KMnO₄ modified biochar derived from swine manure for tetracycline removal

Antibiotics are used widely in animal and human medicine to prevent or treat infectious diseases (Marzbali et al. 2016). Over 25,000 tons of antibiotics are used annually in China, approximately 32% of which are applied as feed additives (Dai et al. 2019). Tetracycline (TC) is the most commonly used antibiotic, is produced at large-scale and widely available in the market. However, more than 70% of TC cannot be digested directly by animals, the undigested TC being excreted and it accumulates in the soil, sediment and aquatic environments (Luo et al. 2011; Guo et al. 2022). Livestock wastes are a major source of undigested TCs and antibiotic resistance genes in the environment (Zhou et al. 2017). Worse still, currently available livestock waste treatment processes, such as composting, do not mitigate their dissemination efficiently into the environment and they may even be enriched after composting (Berendonk et al. 2015). Therefore, there is an urgent need to find a sustainable and practical method of removing residual TC from livestock industry wastewater.

Considering its simplicity and high removal efficiency, adsorption was considered as a good practical choice for TC removal (Chen et al. 2017a; Liang et al. 2018). Biochar is a porous, stable and carbon-rich solid, produced by the pyrolysis of agricultural wastes, animal manure, sewage sludge and other organic wastes, in an oxygen-limited environment (Chen et al. 2017b; Leng et al. 2021). It has been used widely as the adsorbent for pollutant removal due to its high specific surface, heterogeneous pore structure, and large surface functional groups, but its efficiency varies considerably for different feedstocks, pyrolysis conditions, modifying agents and other factors (Cole et al. 2017). The preparation of livestock waste-derived biochar could limit the negative environmental effect efficiently by reducing the CO₂ emitted and treating wastewaters released by the livestock industry, which is often contaminated with TC (Li et al. 2017a; Wang et al. 2018). In addition, in preparing biochar, the antibiotic residues and pathogens in livestock waste would be eliminated, reducing the risk of these hazardous materials entering the environment. However, although much attention has been placed on livestock waste-derived biochar's apparent ability to remove pollutants (Mitchell et al. 2015), its adsorption capacities vary remarkably. For example, the TC adsorption capacities of some biochars exceeds 1536 mg g⁻¹ (Yang et al. 2019), but that of most biochar for TC is below 50 mg g⁻¹ (Luo et al. 2019; Tan et al. 2019).

Various methods have been adopted to alter the specific surface, functional groups, porous structures or other physicochemical characteristics of pristine biochar, to improve its adsorption performance (Vithanage et al. 2016). KMnO₄ is considered an effective adsorbents modification reagent, with great potential to accelerate the adsorption rate and enhance the pollutant removal performance of biochar (Cantu et al. 2014; Song et al. 2014; Zhang et al. 2020). KMnO₄ acts not only as a potent oxidizing agent but also as the precursor of MnO_x, thus forming novel, engineered biochars. Previous studies have found that KMnO₄ modification improves the number of oxygen-containing functional groups (OCFPs), enlarges the specific surface, and provides more active sites on biochar (Feng et al. 2014). However, biochar derived from animal manure usually has high ash content, with a relatively higher oxygen/carbon ratio and lower specific surface (Zhao et al. 2016). Thus, the TC removal performance may differ from other biochars. ‘Treating waste with waste’, using animal manure-derived biochar as a scavenger for wastewater may therefore be a promising and practical measure.

In this study, swine manure was chosen as the biochar feedstock and KMnO₄ as the activator to further improve its sorption capacity. The study's aim's were to: (1) compare the properties of animal manure-derived biochar before and after KMnO₄ modification; (2) evaluate the TC sorption capacity of MnO_x-SBC; and (3) clarify the possible mechanisms of TC adsorption on livestock waste derived biochar.

Swine manure was collected from a pig farm in Liaocheng, China. TC (C₂₂H₂₄N₂O₈, purity >98.5%) was purchased from Macklin Co., Ltd. (Shanghai, China). All other chemicals used, including HCl, NaOH, and KMnO₄, were analytical grade and obtained from Sinopharm Company, China.

Air-dried swine manure was ground to pass through a 2 mm sieve, packed into a ceramic pot and pyrolyzed at 450 °C for 4 hours in a tube furnace (heating rate 15 °C min⁻¹ and N₂ flow rate 50 mL min⁻¹). The resulting biochar, marked as SBC, was washed, dried at 60 °C to a constant weight and sieved for the 0.15–0.45 mm fraction for analysis.

The biochar adaptation with KMnO₄ was conducted as described in the literature with some modifications (Chen et al. 2018): 5.0 g SBC was immersed in 500 mL 5% KMnO₄ solution and concussed ultrasonically for 24 hours, then heated at 450 °C for 1 hour in a muffle furnace and washed with distilled water. The biochar was then dried at 105 °C for 24 hours, and marked as MnO_x-SBC for characterization and batch experiments.

A scanning electron microscope (SEM, SU8010, Japan) equipped with an energy-dispersive X-ray spectroscope (EDS, Horiba EX-350, Japan) was used to determine the biochar's morphology and the main elements on its surface. Surface functional groups were identified by Fourier transform infrared spectroscopy (FTIR, Nicolet 6700, Thermo Electron Scientific Inc., USA). The specific surface, porosity and pore volume were determined by Brunauer-Emmett-Teller analysis (BET, ASAP2020, USA). The XPS spectra were obtained on an Escalab250Xi (Thermo Fisher Technologies, Escalab 250XI, USA). The point of zero charge (pH_pzc) was estimated by the pH drift method (Jang et al. 2018).

Initial solution pH effects on TC removal performance were investigated using batch experiments with the pH ranging from 2 to 11. The initial pH was adjusted using 0.1 mol L⁻¹ HCl or NaOH solution. Approximately 10 mg of adsorbent (SBC and MnO_x-SBC, respectively) was added to brown glass tubes containing 40 mL TC solution (200 mg L⁻¹) at different pHs. The tubes were placed in an opaque shaker and shaken at 180 rpm for 24 hours. After filtration with a 0.22 μm syringe filter, the filtrate's residual TC concentration was measured by spectrophotometry at λ = 365 nm (Xiang et al. 2020). Blank experiments without biochar addition were also conducted to eliminate TC decomposition.

For the kinetic adsorption experiments, approximately 10 mg biochar was added to 80 mL TC solution (200 mg L⁻¹, pH unadjusted) in brown glass tubes, and shaken at 180 rpm for 24 hours. Samples were taken at regular intervals and measured at λ 365 nm.

For the adsorption isotherms, approximately 10 mg of biochar added to 40 mL TC solution at different initial concentrations (25–300 mg L⁻¹, pH unadjusted) and shaken for 24 h. The residual TC concentration in the filtrate was measured as above.

The effects of temperature on TC adsorption were determined in batch experiments at different initial TC concentrations (25–300 mg L⁻¹) and 25, 35, and 45 °C, respectively. Other procedures were as above.

After KMnO₄ modification, the proportion of C in MnO_x-SBC had decreased and that of O increased (Table 1), which is attributed to the oxidizing effect of KMnO₄. The attachment of MnO_x particles also lowered the proportion of other elements. The ratios of O/C and (O + N)/C in MnO_x-SBC were also slightly higher than those of pristine biochar, which was consistent with results obtained by others (Liao et al. 2022). The lower H/C means higher aromaticity, while higher O/C indicates greater polarity (Dieguez-Alonso et al. 2019).

After KMnO₄ modification, the SBC's pore volume and BET surface changed from 0.082 cm³·g⁻¹ and 41.58 m²·g⁻¹ to 0.134 and 66.27, respectively (Table 1). The adsorbent's pores were divided by International Union of Pure and Applied Chemistry (IUPAC) into micro-, meso- and macro-pores, according to the pore diameter (Li et al. 2017b). With a sufficiently large biochar pore diameter, the large microporous structure allowed TC molecules to enter the pores quickly and become distributed evenly on its surface, leading to MnO_x-SBC's superior removal performance. Due to the size exclusion effect, the biochar pore diameter should be at least 1.7 times that of the TC molecule, indicating that the pore-filling participates in TC adsorption.

The predominant component of TC was when the initial solution pH was below 4.0, but became TCH⁻ and TC²⁻ when the pH exceeded 7.7 (Jang et al. 2018). Increasing the pH increased deprotonation of the functional groups on the biochar surface, so that the TC combined with the biochar by co-precipitation, ion exchange, and complexation (Zhang et al. 2020). This reduced the adsorption affinity of biochar for TC, and the adsorption capacity fell at high pH (Xu & Li 2010). However, large amounts of TC were still adsorbed by MnO_x-SBC, indicating that other effects, such as pore filling, π-π action and H bonding, also played essential roles (Wang et al. 2020). In summary, MnO_x-SBC can adsorb sufficient TC across a range of pH values, with broad application prospects in TC removal.

Kinetic models, including the pseudo-first-order (PFO), pseudo-second-order (PSO), interparticle diffusion and Elovich models, were applied to estimate the biochar TC adsorption rate and determine the adsorption equilibrium time (Table 2). The PSO model's R² for MnO_x-SBC was higher than that of the PFO and Elovich models, which is consistent with previous results (Du et al. 2021). According to the PSO model, biochar's TC adsorption has two dynamic steps: a single layer of TC molecules forming first, with TC molecules adsorbed on the biochar's micropore surface, with chemisorption then becoming the dominant process as monolayer physical adsorption approaches saturation (Jing et al. 2014). The TC reaction process on MnO_x-SBC is also rapid because the adsorption rate constant K is less than 1, indicating that TC adsorption on MnO_x-SBC comprises chemisorption and physical capture (Hu et al. 2021).

The intraparticle and liquid film diffusion process results indicate that intraparticle diffusion was involved, and occurred mainly on the surface (Figure 6(c)). However, the large fitting curve intercept for both MnO_x-SBC and SBC suggests that the intraparticle diffusion process was a major speed-limiting step but not the only one – boundary layer diffusion or external mass transfer may also have been involved.

The maximum TC adsorption capacity of MnO_x-SBC reached 95.81, 99.11 and 105.85 mg g⁻¹, respectively, at 298, 308 and 318 K (Table 3), 39.8–46.4% higher than those of SBC. MnO_x-SBC's enhanced TC adsorption capacity may be attributable to its abundant OCFPs, larger specific surface and higher oxygen/carbon ratio. The values of R_L and n⁻¹ were 0.923 and 0.021, respectively, suggesting a high affinity and numerous suitable adsorption sites on MnO_x-SBC for TC species (Huang et al. 2018).

Generally, the π-π EDA interaction was weak at low pH, since TC molecules are cationic and the biochar surface is commonly positive (Mitchell et al. 2015). With increasing pH, TC becomes zwitterionic, and the EDA interaction between TC and biochar is strengthened (Yang et al. 2019). However, MnO_x-SBC's TC removal performance is relatively high with pH between 4 and 11 in this study, indicating that electrostatic interaction participated in TC adsorption but was not the dominant factor.

FTIR analysis showed that, after TC adsorption, the position of the stretching peak of each functional group – C–O–C in the aromatic ring structure, and C=O and O–H in the alcohol and phenol – had shifted. It was concluded that the aromatic group participated in the entire adsorption process (Chen et al. 2018), and π-π conjugation formed between aromatic structures, including C–O–C and C=O in the biochar and the aromatic ring structure in TC. The H-bonding between -OH and the oxygen-containing functional groups also facilitated adsorption (Zhang et al. 2020). In summary, the difference in the adsorption mechanisms could be responsible for the changes in the interactions of TC absorbed by MnO_x-SBC.

KMnO₄ modification alters the surface structures and properties of SBC, further enhancing the TC removal performance of MnO_x-SBC. MnO_x-SBC's TC adsorption behavior was a spontaneous, endothermic, and monolayer chemical process. The process is best fitted by the pseudo-second-order model and Freundlich model. TC adsorption onto the MnO_x-SBC surface is caused mainly by electrostatic interaction, pore-filling, π-π conjugation, H-bonding and complexation of the oxygen-containing functional groups. Biochar derived from swine manure, especially after KMnO₄ modification, may thus be suitable for removing TC from aqueous solution by sorption.

This study was supported by the Natural Science Foundation of China (Grant No. 41807092), The Open Project of Liaocheng University Animal Husbandry Discipline (No. 319312101-18), the Project of Shandong Province Higher Educational Science and Technology Program (KJ2018BAF034), and the Startup Foundation for Ph.D. of Liaocheng University (318051839).

Zan Fu: Conceptualization, resources, writing original draft. Yurong Chen: Investigation, data curation. Yanyan Lu: Methodology, formal analysis. Yue Wang: Investigation, data curation. Jiahui Chen: Methodology, formal analysis. Youxin Zhao: Writing, review and editing. Xiaofei Tian: Supervision, writing review and editing, and funding acquisition.

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.