Biochar derived from traditional Chinese medicine residues: An efficient adsorbent for heavy metal Pb(II)

3.2.1 SEM observations

Fig. 3 and Fig. 4 showed SEM images of the TCMR biomass feedstocks and biochars, respectively. There were distinct morphological differences among the feedstocks and biochars. In comparison with the TCMR feedstocks, the biochars contained obvious disordered pores, which might have been produced by the release of oxygenated volatiles from the biomass during pyrolysis (Jiang et al., 2022). In the case of B_DNS, the pores were more dispersed and some of the structure collapsed after pyrolysis, but the whole framework still had a good open texture. The raw material of CX was hard with a comparatively high density and fewer pores, but the overall appearance of the biochar after pyrolysis was loose and porous. Structures such as the trapezoidal and reticulated channels and cork cells in DG and the xylem channels in DGS remained intact before and after thermal decomposition. At HQ, numerous powdery particles were adhered to the feedstock. After pyrolysis these particles disappeared, which exposed the pore structure on the surface. The biochars obtained from the five types of residues exhibited different surface morphologies, including both lamellar and porous structures. The unique channel-like structures of the residues themselves increased the surface area of the biochars (Ma et al., 2018), which was conducive to the adsorption of heavy metals.

Close

Fig. 3

SEM images of the TCMR biomasses.

Export to PPT

Close

Fig. 4

SEM images of the TCMR biochars.

Export to PPT

Overall, the resulting biochars contained more loose pores than the residue feedstocks. Moreover, certain plant-specific structures such as cork cells and reticulated channels were preserved to some extent. This suggested that the pore structure of the feedstock can be retained in the thermally treated material, as previous studies have also indicated (Komkiene & Baltrenaite, 2015).

3.2.2 Structural features of biochars

In this study, all of the medicine residue biochars had a small specific surface area (0.83–1.60 m²/g) (Table 1). This was similar to the experimental results reported in the literature, where pristine biochar had a small specific surface area (Huang et al., 2023). Interestingly, the research found that the specific surface area of the different biochars varied particularly widely (0.82 to 420.30 m²/g and 0.65 to 642 m²/g, respectively) (Li et al., 2017; Liang et al., 2021). Furthermore, there are large differences in specific surface area between biochars produced from the same feedstock. For example, the specific surface area of rice straw biochar produced at the same pyrolysis temperature (500℃) differs by almost sixty times between the two papers. The smaller one is only 0.84 m²/g (Mehmood et al., 2022) and the larger one is 50.11 m²/g (Fan et al., 2018). In addition, the specific surface area of biochar is also significantly affected by pyrolysis temperature. Biochar produced from the same wood mixture had a specific surface area of 3.54 m²/g at 400℃, but when the pyrolysis temperature was increased to 700℃, the specific surface area was up to 220.41 m²/g (Kalina et al., 2022). Meanwhile, literature reported that the adsorption performance of activated carbon (SSA = 1409 m²/g) for Cu, Co and Pb was much lower than that of ten pristine biochar (SSA in the range of 1.05–5.41 m²/g) (Huang et al., 2023). Regmi reported that commercially available powdered activated carbon (PAC) had very little affinity for copper and cadmium, even though its specific surface area was more than two orders of magnitude larger than that of switchgrass biochar (Regmi et al., 2012). This demonstrated that the heavy metal adsorption performance of a material was influenced by a number of factors, not just specific surface area. In this study, B_DNS and B_CX had relatively great advantages in terms of specific surface area (1.55–1.60 m²/g) and had smaller average pore sizes (11.71–15.79 nm). If only the effect factor of specific surface area was considered, a large specific surface area and multiple pores can facilitate the physical pore filling of heavy metals in the removal process (Li et al., 2022). Therefore, B_DNS and B_CX have a greater potential to play a role if there is pore filling in the adsorption mechanism.

According to the International Union of Pure and Applied Chemistry classification, the adsorption/desorption isotherms for the five biochars conformed to the type III isotherm (Fig. S2), which proved the biochars were nonporous or macroporous materials. Based on the table of particle size distribution (Table S1), the prepared materials were mixtures of mesopores (2–50 nm) and macropores (>50 nm) materials. The isotherms were concave and had no inflection points. The amount of gas adsorbed increased with an increase in the partial pressures of the components. The isotherms also exhibited H₃-type (mesoporous) hysteresis loops with no obvious saturation plateau, which indicated that the biochars had uneven pore size distributions. In addition, the H₃-type hysteresis loops showed that the narrow pores created different pathways for the adsorption and desorption of N₂, which was consistent with the SEM images of the biochars (Mukome et al., 2013).

3.2.3 FT-IR spectroscopic analysis

The FT-IR spectra of surface functional groups in the five biochars were shown in Fig. 5, which indicated the presence of oxygen-, nitrogen-, and carbon-containing functional groups with peaks in the range of 400–4000 cm⁻¹. It could be seen that the types of surface functional groups in the biochars obtained from the five residues were mostly similar, with minor variations in band positions and wavenumbers and slight differences in the sharpness and intensity of peaks. Among the common peaks, the peak at 1575 cm⁻¹ was attributed to COOH groups in carboxylic acids, the peak at 880 cm⁻¹ was attributed to C–H or N–H out-of-plane bending, and the peak at 800 cm⁻¹ was attributed to C–H bending in aromatic compounds (Hossain et al., 2011; Lu et al., 2013). In addition, different biochar materials gave rise to specific peaks. For example, the presence of alcoholic and phenolic hydroxyl groups in the biochars was indicated by the peak at 3400 cm⁻¹, which corresponded to the vibrations of OH groups in water molecules, alcohols, carboxylic acids, or metal hydroxides (Jin et al., 2016). This peak was most prominent in the spectrum of B_CX, whereas it was barely visible in those of B_DNS and B_DGS. The peak at 3040 cm⁻¹ in the spectrum of B_CX was attributed to alkyl-CH₂ vibrations, and in the spectrum of B_DNS the peaks at 2920–2855 cm⁻¹ were attributed to C–H stretching vibrations in CH₃ and CH₂ groups (Mukome et al., 2013). In the spectra of B_DNS and B_CX the peak at 1340 cm⁻¹ was attributed to C-O/C-N stretching vibrations, and the peak near 1120 cm⁻¹ was attributed to C-O stretching vibrations (Liu et al., 2017).

Close

Fig. 5

The FT-IR spectrums of the TCMR biochars.

Export to PPT

The results of FT-IR spectroscopy showed that the biochars obtained from the different plant residues had different types of functional groups on their surface. Combined with the results of the Boehm experiment (Table S2), it was found that the total amount of alkaline functional groups on the surface of biochar was higher than the total amount of acidic functional groups, which was a good explanation for the alkaline pH (Marzeddu et al., 2022). Meanwhile, the distribution of B_DNS functional groups showed relatively higher concentrations of carboxyl (0.15 mmol/g) and lactone groups (0.14 mmol/g) compared to the other biochars, which was consistent with the infrared spectra. In general, the presence of functional groups on the surface of biochar can provide an efficient means of removing heavy metals such as copper and cadmium through adsorption, chelation and precipitation reactions (Regmi et al., 2012). However, the specific removal effect is also influenced by factors such as the pore state of the biochar and the type and content of surface functional groups. Further research is therefore required for practical applications. This demonstrated the potential adsorption capacity of these materials with respect to heavy metals and laid the foundation for further studies.

3.2.4 Phase transformation analysis

The thermal profiles of the dregs obtained from the simultaneous TGA-DSC analyser are shown in Fig. 6. The figure showed that the total mass loss remained around 70 % for all the TCMRs at 1000℃. Looking at the TGA-DTG curves, the mass loss (%) observed around 100℃ could initially be attributed to the loss of moisture on the surface of the TCMRs (Zhao et al., 2021). In addition, two distinct weight loss regions were found to exist in the ranges 210–276 °C and 331–345 °C respectively. The mass loss of the former peak was attributed to thermal degradation of hemicellulose, whereas the latter peak was associated with thermal degradation of cellulose, with an exothermic peak of cellulose located between 311 and 351 °C (Cappello et al., 2023; Zhang et al., 2020). This was due to the fact that the rigid structure of cellulose had a higher thermal stability than that of hemicellulose (Gonzalez-Canche et al., 2021). The DSC curves showed that the thermal equilibrium state was almost reached at the experimental condition of 500 °C, which was a more stable assessment of the state of the material.

Close

Fig. 6

TG-DTG and DSC curves of the TCMRs. (a, b) DNS; (c, d) CX; (e, f) DG; (g, h) DGS; (i, j) HQ.

Export to PPT

Investigating the thermal stability of biochar is essential to achieve natural ageing resistance of the material (Zheng et al., 2023). In this experiment, TGA curves were used to determine the trend of physicochemical properties of BC, mainly describing the weight loss versus temperature. Fig. S3 and Table S3 showed the TGA curves and weight loss of the TCMR biochars. The total weight loss was 18.22 % (B_DNS), 23.96 % (B_CX), 21.15 % (B_DG), 19.28 % (B_DGS) and 21.95 % (B_HQ), respectively. Among them, B_DNS and B_DGS showed a lower loss of mass, which indicated that these two biochars had a better thermal stability. From the thermal decomposition process, the thermal stabilisation stages of the four biochars, except B_DGS, were all three stages, presumably mainly due to the loss of water molecules (stage 1, around 100℃), thermal degradation of lignin (stage 2, around 600℃) and decomposition of other inorganic components (stage 3, around 900℃) (Gonzalez-Canche et al., 2021; Zhou et al., 2021a, 2021b, 2021c). For B_DGS, degradation of cellulose or hemicellulose was assumed to be responsible for the change at 322 ℃. It has been reported in the literature that hemicellulose with an amorphous structure begins to degrade at 220℃, whereas cellulose, due to its highly stable crystalline structure, can be maintained at 300℃, followed by rapid weight loss at 315–400℃ (Dhyani & Bhaskar, 2018). The absence of this stage in other biochars may be due to the fact that the hemicellulose is essentially completely pyrolysed and the cellulose partially pyrolysed at the preparation temperature of the biochar (500℃), which can be verified in the XRD section.

3.2.5 XRD patterns and TEM images of biochars

The structure and phase composition of BCs were shown in the XRD pattern (Fig. 7(a)). A long range disordered peak appeared on the XRD of all biochars in the 2θ range of 20°–30°, demonstrating the presence of amorphous structures on the material (Zhang et al., 2018). The XRD results showed that all biochars were amorphous structures with crystal structures (Das et al., 2021a). Common peaks belonging to graphite microcrystals were present on all biochars, mainly corresponding to several peaks at 2θ = 26.554°, 42.374°, 50.654° and 59.907° (Shao et al., 2021). The graphite microcrystalline peaks on B_DG and B_HQ were more numerous and sharper, indicating a higher degree of graphitisation than the other three biochars. The TEM and HRTEM results of the five biochars were shown in Fig. 8. B_DNS and B_DGS had smooth surfaces and contained some graphitic carbon layer structures with no microporous or mesoporous structures observed. In contrast, B_CX and B_DG had rough surfaces with obvious pore structures. B_HQ showed an outer layer with fine pores and a lamellar structure. The B_DNS and B_DGS surfaces were analysed by TEM and found to have an amorphous carbon layer distribution with some graphitic carbon layers. The graphitic structure results from the graphitisation and aromatisation of cellulose-type components in biomass (Hu et al., 2020). Among them, aromatisation facilitates the formation of cation-π interactions, which can improve the adsorption capacity of the materials for heavy metal cations (Zhou et al., 2022). From the above, it can be speculated that B_DNS and B_DGS have better adsorption potential for heavy metals. Meanwhile, the presence of lattice striations could be detected in the HRTEM images of all biochars. Combined with the analysis of the XRD results, this helped to verify the presence of crystalline carbon.

Close

Fig. 7

(a) XRD patterns and (b) Raman spectra of the TCMR biochars.

Export to PPT

Close

Fig. 8

TEM and HRTEM images of the TCMR biochars.

Export to PPT

In addition, a large number of crystalline cellulose peaks were found on all biochars, corresponding to several peaks at 2θ = 20.639°, 27.593°, 29.454° and 36.649°, suggested that the cellulose in the TCMRs reacted to form crystals under pyrolysis conditions (Zhou et al., 2021a, 2021b, 2021c). This is consistent with literature reports that cellulose conversion of biomass occurs during pyrolysis, resulting in carbon-rich amorphous biochar formed (Biswas et al., 2018; Kaur et al., 2023). This was also consistent with the results of the thermal stability experiments, where the hemicellulose was completely pyrolysed and cellulose was detected. In summary, it was speculated that the carbon material obtained in this experiment was in a structure where amorphous carbon and graphitised carbon coexisted.

3.2.6 Raman spectrum in biochars

The graphitic structure was also confirmed by Raman spectroscopy (Fig. 7(b)). It was possible to distinguish two main peaks detected at approximately 1360 and 1588 cm⁻¹ (Table S4), which implied the point defect degree (D-band) and the graphitic structures (G-band). The characteristic D-band can be attributed to the presence of disorder in the graphene structure, whereas the G-band is due to the stretching band of the sp²-hybridized carbon (Zhou et al., 2023). The Raman signals for lignin and cellulose have been reported in the literature to be between 1000 and 1700 cm⁻¹ (Agarwal & Ralph, 1997). This was consistent with the characteristic peaks of cellulose in XRD. The intensity ratio between I_D and I_G reflects the degree of graphitisation. If this ratio is < 1, it indicates a high degree of graphitisation; if the ratio is > 1, it indicates the presence of a large number of functional groups on its surface (Marzeddu et al., 2022). Comparison of the magnitude of the I_D/I_G values showed that the values of B_DG (0.97) and B_HQ (0.99) were < 1, while the values of the other three biochars were > 1. This indicated that B_DG and B_HQ had a greater degree of graphitisation. This was consistent with the XRD results that more graphite microcrystals were present in B_DG and B_HQ. Meanwhile, more functional groups were present on the surfaces of B_DNS, B_CX and B_DGS, which could be favorable for adsorption of heavy metals.

As shown in Fig. 2, the correlation analysis of the I_D/I_G value with the physicochemical properties of biochar was investigated. It was found that this value does not correspond to a reliable correlation with any of the properties. This was consistent with literature reported that for biochar only, the peak or microcrystalline sizes obtained from different preparation processes did not correlate reliably with the O/C, H/C or C–C sp² of the material in XPS (Gabhi et al., 2020; Ilic et al., 2022).

3.2.7 Structural elements in biochars

As shown in Table 2, the biochars obtained from the five different medicine residues varied considerably in terms of ash content. The ash content of B_DG was as high as 28.34 %, whereas the ash contents of B_DNS, B_DGS, and B_HQ did not vary much (8.54 %–9.91 %). The literature suggests that a higher ash content in the feedstock leads to a higher yield of biochar (Xiao et al., 2022). However, in this study a weak correlation (r = -0.01, p < 0.05) was observed between the ash content of the feedstock and the yield of biochar (Fig. 2), which could be attributed to the characteristics of the raw materials.

Elemental analysis showed that the carbon contents of the five biochars were mostly around 70 %, with B_DG having the lowest carbon content (56.84 %). The contents of both lignin (r = 0.63, p < 0.05) and cellulose (r = 0.84, p < 0.05) were positively correlated with the oxygen content, which also directly confirmed the positive proportional relationship between the contents of biochar components and the content of oxygen-containing functional groups. The stability of the biochars could be determined from the O/C and H/C ratios. The former ratio for all five biochars was less than 0.2, which indicated high stability with half-lives of greater than 1000 years (Ippolito et al., 2020). The H/C atomic ratio was maintained at around 0.5, and for all five biochars it was below the upper limit of 0.7. This indicated a larger, denser aromatic ring structure and that thermochemical transformations were occurring, which further indicated the stability of the materials (IBI 2015). The lower is the H/C ratio, the higher is the aromaticity (Gao et al., 2019). The data indicated that B_HQ and B_DGS had slightly higher aromaticity than the other biochars. Furthermore, it is worth noting that the European Biochar Foundation (EBC) recommends biochar with an H/C ratio of less than 0.6 and an O/C ratio of less than 0.4 for effective carbon sequestration when applied to soil environments (EBC 2012). The results of the thermal stability of the biochar (weight loss ranging from 18.22 % to 23.96 %) are in agreement with the elemental analyses and O/C ratios related to the stability of the biochar, which together demonstrate the stability of the prepared biochar (Fig. S3). As a result, the TCMR biochars have high stability in environmental treatment processes.

3.2.8 Mineral elements in biochars

Tables 3 and 4 showed the contents of four common mineral elements and a heavy metal in the TCMR feedstocks and biochars, respectively. A significant increase in the content of K by a factor of 2.29–3.17 times was observed in the biochars obtained by high-temperature preparation in comparison with the corresponding feedstocks. Moreover, the K content changed from 7.04 to 13.64 g/kg in the raw materials to 20.10–37.21 g/kg in the biochars, and the Na content changed from 0.84 to 1.57 g/kg in the raw materials to 1.11–2.81 g/kg in the biochars. This fully demonstrated that enrichment of mineral elements occurred when the TCMRs were converted into the biochars. The biochars exhibited higher contents of mineral elements than the raw materials. This indicated that a relative enrichment of each element occurred during pyrolysis, which was in accordance with previous findings in the literature (Xiao et al., 2022). Mineral constituents of biochar, such as potassium, calcium, and magnesium, have been reported in the literature to be beneficial for the adsorption of heavy metals in aqueous solutions (Duwiejuah et al., 2020). These mineral components and the heavy metals could be subjected to an ion exchange process, which would increase the binding affinity of the heavy metals on the surface of the biochar and greatly reduce their mobility (Lin et al., 2023a, 2023b; Mazarji et al., 2023), thus achieving the purpose of removing heavy metals.

In addition, the biochars in this study had high pH values, which was presumably due to the fact that most of the mineral elements were retained in the biochars after pyrolysis (Shi et al., 2019). In terms of mineral composition, the TCMRs and their biochars were richest in K (20.10–37.21 g/kg), followed by Mg (4.32–12.42 g/kg) (Table 4). With regard to individual residues before and after pyrolysis, DNS consistently had high levels of K and Mg, whereas DG had the highest Ca content. The Na contents of both DG and HQ were significantly different from those of the rest of the materials pre- and post-pyrolysis. Obviously, probably as a result of variations in the natural composition of the feedstocks, the mineral content of the biochars varied from one source to another (Mehmood et al., 2022; Shen et al., 2015).

Regarding the determination of heavy metal contents, there was little difference in the content of Pb(II) before and after pyrolysis, which proved that no enrichment of heavy metals occurred during pyrolysis of the herbal residues themselves and that there was no secondary pollution of the natural environment with heavy metals (Wu et al., 2016). Although the high K content of the biochars implies that they could act as part of a K fertilizer, potential damage to some K-sensitive plants needs to be considered.

3.3.1 Pb(II) adsorption experiment

Fig. 9 showed the adsorption kinetic curves of five TCMR biochars and their fitted models. The whole adsorption process was divided into two steps: rapid adsorption and equilibrium adsorption. It is clear that all five biochars reached the equilibrium concentration at the experimental setting time (t = 720 min). By analysing the non-linear fit parameters of the pseudo-first-order and pseudo-second-order models (Table 5), the correlation coefficients of the pseudo-second-order models for all biochars with R² > 0.99 were higher than those of the pseudo-first-order models, indicating that the pseudo-second-order models were more consistent with the adsorption process of the original biochars. This suggested that chemisorption is the main adsorption process (Lin et al., 2023a, 2023b). The Elovich model, on the other hand, had a correlation coefficient R² > 0.95, indicating that the adsorption process included inhomogeneous phase diffusion (Zhou et al., 2022). The adsorption isotherms of biochars (Fig. 10) showed that the Langmuir correlation coefficient R² > 0.99, indicating that it was more consistent to be used to describe the adsorption behaviour of biochars. It indicated that the adsorption of biochars belonged to a monolayer adsorption process, where each active site on the adsorption surface had the same adsorption energy (Manoharan et al., 2022). The adsorption capacities of these five biochars were calculated by the Langmuir isotherm model and were 36.42 mg/g (B_DNS), 7.84 mg/g (B_CX), 9.00 mg/g (B_DG), 18.67 mg/g (B_DGS), and 7.45 mg/g (B_HQ), respectively (Table 6). Although all the prepared biochars were able to remove Pb(II) from the solution, their adsorption capacities varied and depended on their individual properties. This was in accordance with been reported in the literature, which showed that different biochars obtained from different TCMR sources exhibited differences in sorption performance for heavy metals (Das et al., 2021a; Ding et al., 2017). The correlation between the amount of adsorption and the physicochemical properties of biochars was analysed (Fig. 2). It was found that the amount of adsorption (Qm) of biochar was strongly and positively correlated with the cellulose (r = 0.82, p < 0.05) and lignin (r = 0.87, p < 0.05) of the feedstock, as well as the oxygen content (r = 0.90, p < 0.05). However, adsorption showed a weak positive correlation with the pH_pzc (r = 0.71) and EC values (r = 0.73) of the materials. The differential contribution of the physicochemical properties of the materials to the effect of their adsorption performance was demonstrated. Among them, the adsorption properties of B_DNS and B_DGS were higher than those of several other biochars. This may be attributed to the high cellulose and lignin content and the associated specific physico-chemical properties, including conductivity, isoelectric point and other factors favourable to enhanced Pb(II) adsorption (Alcaniz-Monge et al., 2022; Li et al., 2020). In order to compare the sorption performance of the studied biochars, a review of literature was done about the biochars in other studies used for adsorping Pb(II) (Table S5). In contrast, B_DNS and B_DGS had relatively high adsorption capacities, and the other three had less satisfactory adsorption performance. In addition, further modification of the original biochar was required to enhance its adsorption performance and make it more effective in removing heavy metals from the environment. This is a challenge that the research group is currently addressing.

Close

Fig. 9

Adsorption kinetic fitting curves of biochars. (A) Danshen (B) Chuanxiong (C) Danggui (D) Dangshen (E) Huangqi.

Export to PPT

Close

Fig. 10

Adsorption isotherm fitting curves of biochars. (A) Danshen (B) Chuanxiong (C) Danggui (D) Dangshen (E) Huangqi.

Export to PPT

3.3.2 Pb(II) adsorption mechanism

In order to further elucidate the mechanism of the adsorption of Pb(II) by pristine biochar, experiments were carried out on B_DNS, and three types of characterization, namely, SEM-EDS, FT-IR spectroscopy, and XPS, were performed before and after adsorption (Fig. 11 and Fig. 12). As shown in Fig. 11, uniformly distributed Pb(II) was visible on the surface of the biochar after adsorption, which demonstrated the successful adsorption of Pb(II). The semi-quantitative results using EDS show that Pb(II) increased by 5 % from the original 0 % (Fig. S4). Moreover, in comparison with the biochar before adsorption, many small particles appeared on the surface of the biochar after adsorption, which were presumed to consist of Pb(II)-related compounds that were generated.

Close

Fig. 11

The SEM-EDS images of biochars before and after adsorption.

Export to PPT

Close

Fig. 12

The spectra of (a) XPS; (b) C 1 s; (c) O 1 s and (d) FT-IR including biochars before and after adsorption.

Export to PPT

Fig. 12(a–c) showed the XPS spectra of the biochar before and after the adsorption of Pb(II), which contained similar characteristic C 1 s and O 1 s peaks. From the semi-quantitative results, it was found that there was a clear silicon-related peak after adsorption, and the content of elemental silicon decreased from 2.02 % to 1.00 %, which demonstrated the involvement of silicon in the adsorption process. The literature reports that silicates can react with lead to form silicate precipitates (Li et al., 2019b, 2019a). The adsorption mechanism was further suggested by the high-resolution XPS spectra of C 1 s and O 1 s. The C 1 s peaks located at 284.8 eV (60.37 %), 285.7 eV (33.20 %) and 289.4 eV (6.43 %) were attributed to C–C/C–H/C = C, C-O/C-N, and C(=O)–OH groups, respectively (Das et al., 2023). After the adsorption of Pb(II), the percentage area of the C–C/C–H/C = C peak decreased from 60.37 % to 38.97 %, whereas the percentage area of C-O/C-N and C(=O)–OH peaks increased to 42.24 % and 18.78 % respectively. This may have been due to complexation or precipitation with Pb(II) resulted in the loss or gain of C in different valence states. Similarly, in the fine spectrum of O 1 s, the peaks located at 531.5 eV (30.97 %), 532.6 eV (35.55 %), and 533.7 eV (33.48 %) were attributed to C-OH, C = O, and O = C-OH(R) groups, respectively (Das et al., 2023; Herath et al., 2021). In the case of Pb(II)-loaded biochar, the percentage of peak area of C-OH increased from 30.97 % to 35.22 % and the percentage of C = O and O = C-OH(R) decreased to 32.65 % and 32.13 %, respectively. This may have been due to interactions between the groups and Pb(II). According to the various XPS peaks, different functional groups may have different adsorption capacities for heavy metals, with hydroxyl and carboxyl groups exhibiting higher adsorption capacities for Pb(II) (Yu et al., 2021a, 2021b).

According to the results of FT-IR mapping (Fig. 12(d)), after adsorption the two peaks at 1260 cm⁻¹ and 1575 cm⁻¹ appeared to be significantly weaker. These were attributed to C-O-C and COOH stretching vibrations, respectively (Li et al., 2019b, 2019a). The peak at 3440 cm⁻¹, which was attributed to O–H vibrations, appeared after the adsorption of Pb(II) by the biochar. This could be attributed to the introduction of Pb(II), which changed the functional groups on the biochar (Zhou et al., 2022). These changes suggested that oxygen-containing functional groups were utilized by Pb(II) and played an important role in the adsorption process. Moreover, abundant mineral elements (K, Ca, Na, and Mg) on the surface of the biochar could be exchanged with metal ions (Lin et al., 2023a, 2023b). The pH_pzc of B_DNS was higher than the initial pH of the solution and the adsorption capacity should be lower due to electrostatic attraction, whereas the adsorption capacity of B_DNS was better, indicating that electrostatic attraction was not the main adsorption mechanism of B_DNS (Zhou et al., 2018a, 2018b).

In conclusion, the adsorption mechanism mainly involved precipitation, complexation of oxygen-containing functional groups, and ion exchange to complete the process.