Introduction

Although industrialization has improved the quality of life of everyone, it has also created a huge problem of air pollution—a major global health and environmental challenge. The chief sources of toxic gases are industries (coal power plants, refineries, and petrochemicals), coal combustion, mining of mineral resources, transportation (land, water, and air), landfills, and waste management (Damma et al. 2019; Li et al. 2022c; Wang et al. 2014; Zhang et al. 2017). Many methods have been developed for controlling air pollution, including selective catalytic reduction (SCR), oxidation, adsorption, and photochemical treatments (Ahmadi et al. 2021; Li et al. 2023a, c; Zhou et al. 2019). The catalysts used in such remediation methods typically have an active site with a support/carrier to facilitate the mineralization of toxic gases. Metal oxide-based catalysts are excellent commercial candidates for removing toxic gases through SCR, oxidation, and adsorption (Anthonysamy et al. 2018; Kim et al. 2018; Murindababisha et al. 2021). However, their narrow operating window, toxicity, high cost, and thermal instability limit their applicability. Hence, such catalysts need to be further improved.

Porous carbon catalysts have shown great potential for environmental remediation due to their distinctive properties such as a high specific surface area, a distinct structure, tunable porosity, and excellent chemical stability (Chiang et al. 2001; Gao et al. 2019; Tian et al. 2001). These carbon catalysts are commonly used in various applications, including for removing pollutants from air and water, for degrading organic compounds, photocatalytic degradation activity, and for electrocatalysis (especially carbon-based single atom catalyst) (Bai et al. 2023a, b; Guo et al. 2020; Gupta and Saleh 2013; Jin et al. 2022b; Li et al. 2015b; Pan et al. 2022; Shi et al. 2021; Shi et al. 2022b, c; Shi et al. 2023; Tang et al. 2022, 2023a, b, c; Wang et al. 2023b; Zhang et al. 2017). Porous carbon materials can also be sintered as active support materials. Several methods that use porous carbon materials have been modified to improve the catalytic performance of such materials (Anthonysamy et al. 2018). A porous carbon-based catalyst with good adsorption properties and a good support/carrier is a crucial component of catalytic systems for removing toxic gases (Cho et al. 2023). However, the insufficient activity of carbon materials, inadequate waste recycling, and poor high-temperature stability need to be effectively addressed before porous carbon-based catalysts could be used for removing pollutants.

This review summarizes the progress made so far in developing porous carbon material-based catalysts for achieving clean environmental remediation. Although several review papers have dealt with carbon-based materials for applications of various catalytic reactions, it still lacks the systematic study that delves into the utilization of carbon-based catalysts for the clean air environments. The current review mainly focuses on the methods developed for removing major toxic gases such as NOx, VOCs (such as toluene), acidic gases (such as H2S), and elemental mercury (Hg0). It also discusses the modification strategies for the development of porous carbon with physical or chemical treatments, metal oxide doping, and heteroatom doping, as well as the effects of various physicochemical parameters on reducing the release of the above mentioned toxic gases (Fig. 1). Finally, future perspectives and possible directions are presented.

Fig. 1 
figure 1

Schematic illustration of the removal of toxic gases by modified catalyst with different strategies

Full size image

Strategies for Fabricating Modified Carbon-Based Catalysts for Removing Air Pollutants

Carbon as a support material plays a vital role in heterogeneous catalysis because of its superior physical and chemical properties compared to those of other supports (Fidalgo and Menéndez 2011). Support materials with a high surface area, suitable chemical composition, and porous structure can accommodate catalytic species/active phases to improve the activity or selectivity of the corresponding catalyst, which can also influence the dispersion and accessibility of catalytic species on the surface of carbon support materials. Catalysts can be further modified by physicochemical treatments, using metal oxides, and through heteroatom (N, O, and S/P, B) doping to improve their activity.

Modification of Carbon Support Materials

Physical or Chemical Treatment/Activation

The aim of activation, which involves physical or chemical treatment, is to improve the porosity, pore volume, and surface area of carbon materials. Activation through physical treatments mainly involves thermal treatment (pyrolysis or carbonization) at high temperatures through steam or gas purging (using gases such as CO2, steam, and N2). The oxidation of the carbon source material at high temperatures (i.e., the conversion of surface carbon and hydrogen into CO2 and H2O, respectively) changes the surface area and porosity in a controlled way (Li et al. 2015b; Panwar and Pawar 2022). CO2 activation forms micropores, while steam activation increases their size. Chemical treatments commonly employ chemical agents [such as KOH, NaOH, H3PO4, H2SO4, and (NH4)2SO4] in one- or two-step processes. These treatments improve the surface area and pore structure of the catalyst and increase its surface oxygen functional moieties (Ku et al. 1994). In a one-step process, the carbonization and activation of the precursor material can be achieved simultaneously by using a chemical agent (Li et al. 2015b). In contrast, the two-step process involves carbonization, followed by activation using chemical agents or pretreatment of the precursor material with a chemical agent before carbonization (Guo et al. 2007).

Metal Oxide Doping

Carbon materials can be modified by doping with metal oxides (mostly transition metal oxides) that improves the physical and/or chemical properties of carbon, which further enhances their air-pollutant-removal efficiency through SCR, catalytic oxidation, or adsorption (Table 2).

Heteroatom Doping

Heteroatom doping refers to the incorporation of noncarbon elements (such as N, S/P, B, and O) into the carbon lattice. Heteroatom doping introduces new chemical functionalities, such as Lewis acid sites, which can interact with toxic gases and facilitate their adsorption and/or reaction. In addition, the heteroatoms can alter the electronic properties of the carbon lattice and make it more polar and enhance its affinity for polar gas molecules, which further helps control the catalytic reaction or remove toxic gases such as NOx, H2S, Hg0, and toluene (Table 3). Heteroatom doping into porous carbon not only serves as a highly efficient catalyst, but can also be utilized as an ideal support for immobilizing metal nanoparticles (Ma et al. 2016).

Heteroatom doped (S/N) carbon-based single-atom catalyst leads to enhance the high stability, catalytic reaction, and/or electrocatalysis activity (Bai et al. 2023b; Tang et al. 2023b, c); wherein, it observed that benefiting from the synergic action of metal active sites, N-doped graphene nanosheets (FeCo-NG) exhibited higher catalytic performance, i.e. ORR. Anchoring metal on S/N co-doped graphene (Mn-S/N-C) improved the electrocatalytic performance of the catalyst system (Bai et al. 2023b). The d-orbital adjustment of the M-center (Mn) in M-N-S-C catalyst material by S/N co-doping benefits to improve the electrocatalytic activity, which is most superior to previously reported Mn-based electrocatalysts and commercial iridium dioxides. The S-doping and presence of asymmetric structure contributes to the ortho-Mn1-N2-S2G site which is more active than Mn_N4G, Mn-N3SG, para Mn-N2-S2G, and Mn-NS3G sites. The ORR rate-limiting steps on the ortho-Mn1-N2-S2Cx have been predicted as the transformation of OH* to H2O (Bai et al. 2023b).

Applications of Carbon-Based Catalyst for Clean Environmental Remediation

NOx Removal

In general, steam/gas purging increases the pore volume (Wang et al. 2016) and develops internal pores by removing trapped volatile gases or particles during carbonization. An increase in the steam-activation temperature enhances the pore volume and specific surface area (SSA) of the catalyst. The SSAs of PCNF10-800, PCNF10-850, and PCNF10-900—polyacrylonitrile (PAN)-based modified carbon materials—are 778, 876, and 1206 m2/g, respectively, while their total pore volumes are 0.272, 0.315, and 0.562 cm3/g, respectively. Increased surface area and pore volume improve the NO-removal efficiency (Wang et al. 2016). Use of different carbonization and steam-activation temperatures has an obvious impact on the NO conversion. The PCNF catalyst carbonized and steam-activated at 800 °C exhibited 60% NO conversion performance, whereas the catalyst only activated at 800 °C afforded 27.6% NO conversion at room temperature (Table 1). Ku et al. (1994) analyzed the catalytic activity of the coconut-shell-derived activated carbon (AC) prepared via chemical activation using O2-NH3, H2SO4, and (NH4)2SO4/H2SO4. The carbon catalyst treated with (NH4)2SO4/H2SO4 exhibited the highest surface area and catalytic activity (Table 1). Li et al. (2015b) examined the effects of physical activation, chemical activation, and co-activation of the Sargassum-based activated carbon (SAC) for low-temperature SCR of NOx. The Co/SAC-2 (co-activated) catalyst afforded the highest NO conversion achieved so far (82.05%) at 125 °C, whereas C/SAC-2 (chemically activated) could afford 75% NO conversion) and P/SAC (physically activated) could afford only 45% NO conversion. Hence, it is clear that chemical co-activation improves the NOx-removal performance.

Table 1 Carbon modified through physical/chemical treatment and its performance in removing toxic gases
Full size table

The porous nature of carbon allows for a better diffusion of gaseous pollutants through the catalyst, which promotes contact between the reactants and the catalyst. Furthermore, carbon modified by metal oxide doping can be utilized for NOx removal, mostly by low-temperature NH3-SCR. In addition, these superior properties can enhance the tolerance to catalyst poisoning. Zhu et al. reported that doping with 5 wt% of V2O5 enhanced the NOx-removal efficiency of the AC catalyst from 20 to ~ 80% at 220 °C with a better SO2 poisoning tolerance, which could be owing to the increased catalyst surface acidity because SO2 oxidizes to SO4−2. In addition, the pre-oxidation of the catalyst plays a crucial role in improving the NOx-conversion performance (to up to 98%) (Zhu et al. 1999). Wang et al. (2014) described that doping activated semicoke (ASC) with 3 wt% of V2O5 improved the SCR catalytic performance (from 25% to > 90%) at 250 °C. Li et al. (2022b) demonstrated that the Ce-modified V2O5/AC afforded a higher NOx conversion (i.e., 90%) than that achieved with the unmodified V2O5/AC (< 80% NOx conversion) at 200 °C (Table 2). They also discussed the effect of the regeneration temperature (470 °C) and regeneration cycling on the improved SCR performance (i.e., > 95%) of the Ce-V2O5/AC catalyst. This result suggests that cyclic regeneration is conducive to desulfurization and SCR. Huang et al. (2007) reported that doping CNTs with 2.35 wt% of V2O5 enhanced the NOx conversion from < 10 to > 80% at 190 °C. However, V2O5/CNTs prepared using CNTs with an outer diameter of 60–100 nm afforded better NOx conversion (92%), perhaps owing to the dispersion of vanadium over the CNT surface and presence of oxygenated groups on the CNT surface with different diameters (during HNO3 treatment). Liu et al. (2021) reported a highly efficient MnOx-doped biochar catalyst obtained by air oxidation for the SCR of NOx, where pre-oxidation (400 °C) and post-oxidation (250 °C) were performed before and after doping MnOx into the biochar. After pre-oxidation, the SSA and total pore volume of the sample increased from 557 to 734 m2/g and from 0.353 to 0.440 cm3/g, respectively, whereas post-oxidation increased the contents of Mn4+ and chemisorbed oxygen, which enhanced the surface acidity and redox capabilities of the catalyst. The catalyst exhibited the highest NOx conversion, i.e., 97% at 150 °C, compared with that achieved without air oxidation (61.8%). Jiang et al. (2019) investigated the low-temperature NOx-removal performance and high poisoning stability of a MnCe-doped AC and a V2O5 co-doped MnCe/AC catalyst. The doping of the AC with MnCe enhanced the NOx conversion from 25 to > 90% at 200 °C. Doping with 0.4 Ce/(Mn + Ce) further increased the NOx conversion to > 97%. The co-doping of V2O5 slightly improved the NOx conversion to > 98%, although it also broadened the reaction window (100–300 °C). In addition, it prevents the active sites from being blocked or poisoned. Figure 2 shows the mechanism of improved SO2 tolerance over the Mn-Ce(0.4)-V/AC catalyst.

Table 2 Metal oxide-doped carbon and its toxic gas-removal performance
Full size table
Fig. 2
figure 2

Mechanism of better SO2 tolerance over Mn-Ce(0.4)-V/AC and effect of V-oxide doping over SCR of NO (Jiang et al. 2019)

Full size image

Jia et al. (2022) reported MnOx-doped biochar and the effect of the modification method for biochar on the NOx-conversion performance at low temperatures. NaOH-modified biochar doped with 25% MnOx resulted in the highest NOx conversion (95%) and a high tolerance to SO2 and H2O at 225 °C, compared to those achieved with acid- and CTAB-modified biochar. Highly dispersed MnOx is easily formed on the NaOH-modified biochar surface, which is crucial for improving the SCR performance. Li et al. (2023a) demonstrated that doping coal-based AC with 10 wt% of Fe2O3 improved the SCR performance from 15 to 30%; however, 10 wt% of Fe2O3/OAC1.5-60-3 (OAC: oxidized AC with APS oxygen functionalization) dramatically enhanced NO conversion to > 99% at 180 °C. In addition, the reaction temperature window broadened (from 120 to 250 °C) with superior NOx conversion (> 90%) and high tolerance to SO2 and H2O. This result suggests that the physical structure was not the key factor in the APS oxygen functionalization strategy, which affects the catalyst performance. Xue et al. (2008) reported 100% NOx conversion of the CuO-impregnated wet oxidized AC (treatment with H2O2) catalyst at 272 °C, wherein the surface oxygen moieties and NO adsorption on the AC played a crucial role in the production of carbon active sites. Notably, it is not the rate-limiting step for the catalytic reduction of NO over CuO by AC. Zhu et al. (2011) reported that surface-modified AC fibers doped with 9 wt% of CeO2 exhibited high NOx conversion (93.96%) at 180 °C; the surface modification of the ACF by HNO3 afforded a better catalytic activity (i.e. > 90%) over a broad temperature window (150–240 °C) compared to that achieved with the oxygen plasma treatment (< 85% NOx conversion).

Heteroatom doping (especially that of N) into carbon can be used to modify its physical and chemical properties to enhance the NOx-removal performance (Table 3). Li et al. (2014) reported the effects of impregnation duration (5 h) and calcination temperature (900 °C) on the N-content and NH3-SCR activity of the modified AC. ACM-5–900 exhibited 51.67% of NO conversion at 80 °C, while that shown by undoped AC was 21.92%. However, they demonstrated that the form of N-containing functional groups {ACM-5–900 had highest proportion of N-6 (pyridinic-N) groups, i.e., 57.3%} influenced NO conversion rather than the total N-content. In addition to the impregnation duration, the N-doped precursor also played a crucial role in improving the NOx conversion of the modified AC, as shown schematically in Fig. 3. The N-doped ACM-5 catalyst with a KHCO3 promoter exhibited a higher NOx-conversion efficiency (52%) than that the reported N-doped carbon catalysts (30%). In contrast, undoped AC showed only 10% NOx-conversion efficiency (Li et al. 2020d). N-6 was reported as an activated site for the catalytic direct decomposition of NO at 500 °C in N-doped porous carbon; in the absence of N-doping, the carbon catalyst showed only < 10% NO conversion (Wang et al. 2018b). Yao et al. (2020) reported the N-doped semicoke-based catalyst, i.e., ASC-10U10Mn, with a high NOx conversion of 94.5% at 275 °C, while the ASC showed 10% NOx conversion. They also demonstrated that N-groups with unpaired electrons (N-6, N-5 pyrrolic, and N-Q- quaternary) play a crucial role in enhancing the adsorption and oxidation of NO and NH3 adsorption owing to abundant Lewis acid sites. N-doping also improves the electron distribution of the catalyst and the electron mobility between the Mn and oxygen moieties. Figure 4 shows the promotion mechanism of N-doping over the MnOx/semicoke catalyst for low-temperature SCR.

Table 3 Heteroatom-doped carbon and its toxic gas-removal performance
Full size table
Fig. 3
figure 3

Promotional effect of N-doping that improves the NOx conversion of modified AC (Li et al. 2020b)

Full size image
Fig. 4
figure 4

Promotional mechanism of N-doping on MnOx/semicoke catalyst for low-temperature SCR (Yao et al. 2020)

Full size image

Lin et al. (2018) reported the effects of N-dopant precursors/additives on the SCR of NOx over the modified AC, where the N-doped AC with the pyridine additive/precursor exhibited a higher NOx conversion (66%) than that achieved with other catalysts. Zhu et al. (2019) investigated the influencing mechanism of N-doping on NO2 adsorption and reduction over the AC using computational studies. AC600 with a higher N-content exhibited the highest NO2-adsorption capacity (135.6 mg/g) and the lowest NO-release percentage. Li et al. (2020c) reported N-doped porous biochar with 82% NOx conversion at 260 °C, where N-6 played a major role in boosting up the denitrification activity. However, raw biomass reacts more readily with N-containing additives, which facilitates carbon formation with an ultrahigh N-content, resulting in the formation of modified carbon with an N-content of 17.71 at. % (N-6: 9.09 at.%). This modified carbon catalyst demonstrated > 90% of NO conversion at 200 °C, which is higher than that achieved with non-N-doped carbon (< 20%). Li et al. (2023b) investigated the synergistic effect of dual heteroatom dopants (N and O) on the carbon-catalyzed NH3 SCR of NOx. The N- and O-doped AC (NOAC) was prepared through one-step NH3·H2O activation of the AC, followed by HNO3 oxidation. The NOAC exhibited enhanced NOx conversion at 200 °C (~ 90%) compared to that shown by the O-doped (~ 62%) or N-doped AC (~ 73%); non-doped AC exhibited only 30% of NOx conversion. In addition to N and O dopants, B and P dopants in carbon catalysts also demonstrated effective NH3 SCR of NOx (Li et al. 2020e; Yang et al. 2021), wherein the Lewis acid properties of the dopants can enhance the adsorption and activation of the reactants (NH3 and O2).

H2S Removal

Physically and chemically activated carbon can also be utilized for H2S removal, mainly by physi-/chemisorption. Physisorption is a relatively weak, nonspecific adsorption that involves intermolecular forces between the adsorbate (H2S) and the adsorbent (with carbon as the solid surface). Physisorption of H2S occurs when weak van der Waals forces form between the sulfur atoms of H2S and the surface atoms or molecules of the adsorbent material. Physisorption is reversible and typically occurs at low temperatures and high pressures. It does not involve any significant chemical changes to the adsorbate or adsorbent. In contrast, chemisorption is a stronger, specific adsorption that involves the formation of chemical bonds between the adsorbate and the adsorbent. Chemisorption of H2S occurs when its sulfur atom forms chemical bonds with the atoms or molecules on the adsorbent surface. Chemisorption typically involves breaking and forming of chemical bonds, leading to a more stable adsorption state. Chemisorption is generally irreversible and can occur at lower pressures and higher temperatures than physisorption. Cattle manure-based AC catalysts are generally prepared through steam and CO2 activation at 850 °C to remove H2S. The AC4 catalyst prepared by pyrolysis at 650 °C and activated with CO2 exhibited the highest SSA (408.36 m2/g) and H2S adsorption capacity (868.45 mg/g) compared to those of AC (AC3) activated with steam. These results indicated that pH and surface area, rather than porosity, are crucial factors for the H2S adsorption (Tuerhong and Kuerban 2022). Su et al. also reported an enhanced surface area with an increase in the pyrolysis temperature, wherein the coconut husk carbonized at 500 °C showed small SSA (0.18 m2/g), which dramatically increased after activation at 650 °C (331.9 m2/g), 750 °C (604.1 m2/g), and 850 °C (811.4 m2/g). However, the surface area does not follow the same trend as that followed by the H2S-adsorption capacity; the adsorption capacities of the catalyst activated at 650, 750, and 850 °C are 29.7, 38.7, and 26.7 mg/g, respectively (Table 1). Therefore, the H2S-adsorption capacity is not entirely dependent on the SSA; instead, it is related to the pore size distribution (Su et al. 2021). Similarly, Bazan et al. suggested that the SSA and micropore volume are crucial for enhancing the H2S-adsorption capacity of pistachio nutshell-based catalysts prepared via H2O/CO2 activation. When the total and micropore volumes of the pistachio nutshell-based catalyst were 0.19 and 0.14 cm3/g, respectively, the H2S adsorption was 7.1 mg/g, which further increased to 11.3 mg/g with an increase in the total and micropore volume (0.64/0.54 cm3/g) of the catalyst. The SSA is another important factor that affects adsorption under wet conditions (Bazan-Wozniak et al. 2017). Nicolae et al. (2022) reported that the pore volume, rather than the SSA, affects the H2S-adsorption performance. KOH activation of the GS_KOH_1 catalyst prepared using guava seeds by hydrothermal carbonization has more micropores, which increase the gas uptake (48.18 mmol/g; in contrast, the GS_KOH_1 catalyst without KOH activation demonstrated an uptake of only 1.52 mmol g−1), compared to the carbon catalyst activated with H3PO4 and K2CO3 that exhibited micropore volumes of 0.46 and 0.58 cm3/g with H2S-adsorption capacities of 7.1 and 11.3 mg/g, respectively (Bazan-Wozniak et al. 2017). Guo et al. also reported that the chemical activation and micropore volume of oil palm fiber-based AC enhanced the H2S-removal performance (68/76 mg/g; KOH/H2SO4 activation) compared to that achieved by physical activation (48 mg/g, CO2 activation) (Table 1). H2SO4-activated carbon has a larger micropore volume (0.28 cm3/g) than that of KOH-activated carbon (0.25 cm3/g) (Guo et al. 2007). Hence, it is clear that H2S removal is affected by the combined effects of the SSA, pore volume (total/micro), humidity, and functional moieties.

Doping porous carbon catalysts with metal oxides can modify the physical and chemical properties of such catalysts, which would in turn improve the adsorption capacity of these catalysts for H2S. AC was doped with various metal oxides (such as ZnO, MgO, CuO, ZnFe2O4, MnOx, V2O5, CoOx, and CeO2) to improve the adsorption and catalytic oxidation of H2S (Table 2). Yang et al. (2020b) reported that MgO0.2ZnO0.8-doped AC exhibited the highest H2S-adsorption capacity (113.4 mg/g among catalysts with different molar ratios of Mg/(Mg + Zn) at 30 °C). The H2S-removal performance was linked to MgO; its basicity continuously increased the formation of HS− for reactive adsorption on active ZnO and for its catalytic oxidation to elemental sulfur. Doping of AC with Zn and Cu oxides (i.e., Cu0.5Zn0.5O/AC) afforded 50 mg/g of H2S-adsorption capacity at 30 °C, which is higher than that achieved with ZnO/AC (33 mg/g). Doping with low amounts of Cu reduced the diffusional limitations in the lattice of the active of the composite phase and through the reacted overlayer—a known issue for ZnO-based sorbents at low temperatures (Balsamo et al. 2016; de Falco et al. 2018). Yang et al. (2020a) reported the use of 10 wt% ZnFe2O4-doped AC as a reusable adsorbent for H2S removal. It is regenerated at 500 °C and exhibits a higher adsorption capacity (122.5 mg/g) at room temperature than that achieved with AC (5.6 mg/g) and ZnFe2O4 (1.6 mg/g) alone. This result reveals the synergistic effect between ZnFe2O4 and AC on the removal of H2S. AC was doped with various metal oxides (such as Mn, Cu, Fe, Ce, Co, and V) to observe the effects of metal oxide doping on their H2S-adsorption capacities. The Mn oxide-doped AC catalyst exhibited the highest H2S-adsorption capacity (142 mg/g) among all the catalysts. This high capacity remained unaffected even after four consecutive adsorption–regeneration cycles. The SSA of all the metal oxide-doped AC catalysts increased only slightly from 876 to 905 m2/g, which is an unusual behavior (Fang et al. 2013). The copper-oxide-doped biochar exhibited an outstanding H2S-adsorption capacity (1191.1 mg/g) at 125 °C, wherein along with the metal oxide concentration, the microwave steam activation and calcination temperature also played a crucial role in enhancing the H2S-adsorption efficiency (Cui et al. 2022). Zhang et al. (2016) reported an MCM-MgO-15 catalyst with the highest H2S-adsorption capacity (2.46 g/g) along with a high total pore volume (1.74 cm3/g) (Table 2). They prepared AC from resorcinol with carbonation at 800 °C, followed by the caustic impregnation of MgO. The caustic impregnation and its loading determine the performance of the catalyst; however, MgO performed better than conventional salts did, such as Na2CO3, NaOH, K2CO3, and KOH. Hence, the surface area was not a major factor affecting the H2S-removal performance; instead, large pores and high mesoporosity are essential for improved H2S removal.

Heteroatom doping is considered an efficient strategy for promoting the catalytic process of carbon, such as desulfurization (Table 3). In particular, N-doped carbon can act as a basic site, enhancing the electron-donating ability of the carbon lattice, and further increasing its redox activity and corrosion resistance. Liang et al. (2020) discussed N-doped carbon with KOH activation via a calcination-induced self-assembly route, which exhibited a high SSA of 1538 m2/g and a H2S-adsorption capacity of 10.5 mmol/g (0 °C, 1 kPas). N-6 is responsible for the improved H2S-removal performance of N-doped carbon. Kan et al. (2019) also fabricated the N-doped ordered mesoporous carbon (N-OMCST-700) with KOH activation; however, the hydrothermally carbonized catalyst exhibited superior performance for the selective adsorption of H2S with an adsorption capacity of up to 13.4 mmol/g (0 °C, 1 bar). KOH activation of the catalyst increased the SSA (1575 m2/g) and pore volume (0.52 cm3/g). Such catalysts also exhibited a high N-content (4.46 wt%). DFT (Density Functional Theory) studies demonstrated strong interactions between the N-6 and N-5 sites. In addition, the basic surface nitrogen sites also participated in the dissociation of H2S to HS− and H+, thus initiating the selective oxidation of H2S to elemental sulfur. Wu et al. (2022) described N-doped AC with a high N-content (17.2%) and a high meso-pore volume, which exhibited a high breakthrough sulfur capacity (1872 mg/g) for the catalytic oxidation of H2S under KOH activation. K2CO3 activation enhanced the SSA (from 187 to 2459 m2/g) and total pore volume (from 0.255 to 1.453 cm3/g) of the N-doped hierarchical carbon compared with those of N-doped and only K2CO3-activated porous carbon. It also exhibited excellent H2S removal with a sulfur capacity of 426.2 mg/g at room temperature compared to that shown by undoped porous carbon (i.e. 12.5 m/g); N-5 and N-6 acted as active sites for H2S adsorption (Chen et al. 2021). Chen et al. (2020) also highlighted the N-doped porous carbon (NPC) with a high SSA (1419 m2/g) and total pore volume (0.80 cm3/g); the H2S-adsorption capacity (205.06 mg/g) is an obvious effect of K2CO3 activation over N-doped carbon prepared by hydrothermal carbonization. The hydrothermal temperature and duration helped increase the proportions of N-5 and N-6 in the porous carbon, which played a crucial role in H2S removal. The proposed mechanism for desulfurization using the NPC is detailed in Fig. 5. Chen et al. (2022) developed the N-doped interconnected mesoporous carbon with H2O2-assisted hydrothermal carbonization. H2O2 played a crucial role in H2S removal by forming a cross-linked mesoporous structure (the mesoporous volume increased from 0.01 to 0.31 cm3/g) and increasing the numbers of abundant N-5 and N-6 sites. H2O2-activated N-doped carbon exhibited excellent H2S-removal performance, with a H2S-adsorption capacity of 181 mg/g, which was five times higher than that for obtained without H2O2 activation. Ma et al. (2021) reported N-doped carbon with an enhanced H2S-adsorption capacity of 54.8 mg/g, higher SSA (1189 m2/g), and larger total pore volume (0.433 cm3/g) than those achieved using porous carbon without N-doping and activation. Among the N-containing groups, N-6 played a crucial role in H2S removal. Lei et al. (2023) reported a 3D cluster like the N-doped mesoporous carbon catalyst with a high SSA (330.5 m2/g), hierarchical nanopores, and a large N-content (8.6%; where % N-6, N-5 > N-Q, i.e., > 60). Such catalysts exhibited 100% H2S conversion at 150 °C, with 94% sulfur selectivity. In addition, Ahmadi et al. (2021) highlighted that N-doping along with S-co-doping in carbon resulted in a high SSA (2579 m2/g), large total pore volume (1.5 cm3/g), and very high H2S-adsorption capacity (17.19 mmol/g at 100 kPas). DFT studies confirmed the prominent role of N-5 in enhancing the adsorption energy of the molecules over the adsorbents. The S atom, as a codopant around the N-5 atom, controls the adsorption of H2S molecules with a lower adsorption energy, indicating that physisorption is more favorable than chemisorption. S co-doping (defects) plays a complementary role in controlling the adsorption mechanism of gas molecules. In addition, a higher S/N ratio results in better regeneration cycles. Thus, they established the source of novel electronic structures and, thereby, that of N-S-defected localities, both theoretically and experimentally.

Fig. 5
figure 5

Schematic illustration showing H2S oxidation in the N-doped porous carbon (NPC) from waste PU (Chen et al. 2020)

Full size image

Hg0 Removal

Hg0 removal can be achieved through both physi-/chemisorption, which needs high surface area and porosity to produce more active sites. To improve the Hg0-adsorption capacity, Liu et al. (2019) prepared seaweed-based novel ACs and demonstrated high Hg0-adsorption capacity (2382.6 µg/g for SAC-800 and 2909.6 µg/g for EAC-800 catalyst) (Table 1). They demonstrated that an increase in the carbonization temperature enhances the SSA (from 399.57 to 524.57 m2/g in the SAC and from 339.04 to 565.82 m2/g in the EAC), with improved Hg0-removal efficiency (from 40.4 to 88.9% in the SAC and from 34.25 to 90.7% in the EAC). The Hg0-removal process involves both physisorption and chemisorption with suitable SSAs, pore structures, and functional moieties on the modified AC. Yang et al. (2018) also reported that the seaweed-based modified AC afforded an enhanced Hg0-removal efficiency (91%). They revealed that the enhanced SSA (from 1.95 to 26.20 m2/g) and Hg0-removal performance (from 12 to 29.07%) were in line with those achieved at the increased carbonization temperature. Further chemical activation (using NH4Br) decreased the surface area of the catalyst; however, it enhanced the Hg0-removal efficiency, which is an obvious effect of the functional moieties (C–Br and C=O) serving as chemisorption sites. Li et al. developed a clean and modified method (a combination of microwave steam activation and H2O2 impregnation) to develop a porous carbon catalyst. The results suggest that microwave activation and H2O2 modification enhance the SSA (from to 154.92 to 433.03 m2/g in RSW10 and 108.80 to 273.91 m2/g in WSW10) and improve the pore structure (increase the pore volume from 0.217 to 0.2877 cm3/g in WSW10 and from 0.186 to 0.1860 cm3/g in RSW10) of the porous carbon (Li et al. 2021), which further facilitate the Hg0-removal performance of the modified catalyst (i.e. 1293.19 µg/g for WSW10 and 1485.61 µg/g for RSW10) (see Table 1). Dou et al. described that H2O2 modification with UV improved the SSA of the modified AC (1408.61 cm3/g), which further enhanced its Hg0-removal efficiency (from 61.71 to 90.04%). The maximum Hg0-adsorption capacity was 3636.43 µg/g. H2O2 modification has a slight destructive effect on the pore structure; however, the significant increase in functional moieties (e.g.,–OH, C–O, and C=O) along with the chemisorbed oxygen (O*) enhances the mercury-removal performance of the catalyst. A schematic of the modification mechanism of AC using the UV/H2O2 advanced oxidation process (AOP) and the Hg0-adsorption mechanism on the modified AC is shown in Fig. 6 (Dou et al. 2023). Liu et al. (2023b) reported a high Hg0-adsorption capacity of 2052.51 µg/g for corn-stalk-based porous carbon developed by UV/H2O2 clean modification. Hence, it can be suggested that the SSA, pore volume, and functional moieties improve the Hg0-removal performance of porous carbon catalysts. They also demonstrated that 3% of H2O2 is the critical concentration to achieve a high Hg0-removal performance.

Fig. 6
figure 6

Schematic diagram of a modification of AC using UV/H2O2 advanced oxidation process (AOP) and b Hg0 adsorption mechanism on the modified AC (Dou et al. 2023)

Full size image

Wang et al. (2023a) reported a porous carbon catalyst prepared by secondary carbonation and activation with K2CO3, which showed SSA values as high as 2925 m2/g and Hg0-adsorption capacities as high as 571 mg/g. In addition, the Hg0-removal performance remained stable for five consecutive adsorption–desorption cycles. A micromesoporous structure with fewer macropores greatly improves the Hg0-adsorption capacity. The possible reaction pathways for Hg0 adsorption onto the layered porous carbon (LPC) are shown in Fig. 7. Song et al. (2020) demonstrated the effects of carbonization temperature and KOH activation on the Hg0-removal performance of anthracite-coal-based AC. They reported that the SSA increased (from 2.60 to 527.43 m2/g) with increasing carbonization temperature and KOH activation, resulting in a catalyst with a better pore structure that afforded better Hg0-removal performance. Li et al. (2015a) reported different solid-waste-based porous carbons developed using microwave and steam activation, which exhibited improved SSA, pore volume (especially micro), and Hg0-removal efficiency (30–70%). Further impregnation with NH4Cl decreased the pore structure and surface area, although an enhanced Hg0-adsorption capacity of 11,400 g/g was observed (Table 1). Wang et al. (2018a) reported HCl-modified porous carbon obtained from various biochar sources with improved Hg0 adsorption. The Hg0-adsorption efficiency increased from 8.2 to 100% along with a better SSA (from 29.9 to 110.1 m2/g); the macropores played a crucial role in this phenomenon.

Fig. 7
figure 7

Possible reaction pathways of Hg0 adsorption on layered porous carbon (LPC) (Wang et al. 2023a)

Full size image

Doping of porous carbon with metal oxides modifies its physicochemical properties, which can then be utilized to enhance the Hg0 removal through adsorption and catalytic oxidation (Table 2). Gao et al. (2023) reported the effects of doping with metal and bimetal oxides on the Hg0-removal efficiency of AC in a broad operating window (80–360 °C). Bimetal-oxide-doped AC (i.e., 6% Co0.5Cr0.5/AC) showed the optimal Hg0-removal performance (88.3%) at 240 °C; in contrast, the use of undoped AC afforded only 40% H2S removal. This result shows the excellent tolerance of the doped catalyst to SO2 and H2O, which contributed to the redox cycle and synergistic effect between CrOx and CoOx. Consequently, a large amount of sufficiently active Co3+ species and surface-active oxygen were produced. In addition, the formation of a CoCrO4 spinel was observed, which entailed the presence of abundant lattice defects, increment in the SSA and the number of active sites with a strong redox capacity, and a higher dispersion of active phases. Addition of CoOx to CeO2-doped biomass AC (15% Co0.4Mn0.6/BAC) resulted in excellent Hg0-removal efficiency (96.8%) at 230 °C through adsorption and catalytic oxidation, owing to the better texture, lower crystallinity, and strong redox ability of this AC. Moreover, the synergistic effect between CoOx and CeOx generated more Ce3+ and Co3+, inducing a large number of anionic defects and producing more active oxygen with oxygen vacancies (Gao et al. 2018). CoOx-modified MnOx-doped biomass AC afforded 98.5% H2S-removal efficiency at 240 °C, which was higher than that achieved with undoped biomass AC (60%).

Addition of 15% CoOx to Mn-doped biomass AC contributed to the synergistic effect, which increased the SSA and Mn4+ content, enhanced the redox ability and the strength, and restrained the MnOx crystallization, which might be responsible for the enhanced catalytic performance and SO2 resistance of this catalyst (Gao et al. 2019). Yang et al. (2019) reported the effect of microwave steam treatment and iron-copper oxide doping on the wheat straw (WS)-derived carbon to improve the Hg0-removal efficiency. The microwave steam treatment improved the Hg0-removal efficiency of WS-derived carbon from 20 to 70% and further enhanced it to 90.58% (2276.45 µg/g of Hg0-adsorption capacity) in CuFe0.3/WSWU10(500) at 130 °C owing to Cu-Fe oxide doping. The Cu-Fe active phases, chemisorbed oxygen, and lattice oxygen play crucial roles in the removal of Hg0. Chemisorption is the key rate-controlling step. Zhang et al. (2021) discussed the promotional effect of FeOx and the doping effect of MnOx on a biochar-based composite prepared by ball milling for Hg0 removal. The MnFe/char-BM4 catalyst exhibited > 95% Hg0-removal efficiency at 100 °C through the physical adsorption and catalytic oxidation mechanism.

Heteroatom doping (with N, S, and P) is an effective method for modifying carbon sorbents and improving their chemical reactivity for Hg0 removal (Table 3). Zhou et al. (2022a) discussed the P-doped biochar (PBC900) without any additional modification, which exhibited a high SSA (1038.34 m2/g) and large Hg0 adsorption (15,047.64 µg/g); in contrast, undoped BC (BC900) showed an SSA of 394.86 m2/g and a Hg0-adsorption capacity of 36.58 µg/g. In addition, more organic functional groups were generated on the PBC surface, such as C-O=P and C=O. With these groups, the O-C=O group can serve as an electron acceptor, accelerating the electron-migration process for Hg0 oxidization. Zhou et al. (2022b) reported S-doped mesoporous carbon (SMC-900), which exhibited a high SSA (993 m2/g) and excellent Hg0-removal performance (> 97% at 50–150 °C) and discussed the effects of the precursor ratio and carbonization on the mercury-removal performance. The sulfur was in the form of thiophene (C-S-C) and oxidized sulfur (C-SOx-C), and the S-content was 10.29%. However, Vakili et al. (2021) highlighted that the N- and S-doped nanoporous carbon (NSDG-10) exhibited the highest Hg0-removal performance (94.5%), whereas that for unmodified or non-doped carbon was 71%. These results indicate that high SSA and pore volume did not contribute significantly to high mercury adsorption because the N- and S-doped carbon showed better Hg0 removal despite having a lower SSA and a smaller pore volume.

VOC (Toluene) Removal

Physically and chemically activated carbon can also be used for removing toluene. Mohammed et al. (2015) reported a toluene-adsorption capacity of 238.10 mg/g for coconut-shell-based AC with KOH activation. Further, the ammonia-treated AC (PHAC-AM) catalyst exhibited a decreased surface area (from 478 to 361.8 m2/g) and a reduced total pore volume (from 0.61 to 0.16 cm3/g). However, the toluene-adsorption capacity enhanced by 10%, i.e., 357.14 mg/g. Ammonia treatment increases the number of basic surface functional groups, which further improves the affinity of AC for VOCs. Li et al. (2020a) reported a sodium lignin sulfonate-based carbon catalyst with very high toluene adsorption (> 2300 mg/g); the porous carbon was activated by metal salts. Among them, ZnCl2-activated porous carbon exhibited the highest SSA (1524 m2/g) and mesopore volume (1092 m2/g), which facilitated VOC adsorption. In addition, many functional groups play crucial roles in enhancing the toluene adsorption. Highly porous carbon with a toluene-adsorption capacity of 1587 mg/g and an extremely high surface area (4293 m2/g) was developed using waste soybean residues with pre-carbonization, KOH activation, and pyrolysis. The activation temperature and KOH dosage affected the SSA and total pore volume (mesopores and micropores). Porous carbon exposes the aromatic sites (e.g., phenol) for π–π interaction with toluene. Presence of metallic K in porous carbon causes extensive intercalation at the surface of graphitic microcrystals, which exposes the microscale space. The exposed aromatic π system may interact with toluene molecules through strong π–π stacking (Li et al. 2022a). Overall, the SSA, pore volume, and functional moieties play key roles in enhancing the toluene-removal rate.

Metal-doped porous carbon can also be utilized to remove toluene by modifying its physicochemical properties (Lei et al. 2020; Li et al. 2020b). The CuO-doped AC exhibited a toluene-adsorption capacity of 701.8 mg/g at 20 °C. CuO doping mainly increases the pore size, which facilitates the migration and diffusion of toluene and improves its adsorption (Lei et al. 2020). Li et al. (2020b) reported the CuCo0.5/C catalyst with better performance, i.e., 90% toluene removal at 243 °C, owing to a large SSA, more chemisorbed active oxygen species, and a high ratio of Co2+/Co3+.

Toluene removal can also be achieved by using heteroatom-doped modified porous carbon. In particular, the NPC enhanced the surface charge distribution and activity of carbon, which further enhanced the adsorption performance for toluene removal (Table 3). Jin et al. (2022a) developed N-doped porous carbon (NPC-1) with an ultrahigh SSA (3235 m2/g) and total pore volume (2 cm3/g), exhibiting a higher adsorption capacity (691.1 mg/g) compared to those reported previously (Fig. 8), which originated from the use of an alkali metal activator (KOH) and an N-dopant (urea). NPC-1 showed a three-dimensional pore structure. Micropore filling is the main mechanism of toluene adsorption. DFT analysis revealed that N-6, as an active site, promoted the adsorption capacity through an interaction mechanism. Cheng et al. (2023) also reported the effect of N-doping and activators on improving the pore structure, increasing the SSA (1000 m2/g), and enhancing the toluene-adsorption capacity (223 mg/g), which maintained 80% adsorption capacity under humid conditions. Besides, a DFT study revealed the dependency of N-containing functional groups over the π–π dispersion and hydrophobic and electrostatic interactions, which promote toluene adsorption. Figure 9 shows the selective adsorption mechanism of toluene on the amino-functionalized modified carbon under humid conditions. Xu et al. (2021) reported the N-doped carbon with good performance for the adsorption of toluene (720 mg/g) and high SSA (2060 m2/g). A DFT study revealed that a large number of amino functional groups (3.6%) exhibited highly strong affinity for toluene than that for N-6, N-5, and so on. Biomass-derived N-doped hierarchical porous carbon (NHPC) exhibited a large SSA (2266 m2/g), a high pore volume (1.14 cm3/g), and a particularly high toluene adsorption (5.94 mmol/g), while corncob-derived N-doped carbon with a higher total cellulose content and lower ash content resulted in a better pore activation effect (Huang et al. 2023). Du et al. (2020) reported an N–O co-doped hierarchical carbon material with a high SSA (1650 m2/g) and toluene-adsorption capacity (627 mg/g), where the hierarchical porosity and high SSA were the key factors for physical adsorption. N-containing functional moieties significantly enhanced the chemical adsorption, while N-5 exhibited the highest affinity for toluene molecules. Shi et al. (2022a) reported N–O co-doped porous carbon with a high SSA (2784.53 m2/g), a hierarchical pore structure, high N (16.16%) and O (15.75%) contents, and excellent toluene adsorption (813.6 mg/g). DFT studies revealed that the interaction among the toluene molecules can be improved by N–O functional groups and multilayer adsorption can be achieved, wherein the optimal adsorption pore size of the N–O co-doped porous carbon was 3–7 times the dynamic diameter of toluene. These optimal adsorption pores provide pathways and adsorption sites that allow the highest adsorption capacity of toluene.

Fig. 8
figure 8

Toluene-adsorption property of NPCs (N-doped porous carbon): a Breakthrough curves for toluene, b adsorption isotherms, c comparison of adsorption capacity of similar carbon materials, and d cyclic performance of NPCs (Jin et al. 2022a)

Full size image
Fig. 9
figure 9

Selective adsorption mechanism of toluene on amino-functionalized modified biochar under humid conditions (Cheng et al. 2023)

Full size image

Conclusions

Different strategies have been developed to improve the air-pollutant-removal performance of carbon-based catalysts, such as physical or chemical treatment/activation, metal oxide doping, and heteroatom doping. In physical treatment, CO2 and steam played a major role in controlling the surface properties (such as SSA, pore size distribution, and pore volume), which further enhanced the removal of toxic gases/pollutants (NOx, H2S, Hg0, and toluene). Chemical treatment/activation (with acid or alkaline salts) provides more functional moieties with an improved pore structure and SSA, which enhance the adsorption of porous carbon (particularly, the adsorption capacity of H2S, Hg0, and toluene). In addition, carbonization and calcination temperatures affect the pore structure of carbon-based catalysts, which is important for reactive gas adsorption during catalysis.

Metal oxide doping enhances the surface acidity and redox ability of the catalyst, which in turn enhances the SCR of NOx and poisoning tolerance. Moreover, pre- and/or post-oxidation of the catalyst helps in improving the surface and structural properties of carbon. Co-doping with metal oxides increases the catalytic performance by broadening the reaction window and resisting poisoning. In addition, metal oxide doping with acid/base modification (i.e., oxygen functionalization/surface modification) provides high SO2/H2O tolerance to catalyst poisoning and enhances the catalytic performance. The basicity and synergistic effect of the carbon catalyst also help improve the H2S-adsorption capacity, wherein the calcination temperature and activation play a crucial role. For Hg0 adsorption, metal oxide doping provides more active oxygen and vacancies, better redox strength, and a synergistic effect, which are beneficial for pollutant adsorption and oxidation. The enlarged pore size with increased SSA and active chemisorbed oxygen due to metal oxide doping enhance the toluene-adsorption capacity of the carbon catalyst.

Heteroatom doped (N/S) carbon-based single-atom carbon leads to enhance the high stability, catalytic reaction, and/or electrocatalysis activity. Regulating the electronic structure of N-doped carbon metal catalyst at atomic level effectively enhances the catalytic activity. Heteroatom doping (with N, O, S/P, and B) plays a critical role in enhancing the catalytic properties of porous carbon. In particular, the N-content with unpaired electrons is crucial for enhancing the adsorption and oxidation of NO and NH3 because of the abundant Lewis acid sites that promote the SCR of NO. In addition, the N-containing functional groups (especially N-6) play a significant role in promoting the SCR performance of the carbon catalyst. The synergistic effect of dual heteroatom dopants (N and O) enhances the carbon-catalyzed NH3 SCR of NOx. The basic surface nitrogen sites (N-6 and N-5) also participate in the dissociation of H2S to HS− and H+, thus initiating the selective oxidation of H2S to elemental sulfur. N and S co-doping along with H2O2 hydrothermal treatment improve the SSA and pore volume, which have a significant effect on the H2S-adsorption capacity of the carbon catalyst with a high N-5 content. P- and S-doped carbon materials enhance the SSA and pore volume, leading to high Hg0 adsorption. However, this is not the case for N- and S-co-doped carbon. In addition, the activator significantly enhances toluene adsorption with a modified carbon catalyst.

After exploring the development of carbon catalysts with various modifications, the following conclusions can be drawn:

  1. 1.

    Development of new precursor materials can be explored for producing carbon catalysts with enhanced gas-pollutant (multipollutant)-removal efficiency, thereby strengthening the anti-poisoning and mechanical properties through commercialization.

  2. 2.

    As the environmental concerns are attracting increasing attention, new activation methods that use renewable energy sources or develop closed-loop systems for carbon recovery and reuse are being explored.

  3. 3.

    Designing heteroatom-doped carbon catalysts with tailored properties and functionalities can help optimize their performances for specific applications. This may improve our understanding of the structure–property relationships of these catalysts. Further research efforts are required to optimize their performance and reduce costs for widespread commercial adoption of such catalysts.

  4. 4.

    In-depth understanding of the mechanisms involved in the catalytic activity of carbon materials is required.

  5. 5.

    A recovery method for used carbon catalysts is needed to conserve the valuable resources.