (Wang et al. 2022d). Solvent exchange is the most popular and efficient methodology to assemble lignin nanospheres owing to its simple and environmentally friendly advantages. The inner interaction of the lignin nanosphere is by weak intermolecular forces, generally hydrogen bonding, π-π interactions, electrostatic and hydrophobic forces (Wang et al. 2020a). The lignin nanospheres have poor solubility resistance and thermostability, and the spherical structure could hardly be maintained in organic solvents, such as tetrahydrofuran (THF), acetone, and ethanol (Xiong et al. 2017; Wang et al. 2020a).

Furthermore, lignin would undergo severe and irreversible condensation under the high-temperature carbonization process (Shuai et al. 2016). Therefore, how to prepare intact LDC spheres from the carbonization of the lignin colloidal nanosphere is another problem to be solved. Hydrothermal and thermal carbonizationa are two efficient ways for the preparation of carbon spheres. Chemical cross-linking could strengthen the three-dimensional network structure and hence improve the physical and chemical properties of the lignin spheres (Souza et al. 2019; Wang et al. 2019a).

Wu et al. prepared lignin nanospheres by solvent exchange method followed by covalent cross-linking by hydrothermal treatment, as illustrated in Fig. 6a (Wang et al. 2021a). Hydrothermal temperature played a significant role in fracturing β-O-4/C_α-C_β bonds, hydroxyl and –OCH₃ groups, in which free radicals form in the lignin nanospheres. The adjacent intramolecular and intermolecular radicals could then be crosslinked via α-5, β-5 and β-β′ bonds. The higher the temperature is, the higher the crosslinking degree is; therefore, the smaller the diameter of the lignin nanosphere is. The carbon spheres prepared from the lignin nanospheres maintained perfect spherical structures and displayed a small graphitic degree and high carbon content. Qiu et al. developed a novel strategy to prepare novel uniform lignin-derived nitrogen-doped (5.73%) carbon nanospheres by direct pyrolysis of colloidal spheres of lignin-based azo polymer (AL-azo-NO₂) at 750 °C (Zhao et al. 2016). The percentage of the rigid segment around the hydrophilic segments increased after diazotization treatment. The parallel arrangement of rigid azo benzene segment drives the formation of uniform and tight colloidal spheres of a size of about 200–300 nm (Deng et al. 2016), where a large number of uniform spheres (500 nm) are also formed in the pyrolysis process.

Fig. 6

a Schematic diagram of the preparation of lignin-derived nanosphere. Reproduced with permission from Ref. (Wang et al. 2021a). Copyright © 2021, American Chemistry Society; b Schematic diagram of the preparation of hollow carbon sphere from Kraft lignin. Reproduced with permission from Ref. (Cao et al. 2021a). Copyright © 2021, Elsevier

Spray drying technique is a scalable and effective method to synthesize carbon spheres (especially hollow carbon spheres) from lignin precursor solution via a simple one-step process. Chen et al. firstly prepared a spherical lignin composite by spray drying of a mixture of lignosulfonate and KOH (Chen et al. 2018). In this synthesis strategy, KOH was used to dissolve lignin so as to form a homogeneous solution. Upon spray drying, lignin/KOH hollow structures were formed. After the simultaneous carbonization and activation at high temperatures, the porous carbon spheres showed an SSA of 1372 m² g⁻¹ and an appropriate pore structure (average pore size of 2.59 nm). Li et al. synthesized plum-like hard carbon microspheres (1–2 μm) from sodium lignosulfonate by spray drying and carbonization. The sodium SLS-derived hard carbon microspheres have enlarged interlayer spacing, fewer defects, and a small SSA of 11.89 m² g⁻¹ (Li et al. 2020a). This hard carbon anode showed superior sodium storage performance. Cao et al. used Kraft lignin as a carbon precursor and KOH as an activation agent to prepare hollow carbon spheres with high SSAs of 1536 – 2424 m² g⁻¹ and hierarchical structure (Cao et al. 2021a) (Fig. 6b). Cao et al. further precisely controlled the spherical morphologies from dense to hollow spheres by tuning the KOH concentration (Cao et al. 2021b). A high concentration of KOH resulted in the formation of hollow structures, while the highest mass ratio of KOH to lignin used was below 1.5:1. Given the low consumption of KOH for the production of activated carbon spheres with a high SSA, it is a promising method for large-scale production. Pang et al. synthesized oxygen/nitrogen co-doped hierarchical porous hollow carbon microspheres (0.2–5 μm) from sodium lignosulfonate by spray drying, thermal stabilization, carbonization, and post nitric acid modification. The as-prepared hollow carbon microspheres have an SSA of 991 m² g⁻¹, a total pore volume of 0.75 cm³ g⁻¹, and abundant surface contents of oxygen (13.12 at.%) and nitrogen (0.97 at.%) (Pang et al. 2018b).

Hydrothermal technique and reverse phase polymerization coupled with a carbonization process can also be used to prepare LDC microspheres. Fan et al. synthesized nitrogen-doped lignin microspheres by lignin self-decomposition and self-polymerization via hydrothermal process at 250 °C and carbonization at 1100 °C. The nitrogen-doped LDC microspheres exhibit well-developed spherical morphology, ultramicroporous (< 0.7 nm) structure, and large interlayer spacing, which endows it with superb sodium storage performance with high specific capacity and stable cycling stability (Fan et al. 2021). Yu et al. prepared a LDC microsphere using reverse-phase polymerization and direct carbonization without pre-oxidization. The morphology, size distribution and thermal stability of LDC microsphere could be controlled by coordinating the solid content and dispersed phase content of lignin emulsion (Yu et al. 2018).

3.3 Carbon foam

Carbon foam has attracted great attention due to its low density, high conductivity, high porosity, and tunable pore size. LDC foam is generally prepared by drying and carbonization of phenolic wet gel. Lignin is an appropriate candidate to replace resorcinol or other phenolic substances to prepare phenolic resin due to the similar chemical moieties of phenolic, carboxyl, carbonyl, hydroxyl groups (Vázquez et al. 1999; Jin et al. 2010). Additionally, the three-dimensional network structure and high carbon content of lignin are conducive to preparing carbon foam from a sustainability perspective.

Seo et al. synthesized LDC foams with porous structural frameworks using a thermal decomposition route at a temperature of 800 °C. The polycondensation reaction of lignin, resorcinol, and formaldehyde into crosslinked phenolic resin network was performed with colloidal poly(methyl methacrylate) microspheres as a sacrificial template. After carbonization, the carbon foam had a bulk density of 0.37–0.60 g mL⁻¹ and porosity of 68.5–82.8%. The LDC foam exhibited excellent mechanical strength (Seo et al. 2014). Qu et al. used lignin to replace 25 wt.% phenol to prepare lignin-phenol–formaldehyde resole resin in an alkaline environment. Then, the carbon foams were synthesized using the resin as a carbon source and polyurethane foam as the template. The as-prepared carbon foams showed open macropores, low bulk density, and good water repellency (Qu et al. 2017). Xu et al. synthesized lignin-based carbon foams by pyrolysis and KOH activation (Xu et al. 2018a), where enzymatic hydrolysis lignin, resorcinol and formaldehyde were mixed under Na₂CO₃ catalysis to prepare lignin-based aerogels after gelation, aging and ambient drying. The as-prepared carbon foam has a high SSA of 779 m² g⁻¹, a large total pore volume of 0.48 cm³ g⁻¹ and a large micropore volume of 0.29 cm³ g⁻¹. Furthermore, the carbon foam displayed an interconnected, hierarchical porous network structure and a high degree of graphitization (Xu et al. 2018a). The carbon foams are essential for some applications that need macropores.

4 Pore engineering of LDCs

The pore structure is an important parameter of carbon materials. Developing the pore engineering method is a long-last research theme since LDPCs with high SSA is important for many applications, such as capacitive energy storage, adsorption of inorganic or organic pollutants, and batteries. The pore engineering strategies for LDPCs can be classified into physical activation, chemical activation, template methods and direct carbonization methods (Fig. 7). There are some review papers focusing on the synthesis strategies of porous carbon materials (Borchardt et al. 2018; Jin et al. 2018; Ghosh et al. 2020; Shao et al. 2020; Yin et al. 2020a) and LDPCs (Zhu et al. 2020; Zhang et al. 2021g; Wang et al. 2022a). Herein, in this paper, we only briefly overview some general synthesis strategies, i.e., the traditional activation and template methods and novel direct carbonization strategies, for LDPCs.

Fig. 7

A summary of strategies of pore engineering for lignin-derived porous carbons (LDPCs)

4.1 Physical activation

Physical activation is also called gaseous activation. The principle of pore engineering of physical activation lies in that gaseous activation agents, such as CO₂, NH₃ and steam, gasify carbon matrix at 600–1200 °C. CO₂ activation has been used as a physical activation agent for carbon materials for a long time. CO₂ activation results in the etching of carbon matrix to form CO. Schlee et al. produced Kraft lignin-derived carbon fiber mats by carbonization at 900 °C for 30 min followed by CO₂ activation at 800 °C for 60 min. The porosity of the LDPC with CO₂ activation was more abundant than that without CO₂ activation. It was proposed that CO₂ activation expanded micropores to form mesopores and created new micropores. The carbon fiber mats with CO₂ activation showed a higher SSA of 1204 m² g⁻¹ than that without CO₂ activation (676 m² g⁻¹) (Schlee et al. 2019). Jayawickramage et al. prepared carbon fiber mats derived from lignin and PVA by electrospinning, and then carbonization at 1000 °C and CO₂ activation at 800 °C. With the decrease of lignin content in the lignin/PVA composite, the SSA, micropore volume and mesopore volume of LDC fiber mats firstly increased and then decreased. The LDC fiber mat derived from lignin/PVA blend of 80/20 exhibits a high SSA of 2,170 m² g⁻¹, a mesopore volume of 0.622 cm³ g⁻¹ and a micropore volume of 0.365 cm³ g⁻¹ (Perera Jayawickramage and Ferraris 2019). Jayawickramage et al. prepared carbon fiber mats from lignin and PAN using a similar synthesis route. With the increase of lignin in lignin/PAN blend, the SSA and mesopore volume increased while the micropore volume decreased. The LDC fiber mats derived from lignin/PVA blend of 70/30 have a very high SSA of 2,543 m² g⁻¹, a mesopore volume of 0.955 cm³ g⁻¹ and a micropore volume of 0.547 cm³ g⁻¹ (Perera Jayawickramage et al. 2019).

As a new physical activation method, NH₃ activation generates pores through the reaction between carbon and NH_3, producing N₂ and CH₄ (Luo et al. 2014). The precise reaction mechanism of NH₃ activation needs to be further explored. Zuo et al. proposed four successive stages in the process of NH₃ activation to engineer pore structure, which correspond to the temperature range lower than 800 °C (< 20% burn-off), the temperature range at 850–950 °C (20–75% burn-off), the temperature range at 950–1000 °C (75–80% burn-off) and the region higher than 1000 °C (> 80% burn-off). In the first stage, the reaction between NH₃ and carbon mainly contributed to the micropore formation with a 19.3% char burn-off. The SSA and total pore volume increased from 124 to 702 m² g⁻¹ and 0.02 to 0.33 cm³ g⁻¹. In the second stage, the reaction between NH₃ and carbon intensified, which mainly contributed to the formation of micropores (1–2 nm) with 74.7% char burn-off. The SSA increased from 702 to 1431 m² g⁻¹ and the total pore volume increased from 0.33 to 0.76 cm³ g⁻¹, showing a linear correlation with increasing the activation temperature. In the third stage, the porosity was remarkably developed with 5% increase in char burn-off, widening the existing micropores to form mesopores (2–4 nm), and generating new micropores (1–2 nm). The SSA increased from 1431 to 2316 m² g⁻¹ and the total pore volume increased from 0.76 to 1.48 cm³ g⁻¹. Particularly, the mesopore volume increased from 0.05 to 0.34 cm³ g⁻¹, and micropore volume increased from 0.71 to 1.13 cm³ g⁻¹. In the fourth stage, the intensified reaction between NH₃ and carbon resulted in the decrease of SSA and pore volume due to the collapse of the pore wall with a burn-off percentage above 80% (Zuo et al. 2020). Zhang et al. prepared enzymatic hydrolysis LDPC by NH₃ activation. They explored the free radical mechanism of the NH₃ activation and the relationship between nitrogen doping configurations, doping levels, and preparation temperatures. With increasing the temperature, the SSA, total pore volume and micropore volume increased from 335.9 to 1464.0 m² g⁻¹, 0.17 to 0.66 cm³ g⁻¹, respectively. Despite the decreased nitrogen content with increased temperature, the relative content of pyrrolic nitrogen increased. Therefore, it is demonstrated that the unsaturated electron-deficient radicals (NH₂· and NH·) reacted with electron-rich N6, leading to a larger decrease of N6 than that of N5 (Jian et al. 2022).

Water steam activation is used to generate pore structure by reacting with carbon to produce H₂ and CO. Generally, the steam activation rate is several times faster than that of CO₂ activation. Fu et al. studied the effect of preparation parameters (carbonization temperature, steam activation temperature and activation time) on the pore texture characteristics of LDPCs. With increasing the carbonization temperature from 250 °C to 450 °C, the SSA increased from 86.41 to 288.79 m² g⁻¹ and the micropore volume increased from 30.46 to 171.04 cm³ g⁻¹. When the carbonization temperature exceeded 450 °C, the SSA remained unchanged while the average pore diameters increased. The SSA increased from 227.43 to 288.79 m² g⁻¹ with the activation temperature rising from 700 to 725 °C, and then decreased to 193.15 m² g⁻¹ if the activation temperature was further increased to 850 °C. An activation time of 40 min was found to be suitable for preparing LDPCs with high SSAs. When the activation time increased beyond 40 min, the SSA decreased significantly (Fu et al. 2013). Carrott et al. prepared an LDPC from enzymatic hydrolysis lignin (EHL) with a SSA of 812 m² g⁻¹ and a micropore volume of 0.27 cm³ g⁻¹ by steam activation at 750 °C (Carrott et al. 2008). Baklanova et al. prepared an EHL-derived porous carbon with an SSA of 865 m² g⁻¹ and a micropore volume of 0.37 cm³ g⁻¹ by steam activation at 800 °C (Baklanova et al. 2003). It needs to be noted here that air could also be used as an activation agent; however, the reaction rate of air with carbon is too fast to be controlled; therefore, air was rarely used to construct LDPCs in practice (Zhang et al. 2021g).

4.2 Chemical activation

The pore engineering by chemical activation is to use chemical activation agents, such as KOH, NaOH, H₃PO₄ and ZnCl₂, to etch carbon matrix at temperatures in the range of 600–900 °C. Recently, CuCl₂ is also employed as a chemical activation agent for LDPCs (Wen et al. 2022).

KOH has been widely used to prepare LDPC owing to its capability to promote the formation of porous carbons with high SSA and large pore volume, which is suitable for many applications (Zhang et al. 2015c, a). The activation mechanisms of NaOH, Na₂CO₃ and K₂CO₃ are similar to that of KOH (Yin et al. 2020a). However, the SSA and pore volume of LDPCs prepared with KOH are higher than those prepared with other chemical activation agents. The KOH activation mechanisms are summarized as follows (Guo et al. 2019): (1) KOH reacts with carbon to produce K, K₂CO₃ and H₂ at 400 °C; (2) K₂CO₃ decomposes into K₂O and CO₂ above 700 °C; (3) CO₂ and H₂O react with carbon to produce CO and H₂; (4) the produced K₂CO₃ and K₂O react with carbon to produce K and CO; (5) the metallic K intercalates the carbon framework to form branch-like micropore channels.

As shown in Fig. 8a, Dai et al. used KOH as an inorganic template and chemical activation agent to prepare sodium lignosulfonate-derived LDPC with an SSA of 2,235 m² g⁻¹ and pore volume of 1.512 cm³ g⁻¹ at a lignin/KOH mass ratio of 1/3 (Tan et al. 2021). Lee et al. investigated the difference in structural characteristics of porous carbons prepared from Kraft and alkaline lignin using KOH activation. KOH could interact with NaOH in Kraft or alkaline lignin to form NaOH/KOH eutectic mixture, which contributed to the microporous and mesoporous structure in Kraft and alkaline LDPCs. In contrast, the KOH activation of an EHL-derived carbon contributed to the microporous structure. The porous carbon materials derived from alkaline lignin and EHL with KOH activation have SSAs of 1825 and 2345 m² g⁻¹, and pore volumes of 1.72 and 1.74 cm³ g⁻¹, respectively (Lee et al. 2018). Chen et al. prepared porous carbon spheres with a SSA of 1372.87 m² g⁻¹ by spray drying a solution of lignosulfonate and KOH and subsequent carbonization (Chen et al. 2018). Tran et al. prepared porous graphite oxide/lignin nanocomposite films with SSA of 1744 m² g⁻¹ using KOH activation (Tran et al. 2017). Wan et al. prepared three-dimensional hierarchical porous carbon from oxidized lignin using one-step KOH-activation and carbonization method. The lignin-derived hierarchical porous carbon shows a SSA of 3094 m² g⁻¹ and a pore volume of 1.72 cm³ g⁻¹ under a KOH/lignin mass ratio of 3/1 (Wan et al. 2021). Wang et al. prepared N-doped porous carbon from urea-modified lignin with a high SSA of 3130 m² g⁻¹ via carbonization and KOH activation (Wang et al. 2016b). Zhu et al. used KOH as both a lignin extraction solvent and an activation agent and melamine as a nitrogen source to prepare LDPC with a SSA of 2646 m² g⁻¹ and a high pore volume of 1.285 cm³ g⁻¹ (Zhu et al. 2017).

Fig. 8

Schematic diagram of the synthesis of LDPCs with hierarchical structure (a) using KOH activation strategy. Reproduced with permission from Ref. (Tan et al. 2021). Copyright © 2018, Elsevier; (b) using hydrothermal carbonization and chemical activation strategy. Reproduced with permission from Ref. (Zhang et al. 2016a). Copyright © 2016, American Chemistry Society; (c) using microwave and KOH activation strategy. Reproduced with permission from Ref. (Wang et al. 2019b). Copyright © 2019, Wiley

Hydrothermal and microwave-assisted KOH activation has also been widely reported to prepare LDPC, as illustrated in Fig. 8b and c. Guo et al. prepared interconnected hierarchical porous nitrogen-doped carbon via hydrothermal treatment and KOH activation of EHL. The obtained LDPC has a SSA of 2218 m² g⁻¹, high electronic conductivity of 4.8 S cm⁻¹ and a nitrogen doping content of 3.4 at% (Zhang et al. 2016a). Zhang et al. used sodium lignosulfonate as a carbon precursor, 1,6-hexanediamine as a crosslinking agent and nitrogen source to prepare nitrogen-doped hierarchical porous carbon by hydrothermal crosslinking reaction and KOH activation. The as-prepared LDPC has an SSA of 1867.4 m² g⁻¹, a moderate nitrogen-doped content of 3.6 at.% and a three-dimensional hierarchical porous structure with rich micropores, abundant mesopores and interconnected macropores (Zhang et al. 2018d). Li et al. prepared EHL-derived LDPC with a three-dimensional interconnected hierarchical structure and a SSA of 1504 m² g⁻¹ (Li et al. 2021a). Guo et al. used KOH to activate EHL-derived hydrochar to prepare EHL-derived LDPC at 800 °C with a KOH/char mass ratio of 1–3. The obtained LDPC has a SSA of 1290–1660 m² g⁻¹ and a three-dimensional hierarchical texture consisting of abundant micropores, as well as some mesopores and macropores, and high electronic conductivity of 4.0–5.4 S cm⁻¹ (Guo et al. 2017). Microwaves could assist the activation process and shorten the reaction time. Zhou et al. explored the effects of preparation conditions on the morphology and pore structure of LDPC sphere by microwave carbonization with KOH activation. The carbon spheres with sizes of 3–6 μm were obtained by activation at 270 °C for 7 h with a KOH concentration of 0.06 g mL⁻¹ (Mao et al. 2018). Zhou et al. prepared LDPC with KOH activation using a microwave heating method under humidified nitrogen atmosphere. The as-prepared LDPC exhibited a high SSA of 2866 m² g⁻¹, hierarchical pore structure with micropores volume of 0.70 cm³ g⁻¹ and considerable meso-/macropores volume of 1.33 cm³ g⁻¹ (Chen et al. 2019b). Zhou et al. prepared LDPC using microwave heating and KOH activation in a very short duration of 10–30 min. The LDPC displayed ultra-high SSA of 3065 m² g⁻¹, a hierarchical porous structure with a micropore volume of 0.73 cm³ g⁻¹, and a meso-/macropores ratio of 64.4% (Chen et al. 2020c). Directly mixing of KOH with lignin (Zhang et al. 2015a, 2018c), or mixing KOH with lignin-derived char (Zhang et al. 2015c) could be used to tune the micropore structure of LDPCs. Relative low SSA with micropore structure could be achieved by direct mixing, while high SSA with hierarchical micro/mesopore structure could be achieved by carbonization-activation strategy.

ZnCl₂ and H₃PO₄ are commonly used as low-temperature activation agents to prepare LDPCs. ZnCl₂ can promote gasification dehydrogenation at low temperatures and act as a supporting template (Yu et al. 2016). Phosphoric acid acts as a catalyst in the activation process to promote the fracturing of macromolecular chains and the dehydration reaction, which hence promotes the formation of pore structures in the process of thermal polycondensation and cyclization (Yahya et al. 2015). Gonzalez-Serrano et al. prepared alkali LDPC with a H₃PO₄/char mass ratio of 1. With increasing temperature from 350 °C to 500 °C, the SSA of porous carbon increased from 534 to 1347 m² g⁻¹. At temperatures above 500 °C, the SSA decreased to 1153 m² g⁻¹ (Gonzalez-Serrano et al. 1997). Yang et al. prepared sulfuric acid hydrotropic LDPC with a high SSA of over 2000 m² g⁻¹ at 450 °C using H₃PO₄ activation (Yang et al. 2020). García-Mateos et al. prepared lignin-derived microporous carbon fiber with a SSA of 2340 m² g⁻¹ at 900 °C using H₃PO₄ activation by direct carbonization lignin under an inert atmosphere or with diluted O₂ (García-Mateos et al. 2020). Brazil et al. investigated the effects of H₃PO₄ loading amount, time, temperature, and microwave power on the surface area of Kraft lignin-derived porous carbon prepared using conventional and microwave-assisted pyrolysis, respectively. The H₃PO₄ loading amount was an important parameter for the surface area of porous carbon. The SSA of the porous carbons obtained at lignin: H₃PO₄ ratio of 1:1 and 1:2 was 1150 and 1030 m² g⁻¹, respectively (Brazil et al. 2020).

4.3 Template-based methods

Template methods, including soft template and hard template methods, have been widely used to synthesize lignin-derived mesoporous carbon. The pore-forming mechanism of the soft template method, in principle, is to form the composite micelles by electrostatic or hydrogen bond self-assembly of soft template and lignin. The soft template decomposes during the carbonization process to produce mesopores. The fully decomposable Pluronic P123 (poly(ethylene oxide)₂₀-poly(propylene oxide)₇₀-poly(ethylene oxide)₂₀), Pluronic F127 (poly(ethylene oxide)₁₀₆-poly(propylene oxide)₇₀-poly(ethylene oxide)₁₀₆, and partially decomposable polyvinyl chloride (PVC) or polyvinyl alcohol (PVA) are the commonly used soft templates.

As shown in Fig. 9a, Saha et al. first reported the synthesis of lignin-derived mesoporous carbon using the Pluronic F127 template. The obtained mesoporous carbon has mesopores in the range of 2.5–12.0 nm, a SSA of 418 m² g⁻¹ and a mesopore volume of 0.34 cm³ g⁻¹ which is twice the micropore volume (Saha et al. 2013). Qin et al. investigated the effect of lignin molecular weight on the structure and morphology of mesoporous carbon. The mesoporous carbon prepared from low-molecular-weight lignin had highly ordered mesoporous channels (3.8 nm) and high SSA (466 m² g⁻¹) (Qin et al. 2018). Herou et al. prepared lignin-derived mesoporous carbon using phloroglucinol, glyoxal, F127, and organosolv lignin. The PPO units of F127 could assemble with lignin to form micelles, and the PEO units could interact with the solvent, while phloroglucinol could surround the PEO segments through hydrogen bonds and glyoxal as a tetra-functional cross-linking precursor. The as-prepared lignin-derived mesoporous carbon with a SSA of 673 m² g⁻¹ and a pore volume of 0.5 cm³ g⁻¹ exhibited a 2D-hexagonal pore structure (Herou et al. 2019). Similar approaches have been employed by Gan and Qi to achieve mesoporous carbon materials (Wang et al. 2020b, 2021e).

Fig. 9

a Schematic illustration of LDPC synthesis using a soft template. Reproduced with permission from Ref. (Saha et al. 2013). Copyright © 2013, American Chemistry Society; (b) Schematic illustrating of LDPC synthesis using soft and hard template strategy. Reproduced with permission from Ref. (Zhang et al. 2016a). Copyright © 2021, Elsevier; (c) Novel template strategy for LDPC synthesis. Reproduced with permission from Ref. (Xi et al. 2020). Copyright © 2020, Roal Society of Chemistry

The pore-forming mechanism of the hard template method is to replicate the structure characteristics of the hard template. Therefore, the characteristics of hard templates play a key role in the formation of the porous structure. Silica-based templates and metal oxides are the common hard templates. Valero-Romero et al. used Alcell lignin as carbon precursors and different zeolite templates as hard templates to prepare lignin-derived mesoporous carbon using the liquid phase impregnation technique. Carbonization at 700 °C and a mass ratio of 1:1 were determined to be the optimal preparation conditions. The lignin-derived mesoporous caron prepared with zeolite Y, zeolite β, ZSM-5 and mordenite templates displayed high SSAs of 952, 864, 619 and 682 m² g⁻¹, and total pore volume of 0.35, 0.32, 0.15 and 0.28 cm³ g⁻¹, respectively (Valero-Romero et al. 2014). As shown in Fig. 9b, Huang et al. used SiO₂ and cetyltrimethylammonium bromide (CTAB) dual templates to prepare lignin-derived honeycomb-like mesoporous carbon with a SSA of 1107 m² g⁻¹, a mesopore volume of 2.06 cm³ g⁻¹ and a total pore volume of 2.35 cm³ g⁻¹ through a self-assembly strategy (Huang et al. 2021b). Li et al. found that silica nanoparticles with a size of 7 nm could lead to synthesis of a highly porous structure compared with that with sizes of 100 and 200 nm prepared using P123 and silica nanoparticles dual templates. With increasing the silica nanoparticles loading from 10 wt.% to 60 wt.%, the SSA of lignin-derived mesoporous carbon increased from 407 m² g⁻¹ to 636 m² g⁻¹ (Li et al. 2020c).

Although silica hard templates could be used to prepare lignin-derived mesoporous carbon with a uniform structure, the silica templates are generally removed by hydrofluoric acid, which is not environmentally friendly. Metal oxide templates have been widely used in recent years because of the green advantage. Song et al. used spherical nano MgO template to prepare alkali lignin-derived mesoporous carbon with a high mesoporous ratio of 68.93% at 1000 °C. The SSA (85 m² g⁻¹) was four times that of the LDC without MgO template (19 m² g⁻¹) (Song et al. 2017). Zhang et al. designed a dedicated self-assembly strategy to synthesize lignin-derived hollow carbon from EHL using a spherical MgO template. EHL coated the MgO template via hydrogen bond self-assembly. After carbonization and template removal, the EHL-derived hollow carbon was obtained (Zhang et al. 2020a). Zhang et al. further used a lamellar MgO template to prepare EHL-derived flower-like carbon by the same hydrogen bond self-assembly strategy (Zhang et al. 2020b). Combining hard templates with chemical activation is also an efficient way to engineer the porous structure of LDPC. Yin et al. prepared N/S-doped carbon from lignin amine using Fe₃O₄ as a template and KOH activation. Fe₃O₄ contributed to the formation of macropores, favoring the formation of micropores and mesoporous on the wall by the subsequent KOH activation, and the as-prepared porous carbon had a SSA of 1199 m² g⁻¹ (Yin et al. 2020c).

The development of new templates to regulate the pore structure of LDPCs has also attracted extensive research attention. As shown in Fig. 9c, Xi et al. used ZnCO₃ with dual roles of gas exfoliation and an in-situ ZnO template to prepare hierarchical porous carbon from EHL. The as-prepared hierarchical porous carbon exhibited 3D porous structure and an SSA of 531 m² g⁻¹. The SSA and pore volume increased with increasing the carbonization temperature to 600 °C (Xi et al. 2020). Fu et al. prepared lignosulfonate/ZnC₂O₄ composite using a hydrophobic bond self-assembly strategy. Owing to gas exfoliation and in-situ ZnO template roles of ZnC₂O₄ during the carbonization process, porous LDPC quasi-nanosheets with a SSA of 1069 m² g⁻¹ and a total pore volume of 1.375 cm³ g⁻¹ were obtained (Fu et al. 2020). Furthermore, Fu et al. investigated the effects of ZnCO₃ and ZnC₂O₄ on the structure of LDPC, and concluded that ZnC₂O₄ had a synergistic effect with lignin, playing a key role in the preparation of hierarchical porous carbon with crumpled nanosheets (Fu et al. 2021).

4.4 Direct carbonization

The direct carbonization method is also called self-template method without additional activation agents, additives, or templates (Zhang et al. 2017; Yin et al. 2020b). A direct carbonization strategy could effectively eliminate the usage of templates and activation agents. Therefore, direct carbonization is a facile, green-environmental, and low-cost route to engineering the pore structure of LDPCs. Direct carbonization of sodium lignosulfonate resulted in LDPCs with moderate SSAs owing to the template role of the pyrolysis product of Na₂SO₄ (Zhang et al. 2022). The pore-forming mechanism of the direct carbonization method is to use the rich inorganic minerals or metal ions in lignin to inhibit the decomposition and shrinkage of lignin or to chemically etch the carbon structure of the lignin at high temperatures. After the removal of the impurities, the developed porous structure was formed in the lignin-derived porous structure.

Liu et al. directly pyrolyzed Kraft lignin at 600–900 °C to prepare O-N-S co-doped hierarchical porous carbon with a large SSA of 338–1307 m² g⁻¹, hierarchical porous structure and a heteroatoms co-doping level of 9.84–19.91 wt.% (Liu et al. 2019a). The Kraft lignin used contained inorganic impurities that could act as templates. Kijima et al. prepared alkaline lignin-derived microporous carbon with a SSA of 740 m² g⁻¹ by carbonization at 900 °C. Furthermore, the SSA of alkaline lignin-derived porous carbon increased to above 1000 m² g⁻¹ via micelle formation and polymer gelation techniques (Kijima et al. 2011). Chen et al. used calcium lignosulfonate as a carbon source to prepare porous carbon by simple carbonization at 600–800 °C. The SSA and pore volume increased from 628 to 1362 m² g⁻¹ and 0.37 to 0.83 cm³ g⁻¹ with increasing the temperature from 600 to 800 °C (Chen and Zhou 2012). Pang et al. prepared interconnected hierarchical porous carbon from sodium lignosulfonate by direct carbonization at temperatures in the range of 600–1000 °C. The SSA and pore volume increased from 248 to 1010 m² g⁻¹ and 0.22 to 0.56 cm³ g⁻¹, respectively, as the temperature increased from 600 °C to 800 °C. With increasing temperature from 800 °C to 1000 °C, the SSA increased to 1328 m² g⁻¹ and pore volume further increased 1.06 cm³ g⁻¹, respectively (Pang et al. 2017). Li et al. directly carbonized the mixture of sodium lignosulfonate and melamine to prepare N–O-S doped porous carbon at 600–800 °C. With increasing temperature, the SSA of the LDPC increased from 164 to 808 m² g⁻¹ and the micropore volume increased from 0.1 to 0.37 cm³ g⁻¹, respectively, while the nitrogen content decreased from 16.79 to 9.72 at.% (Li et al. 2021d). It needs to be noted here that the possible pore generation mechanism of lignosulfonate is that the inorganic metal species generate templates (such as Na₂SO₄). The molecular weight also influenced the pore structure of LDPCs. Jeon et al. investigated the structural properties of porous carbon derived from three types of lignin with different molecular weights. Generally, the SSA of porous carbon prepared from low-molecular-weight lignin was higher than that prepared from high-molecular-weight lignin (Jeon et al. 2015).

The developed porous structure of LDC by direct carbonization can be obtained in combination with other pretreatment techniques. Zhang et al. obtained LDPC with a large surface area of 1831 m² g⁻¹ and abundant micropores and mesopores by a carbonization process assisted with a bacterial culture process (Zhang et al. 2019b). Liu et al. prepared LDC nanosheets by liquid nitrogen freeze-casting and direct carbonization at 900 °C (Liu et al. 2017). Demir et al. prepared sulfur self-doped porous carbon from calcium lignosulfonate by hydrothermal carbonization at 300 °C and thermal annealing at 700–1000 °C. The SSA increased from 78 to 660 m² g⁻¹, and the pore volume increased from 0.05 to 0.25 cm³ g⁻¹ at 700–800 °C, and then decreased from 660 to 260 m² g⁻¹, 0.25 to 0.12 cm³ g⁻¹ at 800–1000 °C (Demir et al. 2018). Liu et al. prepared nitrogen and phosphorus dual-doped LDC microspheres with a high SSA of 938.1 m² g⁻¹ and pore volume of 0.64 cm³ g⁻¹ using pre-oxidation and carbonization of ionic liquid ([Mmim]DMP)-lignin solution (Liu et al. 2022). Pang et al. prepared sodium lignosulfonate-derived hierarchical porous carbon spheres with a high SSA of 1255–1939 m² g⁻¹ and nitrogen content of 1.14–1.66 at.% by air thermal stabilization at 200 °C and carbonization at 700–900 °C (Pang et al. 2018a).

5 Energy storage applications of LDCs

5.1 Supercapacitors

Supercapacitors are promising candidates for energy storage devices due to their coupling merits of high power density, fast charging rate, and long cycle life (Shao et al. 2018). At present, the market share of the supercapacitor is relatively small compared with rechargeable batteries due to its low energy density and high energy storage cost (Yin et al. 2021). Supercapacitors store charge at the electrode/electrolyte interface in two ways: through fast surface ion adsorption/desorption for electrochemical double-layer capacitors (EDLCs) (Chen et al. 2017b); through fast and reversible surface or near-surface Faradic reactions for pseudocapacitors (PCs) (Yu et al. 2013). Porous carbons are commonly used as the electrode materials of EDLCs. A desirable EDLC carbon electrode should have a high SSA to build an electric double layer and a high conductivity to conduct electrons. Activated carbon, carbon nanotube, and graphene have been extensively explored for carbon electrodes, but their complex preparation procedures and high production cost hinder their commercialized applications (Peng et al. 2018). Lignin has been regarded as a promising precursor for advanced carbon materials (Espinoza-Acosta et al. 2018; Zhu et al. 2020). Over the years, diverse LDCs with high SSA have been prepared by the chemical activation method (KOH, K₂CO₃, K₂FeO₄, ZnCl₂, ZnC₂O₄, etc.), template method (F127, SiO₂, MgO, Fe₃O₄, ZnO, P123, F127, etc.), or self-pyrolysis, which show good application performances when applied as the EDLCs electrodes as summarized in Table 1.

Table 1 Representative LDPCs with their electrochemical performances

It is generally accepted that high SSA would lead to high specific capacitance (Wang et al. 2016c). As illustrated in Fig. 10, when the SSA is below 2000 m²·g⁻¹, the specific capacitance of LDPC is positively correlated with the SSA. However, the specific capacitance is limited and does not continue to increase when the SSA exceeds 2000 m²·g⁻¹ or even reaches 3000 m²·g⁻¹. Therefore, the SSA is not the only factor that determines the specific capacitance of LDPCs. In fact, the narrow micropores (< 0.7 nm) in porous carbon are too small in size to accumulate cations, hence making a minor contribution to the capacitance (Shao et al. 2020). Therefore, pore size distribution could be another key factor affecting the capacitive performances of porous carbons.

Fig. 10

A summary of the relationship between specific capacitance and SSA for the reported LDPCs. Data displayed in this figure are obtained from Table 1

Tremendous strategies have been proposed to improve the pore accessibility of LDPCs, including morphology structuring (Liu et al. 2017; Wang et al. 2022c), pore tailoring (Guo et al. 2017; Herou et al. 2019), and heteroatom doping (Tian et al. 2019; Liu et al. 2022). Hierarchical porous carbon materials with interconnected macro-meso-micropore structures are regarded to be desirable electrode materials. In the capacitive charge-storage process, macropore serves as a buffer site for ions, mesopore can shorten the transport distance of ions in internal pores, and micropore can provide active sites for ion accumulation, which boosts the transfer kinetics of electrolyte ions to achieve high capacitance (Dutta et al. 2014). Guo et al. prepared lignin-derived hydrochar with abundant interconnected macropores by hydrothermal carbonization, and then 3D hierarchical porous carbon with honeycomb structure could be obtained through KOH activation (Fig. 11a-d) (Guo et al. 2017). The interconnect macropores, plentiful mesoporous, and rich microporous afford its excellent capacitance performance. The 3D hierarchical porous carbon showed high specific capacitance of 420 F·g⁻¹ at 0.1 A·g⁻¹ and 284 F·g⁻¹ at 100 A·g⁻¹ in 6 M KOH (Fig. 11e). Moreover, the assembled symmetric supercapacitor presented a fast frequency response in both ionic liquid and aqueous electrolytes. It delivered an outstanding energy density of 46.8 Wh·kg⁻¹ and a superior power density of 25,400 W·kg⁻¹ in EMIM TFSI electrolyte (Fig. 11f). Lignin-derived 3D hierarchical porous carbons were obtained through hydrothermal carbonization and activation by Xu’s group (Zhang et al. 2016a) and An’s group (Li et al. 2021a), where the carbon electrodes showed gravimetric specific capacitances of 312 F·g⁻¹ at 1 A·g⁻¹ and 324 F·g⁻¹ at 0.5 A·g⁻¹ in 6 M KOH, respectively. Zhang et al. proposed a novel bacterial activation method to synthesize lignin-derived hierarchical porous carbon (Zhang et al. 2019b). It was found that the bacteria cleave the stable chemical bonds (β-O-4, β-β’, and β-5) of the lignin to reduce its molecular weight, resulting in the formation of interconnected hierarchical porous carbon structure during activation. The prepared porous carbon displayed a specific capacitance up to 428 F·g⁻¹ at 1.0 A·g⁻¹ and 355 F·g⁻¹ at 30 A·g⁻¹ and 82.9% capacitance retention in 6 M KOH. The carbon electrode also showed good cycling performance with a capacitance loss smaller than 3.3% after 10,000 cycles at 5 A·g⁻¹. The fabricated symmetric supercapacitor using this porous carbon electrode exhibited an ultrahigh energy density of 66.18 Wh·kg⁻¹ at 312 W·kg⁻¹ in EMIM TFSI electrolyte.

Fig. 11

a Schematic illustration of preparing lignin-derived 3D hierarchical porous carbons. (b-d) SEM images of the 3D hierarchical porous carbons. e Specific capacitance at various current densities of the 3D hierarchical porous carbons in a three-electrode system in 6 M KOH. f Ragone plots for the 3D hierarchical porous carbon-based supercapacitors in 6 M KOH and neat EMIM TFSI. Reproduced with permission from Ref. (Guo et al. 2017). Copyright © 2017, Roal Society of Chemistry

2D carbon nanosheets can also be candidate electrode materials for supercapacitors owing to their exposed active sites, and good electronic conductivities (Peng et al. 2018). Lignin-derived 2D carbon nanosheets were prepared by freeze-casting of an aqueous lignin dispersion followed by direct carbonization (Liu et al. 2017). With the help of an ice crystal template, carbon nanosheets with a thickness of 50–150 nm were obtained, which displayed a specific capacitance of 281 F·g⁻¹ at 0.5 A·g⁻¹ in 1 M H₂SO₄. Unfortunately, 2D nanosheets were prone to self-stacking, reducing the effective utilization of the surface (Gao et al. 2020). To solve this issue, Fu et al. adopted self-assembly to transform sodium lignosulfonate into 3D porous carbon composed of quasi-nanosheets with the gas-exfoliation and in-situ templating of zinc oxalate (Fu et al. 2020). Benefiting from the fast ion diffusion, boosted electron conduction, and good mechanical stability, the 3D porous carbon electrode exhibited a high specific capacitance of 365 F·g⁻¹ at 0.5 A·g⁻¹ in 6 M KOH, as well as good durability (a 93.5% capacitance retention after 10,000 cycles at 5 A·g⁻¹). The assembled solid symmetric supercapacitor using PVA/KOH gel electrolyte showed a high energy density of 9.75 Wh·kg⁻¹ under a high power density of 6157.9 W·kg⁻¹. Jiang et al. used reduced graphene oxide hydrogels composed of nanosheets as a template to convert lignin into 3D carbon aerogel with a lamellar interconnect structure (Jiang et al. 2020). The hierarchical porous structure and the resulted high SSA rendered the material a good capacitive performance. It showed a high specific capacitance of 330 F·g⁻¹ at 1.0 A·g⁻¹ in 1 M H₂SO₄ and the assembled supercapacitor showed superior cycling performance with a 100% capacitance retention over 10,000 cycles at 10 A·g⁻¹.

In addition to 3D porous carbon and carbon nanosheets, LDC spheres and carbon nanofibers also have ideal performances in supercapacitor applications. Pang et al. prepared carbon spheres from sodium lignosulfonate through thermal stabilization and carbonization (Pang et al. 2018a). The carbon spheres showed a specific capacitance of 276 F·g⁻¹ at 0.1 A·g⁻¹ and impressive cycling stability with 99.5% capacitance retention after 10,000 cycles in 7 M KOH. Chen et al. synthesized LDPC spheres by spray drying a solution of lignosulfonate and KOH, followed by carbonization, which achieved a specific capacitance of 340 F·g⁻¹ at 0.5 A·g⁻¹ in 3 M KOH (Chen et al. 2018). Ma et al. used lignin as a carbon precursor to prepare porous carbon nanofiber by electrospinning with polyvinylpyrrolidone as a spinning agent and Mg(NO₃)₂·6H₂O as an additive (Ma et al. 2018). The obtained carbon nanofiber had a hierarchical porous structure with an SSA of 1140 m²·g⁻¹, and showed a specific capacitance of 248 F·g⁻¹ at 0.2 A·g⁻¹ in 6 M KOH. Titirici’s group designed free-standing LDC nanofiber mats with high packing density (Hérou et al. 2021). Due to the decreased dead volume, the LDC nanofiber electrode achieved a volumetric capacitance of 130 F·cm⁻³ and an energy density of 6 Wh·L⁻¹ at 0.1 A·g⁻¹ with an active mass loading of 3 mg·cm⁻², which outperforms most commercial porous carbons (volumetric capacitance: 50 ~ 100 F·cm⁻³, energy density: 1 ~ 3 Wh·L⁻¹).

EDLCs store charge through ion adsorption/desorption on the electrode surface, and the surface heteroatom doping of carbon would affect the capacitive performance through specific adsorption. Impressive experimental and theoretical results have demonstrated that heteroatom doping could change the charge density distribution of carbon and induce surface polarization, thus accelerating ion adsorption (Liu et al. 2019b; Li et al. 2020b). Meanwhile, the introduction of heteroatoms can also enhance the surface wettability of carbon and increase the accessible surface area, although the wettability of porous materials can not be measured due to the capillarity effect of porous carbons. Moreover, the introduced heteroatomic functional groups can serve as redox-active sites to provide additional pseudocapacitance (Cui et al. 2020).

Pyrolyzing a mixture of lignin, additive with heteroatoms, and activation agent is the prevailing method for synthesizing heteroatom-doped LDPCs. For instance, Wang et al. prepared N-doped porous carbons via activating aniline- and urea-modified lignin with KOH, respectively, which achieved specific capacitances of 336 F·g⁻¹ at 1 A·g⁻¹ and 306 F·g⁻¹ at 0.1 A·g⁻¹ in 6 M KOH (Wang et al. 2016a). In the work done by Zhu et al., lignin-containing black liquor with KOH aqueous solution was used as a precursor and melamine was used as a nitrogen source to prepare N-doped porous carbon with high SSA (Zhu et al. 2017). The as-prepared porous carbon displayed a specific capacitance of 337 F·g⁻¹ at 0.5 A·g⁻¹ in 6 M KOH. Physically mixing lignin and additives is easy to cause heterogeneous doping, which limits the capacitive performance. However, lignin is rich in phenolic hydroxyl and carboxyl groups, which can be functionalized to introduce heteroatom-containing groups, thereby ensuring uniform heteroatom doping in the subsequent carbonization process. Zhang et al. fabricated 3D N-doped hierarchical porous carbons by hydrothermal crosslinking reaction and KOH activation using sodium lignosulfonate as a carbon precursor, 1,6-hexanediamine as a crosslinking agent and nitrogen source (Zhang et al. 2018d). The porous carbon possessed rich micropores, favorable mesopores and interconnected macropores with a high N content of 3.6 at.%, enabling its superior specific capacitance of 440 F·g⁻¹ at 0.5 A·g⁻¹ in 6 M KOH. Moreover, the as-fabricated supercapacitor delivered a high energy density of 18.5 Wh·kg⁻¹ at 300 W·kg⁻¹ in 6 M KOH aqueous system. The work by Liu et al. proposed a facile environmentally friendly and low-cost method to prepare porous carbon from heteroatom-doped lignin (Liu et al. 2019a). Kraft lignin was directly pyrolyzed without any additional activating agents, additives or templates to construct N, O, and S co-doped hierarchical porous carbon. The derived carbon had abundant multi-heteroatoms co-doping (up to 19.91 wt.%) and hierarchical porous structure (Fig. 12a-h), leading to its high specific capacitance of 300.5 F·g⁻¹ at 0.5 A·g⁻¹ in 6 M KOH (Fig. 12i) and a superior energy density of 66.8 Wh·kg⁻¹ at 1750 W·kg⁻¹ for the assembled supercapacitor in EMIMBF₄ ion liquid electrolyte (Fig. 12j).

Fig. 12

a-h SEM images and EDS elemental mappings of the O-N-S co-doped hierarchical porous carbons; i Specific capacitance at various current densities of the O-N-S co-doped hierarchical porous carbons; j Ragone plot of the assembled supercapacitor in EMIMBF₄ system. Reproduced with permission from Ref. (Liu et al. 2019a). Copyright © 2019, Elsevier

Nowadays, with the rapid development of portable electronics and wearable devices towards lightweight and miniaturization, supercapacitors are not only required to have high gravimetric energy density, but also high areal or volume energy density (Li et al. 2019). The traditional powder carbon materials have excellent gravimetric performances, but their low densities lead to poor volume performances, which makes it difficult to be used in compact devices. Micro supercapacitors (MSCs) have attracted wide attention as a new type of capacitive energy storage device because they are easy to carry and integrate with microelectronic systems (Lin et al. 2014). Alshareef et al. developed a lignin-based laser lithography technique to convert lignin film to 3D laser-scribed graphene electrode and fabricate on-chip MSC (Zhang et al. 2018b). The electrode was hierarchically porous and electrically conductive with a conductivity up to 66.2 S·cm⁻¹. The fabricated MSCs exhibited good capacitive performances with a high areal capacitance of 25.1 mF·cm⁻², a high volumetric energy density of 1 mWh·cm⁻³, and a high volumetric power density of 2 W·cm⁻³. Sun et al. produced S-doped porous 3D graphene materials from lignin and polyethersulfone (PES) films via laser direct writing technique (Sun et al. 2021). The as-fabricated MSCs using H₂SO₄/PVA gel as electrolyte showed excellent electrochemical performances with a high areal capacitance of 22 mF·cm⁻², and a high areal energy density of 1.53 mWh·cm⁻² at an areal power density of 25.4 mW·cm⁻². Yuan et al. synthesized superhydrophilic 3D porous graphene with O/S co-doping through a repeated laser scribing technique using sodium lignosulfonate slurry to coat the laser-induced graphene interdigital electrodes (Fig. 13a) (Yuan et al. 2021). The electrode possessed hierarchical porous graphene architecture, superior hydrophilicity, and O/S co-doped chemistry property, which greatly promotes the infiltration and transportation of electrolyte ions (Fig. 13b-d). Consequently, the assembled MSCs showed an outstanding areal capacitance of 53.2 mF·cm⁻², which is 39 times larger than that of the undoped MSC (Fig. 13e). At the same time, the MSCs using PVA/H₂SO₄ gel as electrolyte supplied a high areal energy density of 4.73 μWh·cm⁻² and a high areal power density of 1.6 mW·cm⁻² (Fig. 13f). The above findings demonstrate the good application potential of lignin in MSCs and open a new path to the application of LDC in EDLCs.

Fig. 13

a Schematic illustration of the fabrication of superhydrophilic O/S co-doped graphene electrodes; SEM images of the superhydrophilic O/S co-doped graphene electrode in (b) low-magnification, (c) high-resolution top-view, and (d) cross-sectional view; e Areal specific capacitance of the graphene electrodes; f Areal Ragone plot comparing the assembled MSC with reported graphene-based MSCs. Reproduced with permission from Ref. (Yuan et al. 2021). Copyright © 2021, Elsevier

5.2 Lithium-ion battery

Lithium-ion batteries (LIBs) have received considerable attention in consumer electronics owing to their high voltage output, good charge capability, robust stability, low self-discharge, and high power density (Lu et al. 2013). In general, a LIB is composed of a cathode, anode and organic electrolyte and stores energy physically through the intercalation reactions of Li⁺ at the positive and negative carbon electrode during charge and discharge (Scrosati et al. 2011). The electrochemical performance of LIBs mainly depends on the electrode material, while the theoretical capacity of commercial graphite anode is only 372 mAh g⁻¹ by LiC₆ stoichiometry, resulting in low energy density and power density of the current LIBs (Goodenough and Park 2013). Developing more lithium insertion sites in carbonaceous materials plays a positive role in improving the capacity of anode materials. Inner porosity and high crystallinity could contribute to the reversible lithiation sites (Liu et al. 2021b). LDCs can be efficiently converted into low-cost materials as anodes with highly reversible Li⁺ ion-storage capacity and fast lithium-storage kinetics, providing a new promising pathway to application in batteries fields (Tenhaeff et al. 2014; Espinoza-Acosta et al. 2018; Svinterikos et al. 2020).

The mechanisms of Li⁺ ions storage include pore filling, defect adsorption in the amorphous regions and the intercalation of Li⁺ ions in graphitic layers (Fig. 14a) (Zhang et al. 2021a). The high SSA and hierarchical porous distribution realized by the template and activation methods can achieve effective adsorption and reversible capacity of the lithium ions, which bestows LDPC a higher capacity than carbonized lignin anode. Typically, SiO₂ as a hard template can generate porous structures. As shown in Fig. 14b, Huang et al. pioneered electrostatic self-assembly and dual-template method using cationic quaternized alkali lignin (QAL) (carbon source), sodium dodecylbenzene sulfonate (soft template) and SiO₂ nanoparticles (hard template) to prepare LDPCs possessing ordered mesoporous structure and high pore volume (2.23 cm³ g⁻¹) (Huang et al. 2021b). The obtained carbon materials with high porosity possess fast lithium ion diffusion kinetics and more lithium-ion storage sites for LIBs, leading to LDPCs with a high reversible capacity of 1109 mA h g⁻¹. LDC could also be bonded with conversion-type anodes, e.g., lignosulphonate/MoS₂ composite (Chen et al. 2019a), Si/C from lignin-SiO₂ composite (Du et al. 2018; Li et al. 2021e), lignin/Fe₂O₃ composite (Yi et al. 2017), and lignin/NiO composite (Zhou et al. 2018) etc. All these anode materials displayed remarkable lithium storage performance. Furthermore, Zhang et al. and Xi et al. fabricated the LDPC with KOH, K₂CO₃, K₂C₂O₄ and K₃PO_4, respectively, as both template and activating agents, (Zhang et al. 2015b; Xi et al. 2018, 2019). K₂CO₃ was found to be superior to other potassium compounds in the lignin activation due to the release of CO₂ gas to loosen the lignin and act as an activation agent in the high-temperature activation process (Fig. 14c). The excellent multi-stage pore structure and graphitic lamellar structure of LDPC with K₂CO₃ are favorable for the (de)intercalation of lithium ions (470 mAh g⁻¹ at 0.2 A g⁻¹ after 400 cycles) (Fig. 14d-f). Xi and co-workers prepared LDPC using ZnCO₃ as an in situ templating agent (Xi et al. 2020). The decomposed ZnCO₃ produces gas to peel off lignin for creating micropores and mesopores and the generated ZnO could act as a hard template for generating mesopores. The resultant LDPC achieved excellent volume energy density (730 mAh·cm⁻³ after 200 cycles at 0.2 A·g⁻¹). Recently, electrospun lignin carbon fiber (LCF) also came into view, which might replace traditional carbon fiber materials (Wang et al. 2021c). Tenhaeff et al. eliminated the binder and additional additives in the LIBs anode using LCF mats produced from lignin and polyethylene (Tenhaeff et al. 2014). The LCF mats had comparable electrochemical performance to typical graphite anodes in LIBs. Similar to this work, research has been done on the preparation and application of carbon nanofibers (CNFs) from lignin and other polymers such as polyacrylonitrile (PAN) (Choi et al. 2013; Shi et al. 2017; Ma et al. 2019), polyethylene oxide (PEO) (Wang et al. 2013) and PVA (Stojanovska et al. 2019) as the anodes in LIBs. Recently, Culebras et al. reported CNFs derived from electrospun nanofibers of lignin/polylactic acid (PLA) after carbonization (Fig. 14g) (Culebras et al. 2019). The electrodes produced from 50% PLA blends exhibited a high specific capacity of 611 mAh g⁻¹ after 200 cycles at 186 mAh g⁻¹, owing to the increased porosity. In addition, the surface defect site and the degree of graphitization of LDC could be enhanced through H₂ reduction (Chang et al. 2015), in which the carbonyl content and interlayer spacing of LDC were improved via pre-oxidation process (Du et al. 2021). Thus, LDC as the anode in LIBs can obtain excellent rate ability and cycling stability through ingenious microstructure design.

Fig. 14

a Available Li.⁺ ions storage mechanism in hard carbons. Reproduced with permission from Ref. (Zhang et al. 2021a). Copyright © 2021, Wiley; b Schematic illustration of the fabrication process of the LHC/SiO₂. Reproduced with permission from Ref. (Huang et al. 2021b). Copyright © 2021, Elsevier; c The activation mechanism of K₂CO₃ to EHL; d-f SEM image, TEM and HRTEM images of LDC-K₂CO₃-900. Reproduced with permission from Ref. (Xi et al. 2018). Copyright © 2018, Elsevier; g Schematics of phase separation for lignin/PLA fibers (i), and TEM images of lignin/TPU-50:50 (ii) and lignin/PLA-50:50 (iii). Reproduced with permission from Ref. (Culebras et al. 2019). Copyright © 2019, Wiley; h Schematic illustration for the formation of C-S-6 and C-S-10 composites. Reproduced with permission from Ref. (Yu et al. 2017). Copyright © 2019, Wiley

In addition, LDC also has found a satisfactory application in lithium-sulfur (Li–S) batteries. The poor conductivity of sulfur and the shuttling of polysulfides hinder their commercial applications, even if Li–S battery has a high theoretical energy (2600 Wh kg⁻¹). Micropores could physically hinder the diffusion of S in Li–S batteries. To address this, Yu et al. synthesized lignin-derived macro-/micro-porous carbon by carbonization/activation method and used it as a sulfur-loading matrix (Fig. 14h) (Yu et al. 2017). A porous carbon with a high SSA (1211.6 m² g⁻¹) and pore volume (0.59 cm³ g⁻¹) was obtained. The research found that the sulfur loading time could affect the structure and electrochemical properties of the carbon–sulfur composites. Benefiting from the rich micropores at a long sulfur-loading time, the Li–S batteries exhibited a highly reversible capacity of 791.6 mAh g⁻¹ in the 100th cycle. Subsequently, Yeon et al. grafted nitrogen heteroatoms onto LDPC via facile KOH activation, which formed N-doped honeycomb-like porous carbons (Yeon et al. 2020). The N heteroatoms could strongly interact with S by accelerating the charge-transfer kinetics, and the resulted hard carbons were used as anodes for Li–S batteries, achieving a high initial discharge capacity of up to 1295.9 mAh g⁻¹ and a stable reversible capacity of 647.2 mAh g⁻¹ in the subsequent 600 cycles.

5.3 Sodium-ion battery

Sodium-ion batteries (SIBs or NIBs) have recently been considered as a potential candidate for commercial applications in large-scale renewable energy storage (Zhang et al. 2019e). It is noted that Na⁺ ions are incompetent to insert into graphite layers due to larger ionic radius (0.102 nm) and instability binary compounds of NaC_x, resulting in its low capacity of ~ 30 mAh g⁻¹ in graphite electrode (Liu et al. 2016c; Moriwake et al. 2017). However, LDC as a hard carbon with disordered graphene nanodomain, large interlayer spacing and high electronic conductivity is expected to have a superior Na uptake performance and low-voltage plateau profile in NIBs.

The precise charge storage mechanism of Na⁺ has been under debate for a long time. The storage mechanisms of Na⁺ ions in hard carbon were investigated in details and summarized as follows (Qiu et al. 2017; Bai et al. 2018): “adsorption-intercalation” mechanism (Fig. 15a), “intercalation-adsorption” mechanism (Fig. 15b) and “adsorption-pore filling” mechanism (Fig. 15c). Recently, the “adsorption-pore filling” mechanism has gained more popularity as it can explain the sloping profile, mainly attributed to Na⁺ adsorption at defect sites or the sites with large interlayer spacing and intercalation between layers (Zhang et al. 2021c), and for the plateau region, related to nanovoid filling. Theoretically, in order to fully exploit the sodium storage potential of hard carbon, it would be an effective method to modify its morphology, e.g., defect concentration, porosity and interlayer distance. It is critical to developing an effective design strategy for the preparation of desirable nanoscale structures according to the charge storage mechanism of Na⁺ ions in LDC. For instance, CNFs fabricated from pyrolysis of polymer nano-fibers showed a greatly improved sodium storage capacity, owing to a large number of defect sites to store sodium in addition to improved electronic and ionic conductivity (Luo et al. 2013). Zhang et al. constructed a nanocarbon network as an anode material (E-KL/CA-C) for SIBs by pyrolyzing crosslinked cellulose acetate and lignin (Fig. 15d) (Jia et al. 2018), realizing a reversible capacity of 340 mAh g⁻¹ after 200 cycles and a capacity of 103 mAh g⁻¹ at 400 mA g⁻¹. In addition, heteroatom-doping strategies can produce defects and increase Na⁺ ion-storage sites in LDC and construct pore structure to provide a short sodium ion diffusion path. Chen et al. employed nitrogen-doped porous ultrathin layered carbon as the anode material, prepared by annealing alkaline lignin with melamine and urea (Fig. 15e) (Chen et al. 2020b). From the discharge curves, Na⁺ ions likely underwent two stages: the insertion of Na⁺ ions into the defects and edges of graphite microcrystal at high potential, and into the graphitic layers-like at low potential. The N-doping mainly enhanced the capacity in the discharge voltage at high potential, delivering a high capacity of 138.7 mAh g⁻¹ at 5 C and no obvious capacity decay after 4000 cycles.

Fig. 15

Schematic illustration of the mechanisms for Na-ion storage in hard carbon: a “intercalation–adsorption” mechanism; b “adsorption–intercalation” mechanism. Reproduced with permission from Ref. (Qiu et al. 2017). Copyright © 2017, Wiley; c Schematic diagram of the sodium-storage mechanism in HCs. Reproduced with permission from Ref. (Bai et al. 2018). Copyright © 2017, Wiley; d Schematic illustration of the fabrication procedure of E-KL/CA-C. Reproduced with permission from Ref. (Jia et al. 2018). Copyright © 2018, American Chemistry Society; e Schematic illustration of the porous nitrogen-doped ultrathin layered carbon (LC) preparation. Reproduced with permission from Ref. (Chen et al. 2020b). Copyright © 2020, Elsevier; f Schematic illustration of the lignin-based interpenetrating polymer networks. g Relationship between carbonization temperature and the lignin/epoxy resin mass ratios to layer spacing and specific capacity. Reproduced with permission from Ref (Zhang et al. 2018a). Copyright © 2018, Elsevier; h Schematic illustration showing the preparation of HCM from a single sodium lignin sulfonate source and the process for Na storage. Reproduced with permission from Ref. (Li et al. 2020a). Copyright © 2020, Roal Society of Chemistry

But the low initial Coulombic efficiency (ICE) of the LDC reported in the above-mentioned study (26–52%) is one of the obstacles to their further commercialization. To address this issue, some research work demonstrated that the increasing SSA of LDC could improve ICE: a high SSA may increase Na⁺ consumption and form an unstable solid electrolyte interphase (SEI) layer, resulting in irreversible Na⁺ insertion into the amorphous structure (Huang et al. 2014; Zhang et al. 2016b). Pure kraft lignin-derived LDPC showed poor electrochemical performances in SIBs due to its dominant microstructure (Jin et al. 2014). Similarly, electrospun lignin/PAV-derived nanofibers with high SSA (1049 m²·g⁻¹) also exhibited relatively poor ICE (40%) (Stojanovska et al. 2019). In contrast, lignin/PAN-based carbon nanofibers with low SSA (approx. 27 m²·g⁻¹) showed an initial Coulombic efficiency of 70.5% (Jin et al. 2014). Furthermore, Zhang and co-workers found that adjusting layer spacing can be a crucial parameter for ameliorating ICE, which could provide interconnected channels to promote the rapid transmission of ions (Fig. 15f-g) (Zhang et al. 2018a). Their work successfully synthesized a cost-effective carbon material as anodes derived from the mixture of lignin and epoxy resin via a facile pyrolyzing procedure. In their process, the structure of carbon materials was precisely controlled by altering heat-treatment temperatures and the lignin/epoxy resin mass ratios. Thanks to the enlarged interlayer spacing (0.395 nm) and lower SSA (6.5 m² g⁻¹), the hard carbon microsphere (HCM) possesses distinguished sodium storage behavior with a high reversible capacity of 316 mAh g⁻¹, a superior ICE of 82%, and good capacity retention of 90% after 150 cycles. The full cell with HCM versus O₃-Na_0.9[Cu_0.22Fe_0.30Mn_0.48]O₂ not only demonstrated a high ICE of 80%, but also presented an energy density of 247 Wh kg_anode⁻¹ (vs. hard carbon anode). These results verified that the design strategy is feasible for the fabrication of carbon anode materials for SIBs. In addition, Li et al. obtained a practical HCM anode material using sodium lignosulfonate via facile spray drying and carbonization process (Li et al. 2020a). As displayed in Fig. 15h, the HCM has a plum-like structure with a particle size of 1 ~ 2 μm, formed from the self-assembly of the lignin functional groups and higher evaporation of spherical droplets during spray drying. The excellent sodium storage was reflected by its high reversible capacity (339 mAh g⁻¹) and outstanding ICE (88.3%), which could arise mainly from the enlarged interlayer spacing (0.396 nm), low SSA (11.89 m² g⁻¹) and surface defects limited SEI formation. HCM as an anode material is highly promising for sodium-ion full batteries with Na₃V₂(PO₄)₃ as the cathode. The Na₃V₂(PO₄)₃/HCM full cell can deliver a high reversible capacity (307 mAh g⁻¹ at 0.1 C) and stable cycle performance (221 mAh g⁻¹ up to 200 cycles at 1 C). The energy density of the full cell is up to 190 W h kg⁻¹ (based on the total mass of the anode and cathode).

Importantly, the primary source and molecule structure of lignin are confirmed to have a significant impact on the structural and electrochemical properties of its derived hard carbon products. For instance, Ghimbeu found both kraft lignin- and lignosulfonate-derived carbons have different capacity stability (200 mAh g⁻¹ vs. 120 mAh g⁻¹), attributing to different porosity (1.8 m² g⁻¹ vs. 180 m² g⁻¹) and morphology (dense particles vs. spherical) (Matei Ghimbeu et al. 2019). Moreover, lignin with carboxyl group introduced by pre-oxidation can enhance its crosslinking, which in turn increases the layer distance for facilitating the Na⁺ ion insertion and leads to exceptional rate capability and superb cycle life (Lin et al. 2020). The temperature regulation could precisely control the microstructure of LDC (Navarro-Suárez et al. 2018). Peuvot et al. used lignin-based carbon fiber (LCF) as free-standing anodes for SIBs, and reported that precise temperature control (800 ~ 1700 ℃) could affect SSA (4.7 ~ 355 m² g⁻¹), interlayer spacing (0.367 ~ 0.401 nm), and graphene nanodomain, resulting in a high specific capacity of 310 mAh g⁻¹ and ICE of 89% in 1200 ℃ (Peuvot et al. 2019). Similarly, Alvin et al. combined high-temperature carbonization and low-level heteroatom doping to increase the low-voltage plateau capacity (223 mAh g⁻¹), reversible capacity (328 mAh g⁻¹) and maintain high ICE (72%). The relationship between key structural parameters and sodium storage mechanism and performance, however, needs to be investigated.

5.4 Potassium-ion battery

Recently, potassium-ion batteries (PIBs) have come under the spotlight due to their low cost, rich resources and similar working principles as those of lithium-ion batteries (LIBs) (Xu et al. 2018b). The (de)intercalation mechanism of K⁺ in graphite has been confirmed and showed a low voltage plateau. But the large size of K⁺ could cause the large volume expansion of graphite (KC₈, 60%) (Komaba et al. 2015; Zhang et al. 2021h), resulting in slow kinetics and rate capacity. At present, hard carbons with large interlayer spacing have been widely studied in PIBs (Wu et al. 2019). Absolutely, the charge storage mechanism of K in hard carbon is quite different compared with graphite. In addition, besides the Faradaic contribution caused by K⁺ intercalation, the charge storage contribution of K⁺ adsorption onto the nanovoids and surface defects/functional groups could also play an important role (Fig. 16a) (Yang et al. 2018). Based on the above mechanism, to lower the associated strain energy in hard carbon anode, efforts have been made by controlling structural engineering, including expanding interlayer space, introducing defects, and heteroatom-doping, with which significant K-storage capacity improvements have been achieved (Zhang et al. 2021a). As reported previously, the directly carbonized lignin as a hard carbon could be used as an anode material to store a high amount of K⁺ ion with impressive low voltage platform capacity (at ~ 0.25 V), and the total capacity of the K⁺ ion insertion/extraction attained 246 mAh g⁻¹ (Alvin et al. 2020). The research also found that tuning the carbonization temperature (1000–1500 ℃) could alter the corresponding structure and electrochemical properties of LDC. As the carbonization temperature increased, carbon was gradually arranged from short-range ordered structure to long-range ordered structure with turbostratic graphene nanodomains, leading to the increased graphitization degree and considerably collapsed micropore. Thereby the plateau capacity of KIB increased from 113 to 178 mAh g⁻¹. The adsorption-insertion mechanism of K⁺ ion insertion into hard carbon was further confirmed by experimental observation and theoretical calculation. The sloping voltage capacity was attributed to the adsorption of K⁺ ions on defect sites, edge sites, and micropores (Zhang et al. 2020d, c, 2021d, f), whereas the low voltage plateau capacity was caused by the intercalation of the K⁺ ions into graphitic layers. Heteroatom doping also showed an effective effect on the potassium storage performance of carbon-based materials. Phosphorus-doped hard carbon (PHC) has enlarged interlayer spacing of the graphitic layer. Compared with HC, the capacity of PHC in KIB also increased from 245 to 302 mAh g⁻¹ mostly owing to the increase in the plateau capacity from 153 to 192 mAh g⁻¹. In addition, pre-oxidation could cross-link the structure of lignin precursor and further promote the reaction dynamic, and as a result, the carbonized lignin has a capacity of ~ 300 mAh·g⁻¹ in PIBs (Lin et al. 2020). Even though such potassium storage capacity is comparable to that of other hard carbon materials, it is still lower than the capacity of lithium storage in graphite. Jiang et al. employed a simple hydrothermal condition with graphene oxides (GO) as catalysts to reconstruct lignin (Jiang et al. 2021). Lignin, which loses the aromatic unit side chain, can be polymerized with reduced graphene oxide (rGO) foam to form a three-dimensional layered composite. Carbonized reconstructed lignin has fewer defects, fewer micropores and adjustable interlayer spacing at 750 ℃, which significantly improved the total capacity compared with directly carbonized lignin (~ 355 mA h·g⁻¹ vs 252 mA h·g⁻¹ at 50 mA g⁻¹) and the capacity (~ 200 mA h·g⁻¹) at a high current of 1 A g⁻¹ for K ion storage. Note that reconstructed lignin with the graphitic nanostructures leads to lowered voltage hysteresis between the charge and discharge process, which can be beneficial to practical application. In addition, there is still some possibility of remodeling lignin in response to potassium storage strategy, which can draw lessons from the previous relevant research work, e.g., designing high edge-nitrogen doped carbon (Zhang et al. 2020d, 2021c, e), building hierarchical porous carbon (Huang et al. 2021a), and fabricating hard/soft carbon composite (Jian et al. 2017) etc.

Fig. 16

a Schematic of the mixed mechanisms of potassium-ion storage. Reproduced with permission from Ref. (Yang et al. 2018). Copyright © 2018, Wiley; b Schematic of pyrolytic production of lignin-derived hard carbon; c The overall capacity, bulk-insertion (Qb) and surface-adsorption (Qs) contribution to the capacity of different lignin M_W and pyrolysis temperature; d Proposed K-ion storage mechanisms in low M_W (i), medium M_W (ii), and high M_W (iii) of lignin. Reproduced with permission from Ref. (Wu et al. 2022). Copyright © 2022, Elsevier

To better reveal the applications of LDC in potassium storage, it is important to understand the relationship between its microstructure and the charge storage mechanism. Wu et al. speculated the interplay among lignin molecular weight (MW), pyrolysis temperature, and hard carbon structure is essential in K-ion storage performance (Wu et al. 2022). In this research, lignins with M_W of 2233 (low M_W), 9660 (medium M_W), and 14,300 g mol⁻¹ (high M_W) were directly pyrolyzed at 500 ~ 1000 ℃ (Fig. 16b). As shown in Fig. 16c, the lignin of a higher molecular weight is more inclined to the surface adsorption contribution (Q_b), as the high M_W lignin is difficult to get modification (depolymerized, fragmented, and graphitized). In contrast, the lignin of a lower molecular weight is more inclined to the bulk-insertion contribution (Qs). To be more exact, as summarized in Fig. 16d, low M_W lignin would produce more graphite-like nanocrystals, narrow interplanar distance and fewer defect sites under the modification, causing limited Qb (115 mAh g⁻¹) and limited Qs (136 mAh g⁻¹) (Fig. 16d(i)). High M_W lignin achieves massive Qs (216 mAh g⁻¹) and limited Qb (111 mAh g⁻¹) due to its amorphous dominated carbon structures, least graphite-like nanocrystals with suitable carbon interlayer distance (Fig. 16d(iii)); Medium M_W lignin would realize both high Qs (219 mA h g⁻¹) and high Qb (159 mAh g⁻¹) from their balanced disordered carbon and graphite-like nanocrystals with expanded carbon layers (Fig. 16d(ii)). Moreover, the disordered carbon rose with the pyrolysis temperature increases (500 ~ 700 ℃), resulting to the dominant mechanism of hard carbon shifted from insertion mechanism to adsorption mechanism (Fig. 16c). However, adsorption contribution was slightly attenuated at a higher pyrolysis temperature (1000 ℃) due to the reduced interlayer spacing affecting the K⁺ ion bulk insertion, even with more disordered carbon. Thus, benefited from balanced disordered carbon and graphite-like nanocrystals with a considerable interlayer distance, the medium-M_W lignin at 700 ℃ achieved the best overall performance with a high reversible specific capacity (~ 300 mAh g⁻¹ at 50 mA g⁻¹) and high capacity retention (80% after 100 cycles). So fine control of hard carbon structure is a key to meeting the technical requirements of the anode materials in PIBs.

6 Catalysis and environmental applications of LDCs

6.1 Electrochemical catalysis

Highly efficient catalyst for oxygen reduction reaction (ORR) has become the decisive factor for the employment of fuel cells and metal-air batteries. Currently, the scarcity of platinum (Pt)-based catalysts still can not meet the commercial requirements. In fact, Pt-based catalysts have some limitations. Metal-free or non-precious metal catalysts attract attention for the motivation of developing abundant cheap catalysts. This research topic focuses on the following issues: (i) efficient alternative carbon sources; (ii) 3D structures exposing active sites; (iii) doping heteroatom to change electron configuration, enabling the catalyst with a four-electron (4e) pathway.

N, S co-doped porous carbon enhances the ORR performance due to the synergetic effect of nitrogen and sulfur doping (Fan et al. 2019). Recently, some studies have shown that the introduction of halogen atoms in the carbon matrix could improve the ORR performance (Jeon et al. 2013). The electrochemical ORR catalytic performance of non-precious metals mainly focuses on the high catalytic performance and long-term durability. Shen et al. proposed a dual-template methodology using eutectic NaCl/ZnCl₂ melt to achieve a 3D mesoporous skeleton structured carbon from lignin (Shen et al. 2020). The as-prepared lignin-derived N, S and Cl-doped carbon materials have a high SSA of 1289 m ² g⁻¹, a large pore volume of 2.80 cm ³ g⁻¹ and a well-connected and stable structure. It is assumed that the excellent electrochemical activity and stability of the ORR catalyst in acidic media are attributed to the presence of C–Cl covalent bonds in the carbon material. The prepared carbon materials exhibited excellent catalytic performance and durability in the acidic electrolyte, showing an extremely high half-wave potential (E_1/2 = 0.792 V), and the electron transfer number is around 3.93, indicating an extremely low H₂O₂ yield (< 5%), the Tafel slop (43 mV dec ⁻¹), with high retention of the limiting current density maintaining 89% after 5 thousand cycles. The prepared carbon material was employed in a H₂-O₂ single fuel cell which exhibited a maximum power density of 779 mW cm⁻².

6.2 Photocatalysis

Utilizing lignin waste to develop a value-added material has been attempted in a variety of applications. LDC can be applied as a substrate for the synthesis of a carbon-supported photocatalyst. Basically, titanium dioxide (TiO₂) is generally considered to be the most commonly used photocatalyst under UV light stimulation (Nair et al. 2016).

Carbon-based material could be coupled with photocatalyst as an electron reservoir to trap photo-generated electrons facilitating the separation of photo-generated charge carriers. (Srisasiwimon et al. 2018). Thus, the modification of TiO₂ with lignin, as a carbon source, to form composite photocatalysts for improving their photocatalytic performance is tried in practice. As an example, TiO₂/lignin-based carbon composite photocatalyst was synthesized by a sol − gel microwave technique. Lignin as a carbon precursor for synthesizing composite photocatalysts showed improvement in the performance of TiO₂. Under UV irradiation, the conversion rate of lignin was as high as 40%, and the photocatalytic activity was enhanced. High-value chemicals such as vanillin were detected after photocatalysis by gas chromatography-mass spectrometry (Srisasiwimon et al. 2018). Structural modifications of TiO₂ with carbon have been described as a promising way to enhance the titania activity under visible radiation (Sakthivel and Kisch 2003). A. Gómez-Avilés et al. have tested the degradation of a model drug (acetaminophen) in water by C-modified TiO₂ photocatalysts under solar irradiation (Gómez-Avilés et al. 2019). The photocatalyst achieves complete conversion of acetaminophen in only 1 h under solar irradiation.

LDC could also be used to decorate ZnO photocatalyst. Zhang et al. fabricated a lignin-based hollow carbon (LHC) with a thin shell and opened structure by low-cost EHL and MgO as the template. Then ZnO nanoparticles were immobilized on the exterior and interior surface of LHC (ZnO/LHC). This two-step method avoided excessive carbon coating on the surface of ZnO (Zhang et al. 2020a).

Zhang et al. further synthesized a lignin-based flower-like carbon (LFC) using Mg(OH)₂ as a structural inducer (Zhang et al. 2020b). The LFC presented 3D structure, which is beneficial to anchor ZnO nanoparticles. The ZnO/LFC composite photocatalyst demonstrated extended light absorption and separation of photogenerated electron–hole pairs because of the electronic interface interaction between ZnO and LFC. Wang et al. presented a novel LC/ZnO hybrid composite synthesized through a facile yet efficient one-pot carbonization method using industrial lignin (Wang et al. 2017). The obtained LC/ZnO hybrid composite exhibited significant enhancement in photocatalytic performance compared with pure ZnO nanoparticles.

6.3 Adsorption

Carbon dioxide emissions are considered a major contributor to climate change. CO₂ adsorption by solid adsorbents, such as zeolites and metal − organic frameworks, have demonstrated to be efficient for CO₂ capture. Porous carbons remain superior CO₂ adsorbents given their stability and energy intake during regeneration (Saha and Kienbaum 2019). Bio-based resources are highly attractive as carbon precursors that can be produced sustainably in a large scale. Lignin shows promises to be a valuable source owing to its high carbon yield, but the hydrothermal treatment of lignin usually produces interconnected irregular topologies.

Zhao et al. synthesized multi-scale carbon superaparticles (SPs) by combining lignin nanoparticles and microbeads with cellulose nanofibers (CNFs) via a soft-template method (Fig. 17a) (Zhao et al. 2021). They prepared lignin SPs by evaporation induced self-assembly (EISA) of suspensions containing LPs and CNFs, and the morphology of LP-based SP and 2.5 wt% CNF-assembled carbon SP was compared at different CNF mass fractions (Fig. 17b). The CO₂ adsorption capacity was compared under conditions of SSA, carbonation temperature and nitrogen content (Fig. 17c-f). The carbon SPs formed by controlled sintering during carbonization developed high mechanical strength (58 N·mm⁻³) and surface area (1152 m²·g⁻¹). As a result, the carbon SPs offer hierarchical access to adsorption sites that are well suited for CO₂ capture (77 mg CO₂·g⁻¹).

Fig. 17

a Superstructuring of lignin particles (nanometer-scale) into millimeter-scaled carbon supraparticles (SPs) for use in CO₂ capture. The self-assembled lignin SPs are converted into carbon-based, hierarchically structured materials following controlled carbonization and activation; b Schematic illustration of the preparation of lignin SPs via EISA of suspension containing LPs and CNF. SEM images show the morphology of LP-based SPs and 2.5 wt% CNF-assembled carbon SPs at different CNF mass fractions; c Specific surface area (SSA) and corresponding CO₂ uptake capacity as well as (d) areal CO₂ uptake of carbon SPs obtained at varying carbonization temperatures; e Nitrogen content and SSA and (f) CO₂ uptake capacity as a function of nitrogen content and SSA by carbon SPs-800 activated during given time by steam (N-free carbon SPs) and ammonia-steam (N-doped carbon SPs). Reproduced with permission from Ref. (Zhao et al. 2021). Copyright © 2021, American Chemistry Society

Geng et al. described a new family of carbon aerogel with a hierarchical structure from kraft lignin/TOCNF precursors and synthesized by ice-templating and further carbonization (Geng et al. 2020). This work demonstrated the suitability of using lignin and TOCNFs, which can be extracted from various renewable resources, to prepare multifunctional carbon aerogels. The best results exhibited a CO₂ adsorption capacity of 5.23 mmol g⁻¹ at 273 K and 100 kPa and a gravimetric specific capacitance of 124 F g⁻¹ at a current density of 0.2 A g⁻¹.

Adsorption is considered as an efficient method for capturing contaminants from waste water (Da̧browski 2001). As discussed before, Lignin has abundant oxygen-containing groups (− C = O and − OH) in its carbon skeleton, which is beneficial for the development of functionalized lignin-derived adsorbents. Chemical activation and physical activation strategies could be used to enlarge the SSA of LDPCs to enhance the removal efficiency of dyes. Xu et al. designed and prepared a pH-responsive, easily regenerated, high-capacity, and highly selective lignin-derived sulfonated porous carbon (LSPC) by using Fe₃O₄ as a hard templating agent (Fig. 18a) (Zhu et al. 2021). The effect of pH on adsorption was explored (Fig. 18b-c), explaining the mechanism of adsorption and desorption of α-MB on LSPC (Fig. 18d). After concurrent acid etching and sulfonating process, porous structure of LSPC was obtained, and high density of negatively charged groups (including − SO₃H, − COOH, and phenolic − OH) were anchored onto the LSPC. The maximum adsorption capacity achieved was 420.40 mg/g for methylene blue (a-MB). Moreover, the a-MB adsorbed onto the LSPC could be desorbed in acidic methanol solution. The regenerated LSPC could be recycled for a-MB removal in absorption–desorption cycles.

Fig. 18

a Schematic illustration of the preparation of LSPC; b UV–vis spectra of α-MB solutions (C₀ = 20 mg/L, pH = 2 and 11) before and after 3 min adsorption onto LSPC (inset: photographs of α-MB solutions before and after 3 min adsorption), c effect of pH values on the adsorption of α-MB onto LSPC (C₀ = 500 mg/L), and d the adsorption and desorption mechanism of α-MB onto LSPC. Reproduced with permission from Ref. (Zhu et al. 2021). Copyright © 2021, Elsevier

Chromium is a representative heavy metal, and Cr(VI) is highly carcinogenic. Cr(VI) usually exists in water in the form of oxygen anions such as chromate. Because of the large hydration radius of oxygen anions, Cr(VI) is difficult to be adsorbed by adsorbents. N and S can provide electrons to reduce highly toxic Cr(VI) to nontoxic or low-toxicity Cr(III) (Saha and Orvig 2010). LDPCs could contain a large number of O groups and S groups that can effectively reduce Cr(VI) (Doherty et al. 2011). Hydrothermal-induced assembly technology can add oxygen-containing groups (-OH, etc.) on LDPCs and thus can also be used for incorporating specific elements into the materials (Liang et al. 2021b). The generated hydrothermal carbon has a drawback of low SSA, and the structure is agglomerated, with little pore structure (Liang et al. 2021a). Alkaline activation strategies can ameliorate this defect to prepare adsorbents with more pores and a very high SSA. For instance, Liang et al. prepared lignin-based carbon spheres (L-CSs) by hydrothermally induced assembly using industrial alkali lignin as raw material (Liang et al. 2022). A novel lignin-based hierarchical porous carbon (L-HPC) with enhanced SSA showed excellent adsorption performance of Cr(VI), and the adsorption capacity reached 887.8 mg g⁻¹.

Ni(II) is one of the toxic metals present in wastewater, which poses a serious threat on the health of human beings and the local ecosystem. Activated carbon as an efficient adsorbent is widely applied to remove Ni(II) from wastewater due to its high SSA, extensive adsorption capacity and various surface functional groups. Gao et al. used papermaking black liquor lignin as raw material and prepared activated carbon with high SSA and low cost by pre-carbonized KOH activation method. The feasibility of efficient Ni(II) adsorbents was described (Gao et al. 2013).

6.4 Oil/water separation

In recent years, industrial wastewater containing oil, leaking petroleum products during extraction, storage, transport and discharge worldwide have become major environmental issues and global challenges. Cleaning of contaminants through oil–water separation is an effective method, where high efficiency and low cost adsorbents are essentially needed. There is a growing interest in the development of new materials from biomass with functional additives, such as lignin owing to its intrinsic properties.

3D porous foams or sponges are receiving increasing attention worldwide due to their lightweight, high porosity, low density, high SSA and high adsorption capacity (Priyanka and Saravanakumar 2018). Its large SSA, higher thermal stability, higher electrical conductivity, ultra-lightweight, good reusability and superhydrophobicity provide better conditions for oil–water separation. Vannarath et al. used two biopolymers, lignin (extracted from betel nut shells) and starch, to synthesize a lignin carbon foam by pyrolytic carbonization technology with zinc nitrate as an activation agent (Vannarath and Thalla 2021). The lignin and zinc oxide particles are embedded in the carbon foam, making it an ultra-light, hydrophobic and thermally stable material. The synthesized lignin-carbon foam is ultralight (density = 0.0294 g cm⁻³), excellent hydrophobic (water contact angle was 124^◦), mesoporous (3D cell-like structure), fire-retardant and thermally stable. The foam showed an excellent adsorption capacity for different oils, and the highest sorption was observed for diesel oil (7842.71 mg g⁻¹).

7 A summary of current status and future trends

LDC and LDPC are promising carbon materials for next-generation commercial applications, given the controllable SSA, morphology, surface chemistry, and microcrystalline structures (Fig. 19). The overall target for LDC and LDPC development is to design green, sustainable chemical processes with low production cost (Zhang et al. 2021b), and to tune the microcrystalline structure, pore distribution, and surface chemistry (Yin et al. 2020a). LDC and LDPC can be applied in a variety of applications. Specifically, LDC and LDPC are promising in supercapacitors, batteries, energy storage, catalysts, and environmental remediation applications.

Fig. 19

A summary of the future research trends of LDC materials

Low energy density and high energy storage cost are two obstacles impeding the development of supercapacitor commercialization. LDPCs for environmental applications are price-sensitive, in which the fabrication cost should be much lower than LDPCs for supercapacitors. In this regard, to realize the commercial application of LDPCs toward supercapacitors and environmental application, the following issues need to be addressed: (1) The uncertainty of lignin raw materials with great difference in the molecular structure of lignin results in the instability of the structure and properties of LDC products. Pre-carbonization at low temperatures could be developed to alleviate the difference between different lignin precursors; (2) Production of LDPCs mainly by carbonization and activation at high temperatures, is chemical-consuming, energy-consuming, highly corrosive, and not environmentally friendly. The development of green and efficient activation technology, such as direct pyrolysis and low-temperature activation, to replace the traditional activation of potassium salt is an urgent task for the large-scale production of LDPCs; (3) Although the rich oxygen in lignin is beneficial to improve surface wettability, oxygen functional groups tend to cause self-discharge, and excessive leakage current is not conducive to the practical application of supercapacitors, especially for organic electrolyte systems; (4) The energy density of LDPC-based supercapacitor cannot meet the requirement of practical applications, so the precise control of pore size (sub-nano pores and ordered mesopores) is needed to realize the applications of chemical adsorption and to enhance the energy density of supercapacitors.

The storage of alkaline metal ions with different charge storage mechanisms requires that LDC should have different physicochemical properties and microstructures. Generally, hard carbon for Na ion storage should have plenty of inner closed sub-nanometer-sized pores to realize the quasi-deposition of sodium metal in these pores and large interlayer spacing to enable fast and stable intercalation/adsorption. Therefore, novel synthesis methodologies need to be developed to realize internal closed pores and high capacity. On the contrary, the storage of Li and K should fully utilize the intercalation of Li and K ions in the graphitic regions and the adsorption in defect sites.

The morphology control of LDC is limited to spheres, hollow spheres by spray drying or anti-solvent phase separation techniques and fibers by electrospinning. These methods are tedious in practice. For example, electrospinning is hard to be realized in a large-scale manner. Therefore, novel techniques that use low consumption of organic solvents and are high-throughput need to be developed for commercialization. In summary, it is significant to develop novel and facile techniques for the preparation of LDCs and LDPCs with sphere and fiber morphologies to further advance the development of lignin-derived functional carbon materials for future applications.

Developing synthesis and engineering methods for LDC and LDPCs is in the central area of LDC materials. With the pursuit of sustainable development and green preparation engineering methods, we could elaborate a novel pathway for high-value-added utilization of lignin towards commercialization.