1 Introduction

Tetracycline (TTC) is the most widely utilized antibiotic in China and the second most frequently employed antimicrobial globally (Balakrishnan et al. 2023), (Mohammadi et al. 2023). In recent years, tetracycline has been detected in various environments, including water, soil, and even plants (Zhang et al. 2021a), (Zhu et al. 2021), (Chang et al. 2023), (Lenart-Boroń et al. 2022). This exacerbates concerns regarding public health, as the indiscriminate use of antibiotics is inducing modifications in bacteria and their genetic material, thereby fostering mutations and increased resistance (Leichtweis et al. 2022). Strategies for tetracycline removal include membrane separation (Liu et al. 2019), adsorption, (Minale et al. 2020; Zhang et al. 2023a) advanced oxidation technologies, (Hou et al. 2022), (Zhang et al. 2023b) etc. Adsorption is widely favored due to its low energy consumption, ease of operation, and environmental friendliness (Liang et al. 2022). Nevertheless, conventional adsorbents such as carbon nanotubes (Li et al. 2023), graphene (Zhao et al. 2023), and activated carbon16 (Vinayagam et al. 2023) are constrained by intricate synthesis and relatively elevated costs. Biochar (BC) is a high-value adsorbent generated via biomass pyrolysis under oxygen-limited conditions, and it exhibits a novel negative carbon emission level (Xu et al. 2020; Yi et al. 2020). The biomass for biochar production originates from solid wastes generated at various sources, including agriculture, aquaculture, and industry, and it is abundant and cost-effective (Liang et al. 2021). Additionally, the porous structure and abundant functional groups of biochar provide exceptional adsorption capacities, suggesting its potential as a promising adsorbent (Yang et al. 2021b).

Pristine biochar demonstrates limited capacity for tetracycline adsorption, while modifications significantly enhanced the absorption of tetracycline. Modifications influence the structural characteristics of biochar, such as the surface area, porosity, aromaticity, surface functional groups, and mineral content (Qiu et al. 2021). For instance, acid‒base modifications alter the surface characteristics of biochar, expanding specific regions and pore structures, thereby exerting a significant influence on physical adsorption of pollutants (Cheng et al. 2021).Organic modifications involve the conjugation of biochar with organic compounds containing a multitude of functional groups, thereby increasing the number of adsorption sites (Qiu et al. 2021). Iron modification facilitates the formation of the carbon structure in biochar, concurrently increasing the specific surface area and functional group abundance (Karunaratne et al. 2022). Therefore, modifications of diverse structures in biochar have varying impacts on tetracycline adsorption. Mei et al. (Mei et al. 2021) utilized Fe–N doping to modify biochar at 700 °C. By manipulating the surface area, functional groups, and graphite carbon structure, the maximum adsorption capacity for tetracycline exhibited a 5.4-fold enhancement relative to pristine biochar. Yu et al. (2020) modified grapefruit peel extract and enriched the surface functional groups of grapefruit peel-based biochar prepared at 600 °C. The maximum capacity for tetracycline adsorption was raised to 2.3 times that of the original material. Mu et al. (Mu et al. 2021). formed mesopores in tea biochar by utilizing KOH at 700 °C and concurrently enriched the surface functional groups. However, the maximum tetracycline adsorption capacity of the modified biochar was only 1.827 times that of the unmodified material. Therefore, identifying the key structures that regulate the tetracycline adsorption capacity of biochar and selectively modifying the biochar structure for a specific purpose contribute to the development of more cost-effective and efficient biochar adsorption materials.

However, there has been no prior identification of a correlation between biochar structure and tetracycline adsorption in previous studies. This limitation arises from the inherent heterogeneity in biochar structures, which is influenced by both the feedstock and carbonization degrees (Keiluweit et al. 2010). Therefore, it is important to develop standardized construction of biochar. Xiao and Chen proposed that biochar has a quaternary structure, and despite substantial discrepancies in the composition and characteristics of biochars, their fundamental architectures remain analogous (Xiao and Chen 2017). This provides the possibility of constructing a microscopic structural model for tetracycline adsorption by biochar. Research indicates that the mechanisms for tetracycline adsorption onto biochar involve chemical reactions and intermolecular forces, including electrostatic interactions, hydrogen bonding, π-π interactions, and van der Waals forces (Ngigi et al. 2019; Yue et al. 2019; Zhang et al. 2021b). However, no studies have identified the specific interaction sites, so there is no mechanistic rationale for these interactions.

At a pyrolysis temperature of 300 °C, the biomass exhibits a low pore volume and low crystallinity, with subtle variations indicative of biochar with a low degree of carbonization (Leng et al. 2021). With increases in the pyrolysis temperature, amorphous carbon is transformed into crystalline carbon through condensation, leading to increases in the specific surface area and porosity, elevated ash content, and the formation of a “molten ring two-dimensional structure” characterized by abundant pores (Ahmad Farid and Andou 2022). At 700 °C, the disordered layered structure of biochar transitions to graphite (Chen et al. 2017). Biochar maintained at 700 °C exhibited a graphitic microcrystalline structure. Due to the widespread adoption of straw biochar for investigating the structural properties of biochar, this study focused on the adsorption of tetracycline by biochar prepared within the pyrolysis  range of 300 to 700 °C (Zhang et al. 2023c), (Fu et al. 2021), (Liao et al. 2023). The primary focuses were (1) identifying the biochar structures influencing tetracycline adsorption and determining the key influencing factors; (2)  establishing the relationships among the key factors influencing biochar adsorption and constructing a standardized model for biochar and (3)  investigating the micromechanisms of tetracycline adsorption by biochar and unveiling the adsorption sites. It is anticipated that this study will offer valuable insights advancing the development of more efficient and cost-effective biochar materials.

2 Materials and methods

2.1 Chemicals

More information can be seen in Additional file 1: Text S1.

2.2 Preparation and characterization of rice straw biochars

Rice straw biochar was produced in a vacuum tube furnace (GSL-1600, China)  by slow  pyrolysis method (Hoslett et al. 2021; Fan et al. 2023; Abumelha et al. 2023; He et al. 2019). Briefly, the collected rice straw was washed three times and dried in a hot air oven at 80 °C until reaching a constant weight. Then, the dried biomass was ground and sieved through a 120-mesh sieve. Next, the powdered biomass was placed in a tube furnace and heated at a rate of 5 °C min−1 and carbonized for 2 h under continuous nitrogen flow to avoid oxidation. The heat treatment temperatures were 300 °C, 400 °C, 500 °C, 600 °C, and 700 °C; the products obtained with these different heating treatment temperatures were labeled BC300, BC400, BC500, BC600, and BC700, respectively.

The surface morphologies of the biochar were examined with scanning electron microscopy (SEM; Hitachi S-4800, Japan). The Brunauer–Emmett–Teller specific surface areas and total pore volumes of the biochar were evaluated with N2 adsorption at 77 K and a surface area analyzer (Quantachrome, Autosorb-iQ-Station 3, USA). The elemental contents of different biochars were determined with an elemental analyzer (Elementar: Vario EL cube, Germany). The ash contents were determined by heating the biochars at 800 °C for 4 h in a muffle furnace with ambient atmosphere (Thermo BF51866KC-1, Germany). The percentage of oxygen was determined by subtraction. Solid-state 13C CP/MAS NMR spectroscopy was performed with a Bruker Avance 400 MHz NMR spectrometer (Karlsruhe, Germany) operating at 100 MHz.

Quantification of functional groups was performed with the Boehm titration method. Further details can be found in Additional file 1: Text S2.

2.3 Batch adsorption experiment

The kinetics and isotherms for tetracycline adsorption on the biochars were investigated. The adsorption experiments were carried out in 50 mL brown plastic centrifuge tubes containing 25 mL of tetracycline aqueous solution and 0.01 g biochar. The pH was adjusted to 7 with 0.01 M hydrochloric acid or sodium hydroxide, since this is the optimal pH for tetracycline adsorption by biochar (Zhang et al., 2023c), (Hoslett et al., 2021a). At elevated or reduced pH conditions, electrostatic repulsion diminishes the adsorption performance of biochar (Fan et al., 2023). The pHpzc of biochar and the pH variations before and after tetracycline adsorption are presented in Additional file 1: Table S1. Before adsorption by biochar, tetracycline existed in two states, namely, TCH2+ and TCH− (pKa2 = 7.68, pKa3 = 9.68) (Abumelha et al. 2023). Thus, pH 7 falls within the optimal range to minimize interference from electrostatic repulsion. An initial concentration of 10–120 mg L−1 tetracycline was prepared with ultrapure water for the adsorption isotherm experiments. All experiments involved continuous shaking at 200 rpm at a constant temperature of 25 °C for 72 h. Two milliliters of the supernatant was removed and filtered through a 0.22 μm filter membrane (Tianjin Jinteng, China) for further analyses. Tetracycline concentrations of 40 and 80 mg L−1were used for the kinetics experiments, with aliquots taken at specific time intervals. All batch experiments were conducted in triplicate. The concentrations of tetracycline in the samples were quantified with high-performance liquid chromatography (Agilent, 1260 Infinity II, USA) with a tunable UV (TUV) detector set to a wavelength of 350 nm. Competitive adsorption experiments and detection details are described in Additional file 1: Text S3.

2.4 Fitting models

Four adsorption kinetic models, namely, the pseudofirst-order (PFO), pseudosecond-order (PSO), Elovich, and liquid film diffusion (LFD) models, were used to fit the kinetic data. The details of the model equations are presented in Additional file 1: Text S4. The isotherms for tetracycline adsorption on the biochar showed the best fit with the Dubinin-Ashtahkov model (DAM) (He et al. 2019; Jang and Kan 2019; Ma et al. 2021) . The DA model has been used with carbonaceous materials in many cases (Bai et al. 2021; Shen et al. 2015; Yang et al. 2016). The EDA (kJ mol−1) is the characteristic adsorption energy, which includes all forces for direct interactions between the molecule and the surface (Crittenden et al. 1999). More details are shown in Additional file 1: Text S5.

2.5 Density functional theory computational simulations

The 13C NMR spectra of the biochar produced at 300–700 °C were analyzed with MestReNova. The aromaticity (Far) was obtained by integrating the peaks found in the 13C NMR spectra at 95 to 165 ppm. The degree of aromatic condensation (χb) was described by the sizes of the aromatic structures, and it was calculated from the ratio of the fraction of aromatic bridgehead carbons relative to all aromatic carbons (Yang et al. 2021a; McBeath et al. 2011).

$$\chi ^{{\text{b}}} = F_{b} /F_{{ar}}$$
(1)

where Fb is the fraction of aromatic bridgehead carbons relative to total carbons, obtained by subtracting the integration results for aromatic C−O groups (145–165 ppm). The mean weighted square error (MWSE) was calculated to examine the goodness of fit with the model. More details can be found in Additional file 1: Text S6. Fits and calculations were determined with SPSS 23 and Origin v8.0 software.

Density functional theory (DFT) calculations were performed with Gaussian16 and the Vienna ab initio simulation package. The adsorption energies were calculated with projector augmented wave potentials and the generalized gradient approximation (GGA) with the Perdew-Burke-Ernzerhof (PBE) functional (Kresse and Furthmüller 1996; Kresse and Joubert 1999; Perdew et al. 1996). The van der Waals interactions were described by the DFT-D2 method in this work (Grimme et al. 2010). The binding energies (Eab) were calculated with Eq. 2. For geometry optimization and frequency calculations, 6–311 + G(d) was used as the basis set. The PBE0 functional was adopted with the D3 (BJ) empirical dispersion correction. The solvent effect of water was examined with the self-consistent reaction field (SCRF) method based on the SMD model.

$$E_{\text{ab}}=E(\text{slab}+\text{organics})-E(\text{slab})-E(\text{organics})$$
(2)

where E(slab + organics) and E(slab) are the total energy of the substrate system that adsorbs or does not adsorb organics, respectively. E(organics) is the total energy of the organics.

3 Results and discussion

3.1 Characterization of the biochar and adsorption of TTC

The capacities for TTC adsorption on biochar increased with increasing treatment temperatures, as shown with two different concentrations (Additional file 1: Figure S1). Among the biochar samples examined, BC700 exhibited the most efficient adsorption and removed up to 54.5% and 49.17% of the TTC from the two water systems, respectively. This  is consistent with the TTC adsorption capacities reported in prior studies (Additional file 1: Table S2), suggesting that the trend is generalizable across various feedstocks. The specific surface area (SSA) of BC700 was significantly higher than that of the other biochar examined (Additional file 1: Table S3), implying the presence of more adsorption sites in BC700. This was corroborated by the total pore volume. However, the volume of the micropores in BC700 was lower than that in BC600, possibly due to pore  collapse.

resulting from charring deactivation and carbon melting during high-temperature combustion (Lian and Xing 2017). Additionally, the increased ash content led to a reduction in the microporous surface area (Leng et al. 2021). Micropores facilitate the adsorption of tetracycline; however, they restrict its diffusion in liquid, especially within nanometer-sized pores with diameters less than 3 nm (Wang et al. 2022) An increase in the mesopore volume of BC700 was observed, with an average pore diameter of only 4.391 nm. This may compensate for the deficiency in adsorption sites due to the reduction in micropore volume.

3.2 Possible chemical bond in adsorption

The taxonomic categories of the functional groups present in the five types of biochar, along with their respective proportions are illustrated in Fig. 1a. At a pyrolysis temperature of 700 °C, a marked reduction in functional groups, particularly acidic functional groups, was observed (Fig. 1b). This phenomenon was attributed to the thermal decomposition reactions that occurred during high-temperature pyrolysis of the biochar. The FTIR data in Additional file 1: Figure S3 and Table S4 show a broad peak in the range of 2800 to 3100 cm−1 for O–H stretching vibrations of BC300. The C = O stretching vibrations gave a peak at 1730 ± 20 cm−1 and the O–H bending vibration observed at 960 ± 20 cm−1 indicated the presence of carboxylic functional groups. With increases in the adsorption time, the intensity of the carboxylic acid peak gradually decreased, indicating a reaction between the carboxyl groups and TTC. The nucleophilic character of the carboxyl group in TTC enabled it to attack the carbonyl groups (C=O) in functional groups such as carboxyls, carbonyls, and esters on the biochar surface, resulting in the formation of ester linkages (−COO−). This was supported by previous studies on the reactivity of carbonyl functional groups with organic compounds (Zhou et al. 2019). Figure 1c shows that the content of acid functional groups in BC700 was lower than  that in the other biochar, suggesting that acidic functional groups were not the primary sites for adsorption of TTC. However, BC700 exhibited the highest content of alkaline functional groups. The peak at 3300–3600 cm−1 was attributed to N–H stretching vibrations, indicating the presence of amines. The peaks at 688 ± 20 cm−1 and 1580 ± 20 cm−1 indicated the presence of primary amines, which could react with the carbonyl groups of TTC in condensation reactions to form amide bonds. The peak at 1630–1690 cm−1 for BC700 confirmed the presence of amide groups (−CONH−). However, these results were not observed with copper sulfate tests, possibly due to their relatively low content.

Fig. 1
figure 1

The functional groups by XPS (a) and distribution of quantification by Boehm titration (b, c)

Full size image

3.3 Adsorption behavior

Adsorption kinetics were used to assess the adsorption of TTC with various concentrations on biochar prepared with different pyrolysis temperatures (Additional file 1: Figure S4–S6) (Hoslett et al. 2021). At a relatively low concentration of 40 mg L−1, all adsorption processes exhibited characteristics consistent with the Elovich model, which is an empirical model derived from the assumption of chemical adsorption. The rate-limiting step resulted from the energy barrier present on the surface where adsorption occurred (Choudhary et al. 2020). The adsorption rate constants ranged from 0.26 to 2.74 mg·g−1·min−1, indicating slow adsorption via chemical reactions (Additional file 1: Table S5). It is noteworthy that the desorption rate for BC300 exceeded the adsorption rate, which could account for its relatively low adsorption capacity. Nevertheless, it is theoretically expected that BC300, owing to its high concentration of acidic functional groups, would form stable chemical bonds to enhance adsorption. This apparent discrepancy suggested that the adsorption capacity resulting from acidic functional groups was limited, underscoring the operation of other processes. Furthermore, it was evident that the remaining biochar was better described by the intraparticle diffusion model, as indicated by the relatively low ki3 value, which suggested that insufficient adsorption sites  were one of the factors limiting the adsorption process; additionally, a plot of the intraparticle diffusion model did not pass through the origin, indicating the operation of other rate-determining steps (Liu et al. 2021).

The Dubinin-Ashtahkov model is based on the assumption of a single site; that is, adsorption sites have different adsorption capacities, which are related to the temperature. Accordingly, thermodynamic experiments were conducted at three temperatures and yielded good fits (0.97–0.99), which attested to the strong correlation between the experimental and theoretical data (Fig. 2 and Table 1). The EDA was used to determine the adsorption affinity and was found to be independent of Q0. These results indicated that the capacity for tetracycline adsorption by biochar gradually increased with an increase in heat treatment temperature, suggesting that biomass carbonization was a contributing factor to tetracycline adsorption by biochar. During carbonization, the biomass underwent continuous deoxygenation and dehydrogenation, resulting in the formation of carbon flakes. Furthermore, as the temperature increased, these carbon flakes coalesced, resulting in the creation of more bulk carbon.

Fig. 2
figure 2

Isotherms for tetracycline adsorption on biochar fitted with the Dubinin-Ashtahkov model; a 15 °C, b 25 °C and c 35 °C

Full size image
Table 1 Fitting parameters with the Dubinin-Ashtakhov model for tetracycline adsorption on biochar
Full size table

3.4 Carbon structure and correlation

To develop a more comprehensive understanding of the structure of the biochar, the carbon distribution was examined and is presented in Fig. 3b; it was classified into five primary regions based on the 13C NMR spectra shown in Additional file 1: Figure S7. The carbon structures contained methoxy (50–60 ppm), carboxyl and carbonyl carbons (160–210 ppm), and the side chains contained alkyl (0–50 ppm) and O-alkyl (60–110 ppm) groups. It is noteworthy that the composition of the biochar evolved considerably with the heating temperature. For instance, at the lower heating temperature of BC300, the proportion of aliphatic carbon (such as alkyl and methoxy carbon) was the highest, mainly due to incomplete pyrolysis of the lignin. However, as the carbonization temperature was increased, more aliphatic carbon atoms were transformed into sp2-hybridized carbon atoms (such as furan and aromatic carbon atoms, 110–160 ppm), thereby increasing the level of polyaromaticity (Li et al. 2020). Therefore, the aromaticity (Far) and aromatic condensation (χb) indices also increased and were valuable indicators providing information about the carbon structure of the biochar. Aromatic condensation refers to the combination of aromatic compounds to form larger molecules, which is often accompanied by the loss of small molecules such as water or hydrogen chloride. As χb increases, the biochar exhibits a higher carbon mass and an increased proportion of aromatic structures. These aromatic structures increase the SSA and porosity of the biochar, thereby enhancing its adsorption capacity and chemical stability (Wang et al. 2018). The values of Far and χb for the biochar are presented in Table 2.

Fig. 3
figure 3

a Heatmap showing correlation analyses among selected parameters (* p <  = 0.05, ** p <  = 0.01, *** p <  = 0.001). b Carbon distribution of the biochar calculated from the 13C NMR spectrum. c Relative contributions of the interactions occurring during TTC adsorption by biochar

Full size image
Table 2 Selected physical–chemical properties of the biochar
Full size table

Consistent with the aforementioned results, the polarity index [(O + N)/C] and H/C atomic ratio exhibited decreasing trends, shifting from 0.7789 to 0.2726 and 0.8997 to 0.3410, respectively, as the heating temperature was increased (Table 2). In addition, the hydrophilic index [O/C] decreased significantly, from 0.7543 to 0.2540, indicating the loss of polar functional groups. However, as the polarity decreased, both the SSA and the pore volume increased, providing more binding sites for tetracycline. This suggests that factors other than polar functional groups had a greater impact on the adsorption of tetracycline onto the biochar.

Therefore, thirteen different parameters were used to evaluate the correlations between each factor and the adsorption capacity with Pearson correlation analyses (Fig. 3a). The heatmap revealed that qe was highly correlated with EDA at the 0.001 level, with a correlation coefficient (p) between 0.8 and 1. This was explained by the fact that EDA includes all forces that contribute to the adsorption affinity, and it plays a crucial role in determining the adsorption capacity. Furthermore, qe demonstrated a strong correlation with χb at the 0.01 level (p = 0.8–1), suggesting that the level of aromatic condensation in the biochar played a significant role in influencing the adsorption capacity. This is not surprising since a higher χb corresponds to higher condensation of the biochar, resulting in larger SSA and pore volumes, as confirmed by the strong correlation between SSA and χb. Moreover, a significant correlation was observed between qe and the basic functional groups at 0.05. This was attributed to the higher reactivity of the two amino groups in TTC compared to the two carboxyl groups, leading to stronger reactions with basic functional groups rather than acid functional groups. Additionally, qe was negatively correlated with O/C and (O + N)/C, which was consistent with previous findings.

The impact of χb on EDA was determined by plotting a linear correlation graph, as shown in Fig. 4. The results demonstrated a significant correlation between χb and EDA at all three temperatures (R2 = 0.94–0.98). As the degree of aromatic condensation in the biochar increased, there was a corresponding increase in its affinity for tetracycline. This suggested that biochar with higher levels of χb may be a more effective adsorbent for tetracycline removal (Peng et al. 2016). Concurrently, competitive adsorption experiments demonstrated that as the degree of aromatic condensation increased, the proportion of chemical reactions gradually decreased (refer to Additional file 1: Figure S8 and Figure S3c). This was attributed to a decrease in the number of acidic functional groups at the chemical reaction sites, resulting from the increased treatment temperature.

Fig. 4
figure 4

Linear correlation between χb and EDA; a 15 °C, b 25 °C and c 35 °C

Full size image

3.5 Establishment of microscale models and adsorption mechanisms

Hückel’s rule is a fundamental principle that describes the aromaticity of molecular systems. Specifically, it states that a planar cyclic molecular system exhibits aromaticity when the number of π electrons in the system is equal to 4n + 2 (where n is a nonnegative integer). Based on Hückel’s rule of aromaticity, χb can be used to calculate the theoretical sizes of aromatic ring clusters. Additionally, DFT can be employed to establish different microunit carbon ring structures for biochar formed with various treatment temperatures, as shown in Additional file 1: Figure S9 (Yang et al. 2016). All constructed models were optimized to achieve the most stable configurations. Furthermore, different adsorption configurations between the biochar and TTC were studied, and three positions were selected, as shown in Additional file 1: Figure S10. The three positions described in this study were ultimately selected based on multiple experiments and represent different binding characteristics. The binding energies of the resulting complexes were calculated and are presented in Table 3. The positive BC300 adsorption energies indicate that the reaction with tetracycline had thermodynamic barriers, which may impede the progress of the reaction and require a greater energy input. The observed effect was ascribed to the relatively low degree of carbonization exhibited by BC300, which could lead to a reduction in the extent of carbon framework condensation and a consequent decrease in the surface charge density (Yang et al. 2021b).

Table 3 Computational binding energies (Eab) and minimum distances for the sheet-compound configurations
Full size table

The adsorption energies of BC300 were all relatively low positive values, indicating that adsorption may be governed by weak van der Waals forces (0.4–4  kJ mol−1) or π-π interactions (1.5–3 kJ mol−1) (Aliakbar Tehrani and Kim 2016). The carbon rings in the biochar have π electron densities, which enable π-π interactions with the four arene rings in tetracycline (Zeng et al. 2019), (Chen et al. 2007). Studies have shown that for most compounds containing arene rings, the distance at which π-π interactions occur is commonly less than 3.5 Å (Fiori et al. 2017). Within this range, the interactions between π electron clouds reach a maximum value, and stable π-π bonds are formed. Except for BC300, Position 1 was selected as the optimal binding model for investigating the adsorption of tetracycline on the biochar; all other biochar exhibited exothermic adsorption of tetracycline at this position, suggesting that the binding reaction was thermodynamically feasible and facile. The maximum binding energy of BC700 ( −0.946   eV) indicated that in addition to the van der Waals forces and π-π interactions that are known to operate in such systems, electrostatic interactions or hydrogen bonds (25–40 kJ mol−1) may also be contributing factors (Deng et al. 2023).

However, the binding configuration observed at position 1 revealed that the tetracycline binding site on the biochar contained many oxygen-containing functional groups with negative charges similar to those of the biochar (Additional file 1: Figure S11a), making electrostatic interactions highly unlikely. Therefore, it is suggested that hydrogen bonding and the combined effects of van der Waals and π-π interactions may be responsible for the observed binding behavior. The oxygen functional groups in tetracycline are capable of serving as hydrogen bond donors, thereby forming hydrogen bonds with the π electron clouds of the carbon structures present in the biochar (Yang et al. 2011). It was observed that the shortest bond lengths between tetracycline and the five types of biochar ranged between 2.66 and 2.84 Å, which were consistent with the known bond lengths for intermolecular forces (Das et al. 2014). Furthermore, the  size of the carbon clusters present in the biochar, which was the only variable in the simulated adsorption experiments, was strongly correlated with the strength of the π electron cloud, indicating that larger carbon clusters resulted in stronger and more stable adsorption interactions. The optimal adsorption configuration and minimum bond distance for the tetracycline on biochar are shown in Fig. 5.

Fig. 5
figure 5

Optimal adsorption configuration and calculated bond distances for tetracycline adsorbed on biochar: a BC300; b BC400; c BC500; d BC600 and e BC700

Full size image

The microscopic interactions between the biochar functional groups and tetracycline were investigated with the syntheses of various functionalized biochar. X-ray photoelectron spectroscopy (XPS) was employed to characterize the biochar surfaces, revealing that hydroxyl (−OH), carboxyl (−COOH−), pyridine-N (−PDN), pyrrole-N (−PLN), and graphitic-N (−GN) were the primary functional groups present on the biochar (Fig. 1c). These functional groups are common in biochar (Xiao et al. 2018). Therefore, based on the BC700 simulation model, the functionalized biochar was composed of five distinct functional groups, namely, BC-OH, BC-COOH, BC-GN, BC-PDN, and BC-PLN, as illustrated in Additional file 1: Figure S12. The molecular electrostatic potential map revealed that the carbon framework of the biochar was rich in π electrons, thus displaying strongly electrophilic character (Additional file 1: Figure S12) (Ai et al. 2019). It was observed that, with the exception of BC-GN, all of the functionalized biochars exhibit negative charges (Xiao et al. 2020). Moreover, all of the functionalized biochar exhibited a higher binding energy with tetracycline (−1.0866 to −1.8370 eV) than BC700, indicating the possibility of chemical adsorption (Table 4). This may be attributable to the high negative charge carried by the functionalized biochar, which could interact electrostatically with the nucleophilic nitrogen (amine functional group and acyl amide functional group) in tetracycline. The calculated frontier orbitals revealed that the Eg1 values of all biochar specimens were lower than Eg2, thereby signifying that the electrons present on the surface of the biochar were stimulated and transmitted to the tetracycline molecules (Additional file 1: Table S6) (Hou et al. 2021). Moreover, the charge density images provided a lucid depiction of the binding locations between the functionalized biochar and tetracycline (Fig. 6) (Guo et al. 2016). As demonstrated in Fig. 6c, the incorporation of graphite nitride decreased the electrostatic potential and energy, thereby generating overlap between the nucleophilic regions of tetracycline and the nondoped electron-rich zone of the biochar (Dai et al. 2020). However, the higher adsorption energy (−1.0866 > −0.946 eV) suggested that adsorption may have involved electrostatic interactions. Moreover, the intermolecular forces also increased the binding energy. BC-COOH exhibited the highest overlap with tetracycline and showed the greatest adsorption binding energy (−1.8370 eV), possibly due to the formation of hydrogen bonds between the carboxyl groups of the biochar and the nitrogen atoms of the amine functional groups in tetracycline, thereby enhancing the intermolecular interactions (Ai et al. 2019). Likewise, BC-OH bound in a similar manner (Peiris et al. 2017), (Zhu et al. 2019). Binding between the BC-PDN and BC-PLN with tetracycline occurred through the formation of hydrogen bonds with the acyl amide functional groups (Zhang et al. 2021b).

Table 4 Binding energies (Eab) for tetracycline adsorption on functionalized biochar
Full size table
Fig. 6
figure 6

Charge density differences for TTC adsorption on functionalized biochars; a BC-OH; b BC-COOH; c BC-GN; d BC-PDN and e BC-PLN

Full size image

3.6 Potential adsorption mechanisms and binding sites

The possible micromechanisms and binding sites are summarized in Fig. 7. BC300 mainly adsorbed tetracycline (69.21%) through chemical bonding. This is attributed to the high acidity functional group content (4 mmol g−1) of low-carbonized biochar. The nucleophilic nature of the carboxyl groups in TTC allows them to attack carbonyl groups (C=O) in functional groups on the biochar surface, forming ester bonds (−COO−). By comparison, due to low surface charge density, intermolecular forces dominated by van der Waals or π-π interactions play a limited role. With increasing carbonization, the functional groups in biochar decrease, and the carbon structure becomes more ordered. High-carbonized biochar is mainly controlled by intermolecular forces (91.1%), with strong π electron clouds on large cluster carbon structures forming π-π interactions with the conjugated benzene rings of tetracycline. Additionally, oxygen functional groups in tetracycline can act as hydrogen bond donors, forming hydrogen bonds with the π electron clouds in the carbon structure of biochar. Van der Waals forces also occur in the adsorption system. The high affinity of functionalized biochar for tetracycline is attributed to the strengthening of electrostatic interactions. The electron-rich regions of biochar can engage in electrostatic interactions with the nucleophilic regions (amine and amide functional groups) in the nitrogen area of tetracycline. BC-COOH exhibits the highest degree of overlap with tetracycline and has the maximum adsorption binding energy (−1.8370 eV). This may be due to the formation of hydrogen bonds between the carboxyl groups of biochar and the nitrogen atoms of the amine functional groups in tetracycline, enhancing intermolecular interactions.

Fig. 7
figure 7

Schematic illustration of possible micromechanisms for TTC adsorption on biochar

Full size image

3.7 Comparative study

Regardless of the modification method employed, the efficient adsorption of tetracycline is the criterion for assessing the performance quality of biochar. Additional file 1: Table S7 provides a comprehensive comparison of the adsorption efficiency and mechanisms of modified biochar on tetracycline over the past two years, encompassing various feedstock types. It is noteworthy that 86.7% (13 out of 15) of the studies opted for elevated temperatures (≥ 500 °C) as the preparation temperature. This indirectly corroborates the favorable adsorption of tetracycline by biochar under conditions of high aromatic condensation. The impact of modification on the structural regulation of biochar for the adsorption of tetracycline can be broadly categorized into three types: increased functional groups, enhanced graphitization, and the doping of metals. Simultaneously, an increase in SSA and pore volume was observed across all studies, thereby augmenting the adsorption sites. The modification of functional groups alone has limited impact on enhancing the adsorption efficiency of tetracycline. The La-activated orange peel biochar (La-BC) prepared by Chen et al. (Chen et al. 2023) exhibited an impressive specific surface area (SSA) of 2888.33m2 g−1, accompanied by an augmentation in both the quantity and diversity of functional groups. However, its adsorption capacity for tetracycline is only 143.2 mg g−1, indicating a low enhancement efficiency attributed to the surface functional groups and adsorption sites. However, with the concurrent increase in graphitization and functional groups, the adsorption efficiency of biochar for tetracycline can be significantly enhanced. Wang et al. (Wang et al. 2023) synthesized microalgae-derived biochar (CVAC) with nitrogen doping modification, concurrently enhancing the presence of oxygen-containing functional groups and graphitization level. Compared to La-BC, CVAC exhibits a significantly lower specific surface area (SSA) of only 491.16 m2 g−1. Nevertheless, it achieves a remarkable adsorption capacity of 381.584 mg g−1 for tetracycline. Therefore, enhancing the aromaticity of biochar is crucial for improving its adsorption capability for tetracycline. Additionally, the impact of metal doping on the performance of biochar varies. Zhang et al. employed Iron-zinc modified biochar, resulting in a significant enhancement of SSA (747.76 m2 g−1); however, the adsorption capacity was only 94.63 mg g−1. This is attributed to the potential occupancy of certain microspatial domains by iron and zinc ions, resulting in a limited increase in adsorption sites. In contrast, Luo et al. (Luo et al. 2022) incorporated Fe3O4 nanoparticles into biochar, effectively reducing pore size and introducing additional functional groups. Despite a SSA of only 75.34 m2 g−1, the adsorption capacity for tetracycline reached 184.5 mg g−1.

4 Conclusions

In this study, key factors and microscopic adsorption sites of biochar with varying carbonization degrees (300–700 °C) in the adsorption of tetracycline were explored. The results indicated that biochar pyrolyzed at higher temperatures exhibited superior adsorption performance for tetracycline, confirming the aromatic condensation of biochar as a key factor influencing adsorption. Thus, different biochar-tetracycline adsorption configurations were established under various carbonization temperatures, which circumvents the limitations imposed by variations in raw material types and pyrolysis temperatures, contributing to the standardization of biochar research. Although low-carbonized biochar predominantly adsorbs tetracycline through chemical bonding, the available adsorption sites are limited. On the other hand, high-carbonized biochar, with larger carbon clusters offering numerous adsorption sites, predominantly adsorbs tetracycline through intermolecular forces. Overall, this study reveals the most valuable directions for the modification of biochar to enhance tetracycline adsorption, offering insights for the development of more cost-effective and efficient biochar adsorption materials.