Improved Adsorption of Tetracycline in Water by a Modified Caulis spatholobi Residue Biochar.

A potassium modified biochar (KBC) using Caulis spatholobi residue as the raw material was prepared by adopting a two-step method of pyrolysis followed by high-temperature potassium hydroxide activation, and its properties were characterized. Activation using potassium hydroxide under high temperature induced the loss of CaCO3 and partial C on biochar, which created a high specific surface area (1336.31 m2/g) together with a developed pore structure. pH displayed a slight influence on tetracycline adsorption, which signified the slight influence of the existence of tetracycline and the charge potential of biochar. Besides, pore filling, hydrogen bonding and π-π EDA stacking interactions possibly resulted in tetracycline adsorption on biochar. Tetracycline adsorption was fast in the original period, followed by a slower rate of adsorption until equilibrium was reached. Adsorption kinetics of tetracycline could be described using secondary and Elovich kinetic models. Adsorption isotherms for tetracycline were well fitted to the Langmuir isotherm model, and the maximum adsorption capacity of KBC was 830.78 mg/g at 318 K. According to a study of the thermodynamics, the adsorption of tetracycline on KBC was an endothermic reaction process. Corresponding results in the present study demonstrated that high-temperature potassium hydroxide activation enabled biochar to effectively eliminate tetracycline from water and wastewater.

Introduction

Since penicillin was discovered in 1928, the application of antibiotics has risen sharply.[1] However, this sharp rise in use has caused the unintentional release of antibiotics into the environment, a potential source of ecological problems. Trace quantities of veterinary and human antibiotics have recently been identified in various matrices,[2−4] especially in the aquatic environment.[5] Tetracycline (TC) is one of the most extensively applied antibiotics globally, primarily in the treatment of animal diseases and in animal husbandry.[6] TC can be released into the aquatic environment in both its primary and metabolic forms,[6] and TC pollutants can adversely affect the entire ecosystem, not only spawning highly resistant strains of bacteria[7] but also seriously harming aquatic life forms.[8] In addition, these pollutants can threaten human health by means of drinking water and the food chain.

There are many approaches to TC removal in current use, including biological treatment,[9] oxidation techniques,[10] adsorption,[11] membrane osmosis,[12] and ultrasonic cavitation.[13] Adsorption is one of the most normally employed approaches based on the ease of use, low cost, high efficiency, as well as absence of secondary pollution. Biochar—a carbon-rich solid that can be acquired through heating biomass in an oxygen-deficient environment—possesses a porous structure similar to that of activated carbon and, as such, is one of the most frequently applied and efficient adsorbents for use in eliminating a range of pollutants from water.[14−16] Biochar features are a porous structure, high specific surface area, rich surface functional group structure, and high mineral content. It has proven effective in removing TC from wastewater and is therefore often used in the treatment of antibiotic wastewater.[17] Various types of biomass are used to make biochar, with agricultural and forestry biomass—one of the most abundant renewable resources on Earth[18]—such as bamboo, straw, and fruit shell[19−21] being the primary source.

Waste products from Chinese herbal medicine are abundant in China.[22,23] Thus, it is both economical and environmentally friendly to use such waste products as raw materials to prepare biochar. However, the adsorption capacity of biochar prepared by direct high-temperature cracking is generally low, which could be ascribed to limited pore structures, specific surface area, and mere functional groups on biochar with high charring temperature. Thus, an activator is needed to improve pore structures and specific surface areas of biochar, consequently reaching a higher adsorption capacity.[24] For instance, de-ashing treatment using HCl/HF enabled sludge-derived biochar was used to create an extremely high specific surface area together with a developed pore structure, which behaved excellently on adsorbing organic pollutants.[25]Caulis spatholobi is used in herbal medicine due to its various pharmacological effects, such as improving the hematopoietic system, regulating immunity, and antitumor, antivirus, and antioxidation properties.[26] It is rich in lignin, cellulose, hemicellulose, polysaccharides, phenolic ketones, and other macromolecular compounds that contain hydroxyl, amino, ether, and other functional groups.

The present study aimed to prepare high-performance biochar for removing TC. Activation using potassium hydroxide (KOH) under high temperatures was used to prepare biochar with a developed pore structure and specific surface area. The effect of biochar dosages and pH on TC adsorption, as well as the adsorption kinetics, thermodynamic properties, and adsorption capacity for TC, were studied.

Materials and Methods

Materials and Reagents

Caulis spatholobi (“chicken blood vine”) residue (BR) applied in the current work was acquired from a pharmaceutical company in Hangzhou, Zhejiang Province. Sodium hydroxide (NaOH, AR) and hydrochloric acid (HCl, 37%, AR) were acquired from Xilong Science Co., Ltd., and potassium hydroxide (KOH, AR) was from the Sinopharm Chemical Reagent Co.

Preparation of the Biochar

The collected dregs of Caulis spatholobi were cleaned with deionized water and dried in an oven at 60 °C. Besides, the dried samples were then crushed and sieved using a 0.18 mm sieve with the purpose of obtaining a uniform fine powder for use as a raw material.

Preparation of BC

The raw material was transferred to a quartz boat, which was subsequently positioned in the center of a tube furnace and heated up to 900 °C at 5 °C/min based on a nitrogen atmosphere. Besides, the sample was pyrolyzed for 2 h and then permitted to cool naturally to ambient temperature. Then, the sample was removed and rinsed with 1 M hydrochloric acid to remove impurities, followed by rinsing with deionized water to remove residual chemical reagents on the biochar, and subsequently dried in the oven at 80 °C for 12 h with the purpose of obtaining the product labeled “BC”. The yield of biochar was 35.4%.

Preparation of KBC

The BC obtained above was mixed with KOH to produce three mixtures with diverse mass ratios (KOH/BC = 1:1, 2:1, 3:1), which were then transferred to a mortar and pestle and ground thoroughly to ensure that the two were thoroughly intermixed. The ground mixture was then transferred to a quartz boat and subjected to the same heating and chemical treatment as the BC. The final product was labeled as “KBC”. A yield of 27.4% could be obtained on KBC, lower than that for BC. Activation using KOH and HCl can remove minerals in biochar, thus resulting in the decline in the yield of activated biochar.

Characterization

With parameters of 15 s equilibration intervals, 1.00 g/cm3 sample density, and 15.0 cm3/g low pressure dose, the Brunauer, Emmett, and Teller (BET) test was adopted for determining the specific surface area and pore volume of the adsorbents (ASAP 2020 PLUS, MICROMERITICS, USA). Then, the surface functional groups were identified using Fourier transform infrared spectroscopy (FTIR) and a Nicolet 6700 spectrometer (Thermo Nicolet, USA) in wavenumbers within the range from 500 to 4000 cm–1. The surface morphology of the samples was studied based on scanning electron microscopy (SEM) (SU8010, Hitachi, Japan) in 10 kV of acceleration voltage. Besides, X-ray diffraction (XRD, X’PertPRO, PNAlyptal, The Netherlands) in a 2θ range of 10–80° was used to study the crystallinity of the samples at 40 kV of working voltage and 40 mA of working current. In addition, the zero-point charge (pHpzc) of the modified biochar surface was determined using a Nano ZS90 nanoparticle sizer (Zeta potential, Malvern, UK), and the absorbance of TC was measured using a Shimadzu 1800 UV–vis spectrophotometer (Japan). The maximum wavelength of TC obtained by scanning was 353 nm.

Batch Adsorption Experiments

The impact of the quantity of adsorbent on the degree of TC adsorption was studied using a range of concentrations of BC and KBC (0.25, 0.5, 0.75, 1.0, and 1.25 g/L). The impact of pH on TC adsorption was also explored through controlling the pH in the range 1–10 using 1 M NaOH and 1 M HCl solutions. In all experiments, a certain amount of adsorbent was supplemented in a 100 mL beaker including 50 mL of a given concentration of simulated TC wastewater solution; this beaker was then covered to protect the contents from light and stirred using a magnetic stirrer for a day before the sample was then filtered using a 0.45 μm filter membrane.

Measurement of the adsorption kinetics was performed at 298 K, which was controlled by a water bath. The adsorbent dose was 0.5 g/L supplemented to three different concentrations of simulated TC contaminated wastewater solution (150, 200, 250 mg/L). These solutions were sampled at various time intervals from 10 to 1440 min following mixing. Adsorption equilibrium experiments were made at 298, 303, 308, and 318 K with solutions with initial TC concentrations within the range from 150 to 400 mg/L. Both adsorbent dose and pH values were kept constant in these experiments. In terms of all of the samples, the absorbance of TC was calculated based on the adsorption at 253 nm by a UV–vis spectrophotometer and a previously constructed TC calibration curve. Moreover, we repeated all of the experiments three times. Besides, the results were averaged. Then, the adsorption capacity (qe) and the removal precent of TC (R, %) were computed based on eqs and 2, separately:[27]where C0 refers to the initial concentration, C is the concentration at time t, W is the mass of the adsorbent, and V is the volume of solution.

Assay of Desorption and Regeneration

As illustrated in a previous study, sodium hydroxide (NaOH) could desorb residual TC on biochar.[28] In the present study, 0.1 M NaOH was mixed with biochar at 20 °C for 24 h. Repeated desorption using NaOH was conducted until the desorbed NaOH solution became colorless. The amount of TC in desorbed NaOH was analyzed. Afterward, deionized water was used to clean biochar upon desorption process, until the pH of washing water remained unchanged. Next, we dried the washed biochar at 80 °C to obtain the regenerated adsorbent. By performing four cycles of adsorption and desorption, the efficiency of regeneration was determined.

Results and Discussion

Characterization of the Adsorbent

N2 Adsorption/Desorption Isotherm and Pore Properties

With the aim of high-temperature KOH activation, the greatest specific surface area and most developed pore structures should be detected on KBC, rather than BC and raw BR. Figure a presents the N2 adsorption/desorption isotherms of the BC and KBC adsorbents. Both BC and KBC exhibit both type I and type IV isotherms, corresponding to microporous and mesoporous structures on biochar.[29] The rapid increase in adsorbed N2 at very low P/P0 is indicative of the existence of microporous structures in the two adsorbents, while the poor overlap of the N2 adsorption/desorption curves at higher P/P0 values shows that the pore inner surface area of both biochars was much larger when compared with the pore outer surface area, a sign pointing to the existence of mesoporous structures as well as micropores.[30] The pore size distributions within BC and KBC were obtained by the Barrett, Joyner, and Halenda (BJH) and Horvath–Kawazoe (HK) methods, as illustrated in Figure b. The pore-size distributions of BC and KBC were found to be mainly concentrated within 1–5 nm, with many mesopores and micropores, conforming to the results obtained from the N2 adsorption/desorption curves. Moreover, more pores with diameters of 0–10 nm were detected on KBC, as verified in Figure b, demonstrating the function of high-temperature KOH activation on forming pore structures.

Figure 1

N2 adsorption/desorption isotherms (a) and pore size distribution (b) for the BC and KBC adsorbents.

Table shows the distribution of the specific surface area and pore structure of BC and KBC. Both BC and KBC mainly consist of micropores, the values for total surface area and total volume of KBC were obtained using the BET, and Langmuir approaches are much larger than those of BC, which is consistent with other findings. Previous studies have shown that microporous activated carbon is produced when chemical activation is performed with excess alkali (e.g., NaOH and KOH). Under a high temperature, KOH was transformed into K2O and H2O, and thereinto, fragile C could react with H2O to produce H2 and CO, thus inducing C loss and production of pore structures. K2O could catalyze C reacting with H2O, which aggravated C loss and production of pore structures. The increase in micropores may ultimately lead to pore collapse and the formation of more mesopores and macropores.[31] Thus, the addition of KOH can promote the progress of microporous and mesoporous structures.

Table 1

Determinations of the Specific Surface Area and Pore Structure of BC and KBC

sample	SBET (m2/g)	SLangmuir (m2/g)	SMicro (m2/g)	VTotal (cm3/g)	VMicro (cm3/g)
BC	694.35	992.41	511.25	0.367	0.209
KBC	1336.31	1882.36	986.03	0.695	0.409

SEM and EDS

The surface morphologies of BR, BC, and KBC were studied with the use of SEM, and their elemental distribution was determined by EDS using the same instrument. The results are shown in Figure a–f and Table . Upon high-temperature pyrolysis, obvious porous structures could be detected on BC and KBC (Figure c–f. The EDS analyses of BR, BC, and KBC are shown in Table . C and O are the main elements, and also there is a small amount of N and Ca elements. Furthermore, the analyses demonstrated an increased C content and decreased O content on BC and KBC, upon high-temperature pyrolysis, which could result from the loss of O-including functional groups and water during pyrolysis.[32]

Figure 2

SEM images of the raw BR (a, b), pyrolyzed BC (c, d), and modified pyrolyzed KBC (e, f).

Table 2

EDS Elemental Analysis of the BR, BC, and KBC Samples

element (wt %)	BR	BC	KBC
C	53.29	86.68	82.91
N	2.58	1.54	1.58
O	40.76	7.42	8.84
Al	0.06	0.08	0.16
Ca	2.98	3.96	3.03
K	0.34	0.32	3.48

FTIR

Figure a illustrates the FT-IR spectra of BR, BC, and KBC. The broad peak near 3380 cm–1 originating from O–H stretching of the lignin, cellulose, and hemicellulose content of BR[33] was decreased on BC and KBC, which could be ascribed to the dehydrogenation and dehydration effect of high-temperature treatment. Meanwhile, upon high-temperature pyrolysis, weak intensity occurred on C–H stretching of 2891 cm–1 in the hemicellulose backbone[34] and carbonyl groups stretching of 1680 cm–1 in hemicellulose, which indicated the high temperature and KOH activation process decomposed part of the hemicellulose in BR.[35] Compared to BC, KOH activation decreased the intensity of C–C at 1425 cm–1, and C–O at 1046 cm–1, which could be ascribed to the reaction between K2O and C under high temperature. However, although the number of functional groups of KBC was reduced, some organic functional groups were retained on the surface.[36]

Figure 3

FTIR spectra of BR, BC, KBC, and KBC + TC (a) and the XRD patterns of BC and KBC (b).

XRD

Figure b presents the XRD patterns of BC and KBC. The distinct broad diffraction peak at 23.69° in both patterns indicates the presence of graphite-like structures (amorphous carbon) in both biochars, while the characteristic diffraction peak at 43.16° indicates the formation of crystalline graphitic structures.[37] The intensity of the latter diffraction peak is weaker in KBC than in BC, indicating that BC is slightly more graphitized than KBC, which may be the result of the destruction of some functional groups by KOH. The characteristic peaks at 29.34°, 35.95°, 37.34°, 39.42°, 47.34°, and 48.47° seen in the BC pattern correspond to CaCO3 crystallites. After modification by high-temperature KOH and acid washing, most of these peaks disappeared, which indicated that CaCO3 was removed.

Impact of Adsorbent Dosage

As shown in an adsorption system, the adsorbent dosage plays a vital role determining the adsorbent-adsorbent equilibrium. Figure presents the impact of varying the dosage of BC and KBC on TC removal efficiency as well as adsorption capacity. Although the adsorption capacity of BC for tetracycline was limited, the removal efficiency progressively enhanced with increasing adsorbent dosage. At any given dosage, KBC had a higher removal efficiency and larger adsorption capacity in comparison with BC, indicating that KOH activation could obviously improve the TC adsorption performance of the adsorbent. With the dosage being enhanced from 0.25 g/L to 1.25 g/L, the removal efficiency of TC was initially enhanced and subsequently lowered. The initial increase could result from the increase in functional groups and effective contact area with increasing dosage, while the decrease at high dosage was due to the decrease in the effective contact surface of TC because of the overlapping of active sites. The highest adsorption of TC was achieved based on a KBC dosage of 0.5 g/L. For subsequent experiments, this level was maintained.

Figure 4

Impact of BC and KBC adsorbent dosage on removal efficiency and adsorption capacity of TC.

Effect of pH

Figure a shows the impact of pH on the adsorption and removal behavior of BC and KBC. The adsorption of TC by KBC reached the maximum value of 380.75 mg/g at pH 2 and subsequently reduced slightly at higher pH. TC refers to an amphoteric molecule which has three pKa values (3.3, 7.7 and 9.7) and, thereby, will occur in four different forms at different pHs: the cation H4TC+ (pH < 3.4), the neutral molecule H3TC0 (3.4 < pH < 7.6), the anion H2TC– (7.6 < pH < 9.7), and the double anion HTC (pH > 9.7).[38] In addition, the measured pHPZC values of BC and KBC were 3.80 and 3.38, respectively. According to Figure b, the surface charge of the absorbents is therefore positively charged with solution pH < pHPZC and negatively charged with solution pH > pHPZC.[39] At 3.4 < pH < 7.6, H3TC0 is the predominant form, KBC carries a slightly negative charge, and the electrostatic interaction and repulsion forces reach a minimum value, so there exists no significant change in the TC adsorption capacity of KBC at pH values of 4–7. At pH 2.3, TC molecules exist in cationic form, and the KBC surface also has a positive charge. As a result, the high TC adsorption capacity of KBC at this pH despite the electrostatic repulsion suggests other interactions between KBC and TC molecules, namely, strong hydrogen bonding and π–π EDA interactions.[40] And the slight decrease in the TC adsorption capacity of KBC at pH > 7 can be attributed to both hydrogen bonding and π–π EDA interaction forces being weakened under alkaline conditions, in addition to the presence of strong electrostatic repulsion between the negatively charged TC and BC surfaces. In addition, BC displays similar behavior, having the highest adsorption at pH 2–3, although, overall, the adsorption capacity of BC is much lower when compared with that of KBC.

Figure 5

(a) Impact of solution pH on the adsorption of TC on BC and KBC, and (b) the variation of zeta potential of BC and KBC with solution pH.

Impact of Adsorption Time and the Adsorption Kinetics

Figure a shows the variation in TC adsorption on KBC with TC concentration. The trends are broadly TC concentration-independent, showing fast and then slow adsorption and eventually reaching the adsorption equilibrium. During the initial stage of adsorption, KBC is filled with a large number of adsorbable sites, and TC is rapidly adsorbed onto the surface of KBC; however, as the adsorption proceeds, the number of free adsorption sites progressively decreases with the surface adsorption reaching saturation; thereafter, the TC molecules migrate from the surface to the interior, and the rate of adsorption gradually decreases due to increasing diffusion resistance. When all of the adsorption sites of the biochar are occupied, the adsorption reaches equilibrium.[41]

Figure 6

(a) Influence of contact time on TC adsorption of KBC and model fits to the experimental data: pseudo-first-order (b), pseudo-second-order (c), Elovich (d), and intraparticle diffusion (e).

The adsorption kinetics data were fitted using quasi primary, quasi secondary, and Elovich kinetic models. The relevant models are shown below:[42]where q refers to the adsorption amount at time t (mg/g), qe indicates the adsorption amount at equilibrium (mg/g), k1, k2 are the pseudo-first-order and pseudo-second-order reaction rate constants, and α, β stand for the initial adsorption rate constants and parameters associated with the activation energy during the adsorption process, separately.

The fitted curves and extracted parameters of the three adsorption models are shown in Figure b–d and Table . For the purpose of assessing the fitting level of the adsorption model, the determination coefficient (R2) could be adopted. As the initial TC concentration remained unchanged, a greater R2 could be attained in pseudo-second-order and Elovich kinetic models, demonstrating they fit TC adsorption kinetics, which is consistent with the findings of former research studies.[43,44] Pseudo-second-order kinetics are consistent with adsorption via pore filling, covalent bond formation, and electron exchange, while the Elovich model supposes that the adsorbent surface can be energetically heterogeneous and that chemisorption interactions occur.[45] Therefore, the adsorption of TC by KBC proceeds via both electrostatic interactions.

Table 3

Extracted Kinetic Parameters of TC Adsorption on KBC

initial TC concentration C0 (mg/L)	150	200	250
qexp (mg/g)	247.11	298.70	350.30
pseudo-first-order
qcal (mg/g)	221.96	258.59	313.32
k1 (min–1)	0.0631	0.0873	0.0984
R2	0.913	0.909	0.931
pseudo-second-order
qcal (mg/g)	235.75	272.21	328.16
k2 (g·min–1·mg–1)	4.08 × 10–4	5.17 × 10–4	5.07 × 10–4
R2	0.975	0.962	0.976
Elovich
α (g·min–1·mg–1)	87.69	129.07	174.62
β (g/mg)	0.0418	0.0422	0.0393
R2	0.976	0.991	0.979

The particle diffusion model was performed with the purpose of further exploring the TC adsorption mechanism of KBC. The equation for this model is[46]where kid indicates the intraparticle diffusion rate constant (mg·g·min–1/2), and C refers to the boundary layer constant.

Figure e shows the results of the fit. The curve consists of three linear elements, suggesting that the adsorption process is categorized into the following three stages: (1) an external diffusion stage, where TC is transferred to the surface of KBC through the water film; (2) an internal diffusion stage, where TC molecules enter the internal surface through the pores of KBC; and (3) an adsorption equilibrium stage, where the adsorption of TC by KBC reaches saturation. Since none of the linear curves crossed the origin, apart from intraparticle diffusion, other processes were possibly involved into TC adsorption.[47]

Impact of Initial TC Concentration and the Adsorption Isotherms

As Figure displayed the adsorption capacity under different equilibrium concentration, the adsorbed amount of TC increased with its increased concentration in solution under an equilibrium state. With the further increase on TC equilibrium concentration than 100 mg/L, TC adsorption capacity was slightly increased, which could be ascribed to the saturation of adsorption sites on biochar. Langmuir and Freundlich isotherm was adopted for assessing the adsorption capacity of biochar on TC, as illustrated in Figure . Besides, Table shows the extracted fitting parameters. Equations and 8 show the Langmuir and Freundlich models, respectively:[48]where qe denotes the adsorption capacity at equilibrium adsorption, Ce refers to the equilibrium concentration of TC, qm is the maximum adsorption capacity (mg/g), n represents the heterogeneity factor, KL suggests the Langmuir constant (L/mg), and KF means the Freundlich constant (mg/g).

Figure 7

Langmuir (a) and Freundlich (b) isotherm model fits for TC adsorption of KBC.

Table 4

Extracted Fitting Parameters of the Langmuir and Freundlich Isotherm Models of TC on KBC

		Langmuir model				Freundlich model
sample	temperature (K)	qm (mg/g)	KL (L/mg)	K (dimensionless)	R2	KF (mg1–1/n·g–1·L–1/n)	1/n	R2
KBC	298	473.66	0.0286	705602.6	0.992	43.99	0.4418	0.952
	308	631.86	0.0263	648858.3	0.996	57.94	0.4426	0.946
	318	830.78	0.0354	873368.3	0.994	91.35	0.4277	0.944

Adsorption capacity at equilibrium at first increased with enhancing the initial TC concentration—which may be caused by the enhancement in concentration leading to a corresponding increase in driving force[49]—before leveling off. The Langmuir model proved a better fit for the observed adsorption of KBC on TC, suggesting that the adsorption of KBC on TC is mostly monolayer adsorption.[50] Monolayer adsorption from the Langmuir model was positively associated with the specific surface areas and developed pore structure, and such a result could explain the high performance of KBC on TC adsorption, with the maximum adsorption of 830.78 mg/g at 318 k. It was also seen that the values of 1/n in the Freundlich model fit are less than 0.5, showing that KBC can easily adsorb TC.[51] It should also be noted here that KBC shows better TC adsorption than many other adsorbents described in the literature (Table ).

Table 5

Comparison of the TC Adsorption Capacities of KBC as well as Other Biomass-Based Adsorbents Reported

adsorbent	adsorption conditions	Qmax (mg/g)	ref
wheat straw derived biochar	45 °C	584.19	(29)
Eucommia ulmoides lignin-based biochar	55 °C, pH 6	1163	(52)
magnetic sludge biochar	0.2 g/L, 24 h, 35 °C, pH 6	145	(48)
agricultural biochar	5 g/L, 6 h, 25 °C, pH 7	9.45	(47)
rice straw derived biochar	1.2 g/L, 1 h, 25 °C	98.33	(53)
leaf-derived biochar	0.1 g/L, 5 days, 20 °C, pH 6	361	(42)
manganese dioxide modified biochar	0.8 g/L, 3 days, 35 °C, pH 3	131.49	(54)
reed-based biochar	24 h, 40 °C, pH 6	173.61	(36)
alfalfa-derived biochar	0.1 g/L, 5 days, 20 °C, pH 5	302.37	(44)

Impact of Temperature and a Study of the Thermodynamics

The parameters of the thermodynamic processes involved in TC adsorption on KBC can be defined and calculated based on the subsequent equation.[55]where Kd is the distribution coefficient, ΔG0 refers to the change in Gibbs free energy (kJ·mol–1), ΔS0 refers to the change in entropy (J·mol–1·K–1), ΔH0 suggests the change in enthalpy (kJ·mol–1), the absolute temperature (K), and R represents the universal gas constant (8.314J·mol–1·K–1).

Figure shows the plot of ln Kd versus 1/T. ΔH0 and ΔS0 are computed from the slope and intercept of the linear plot, allowing ΔG0 to be calculated according to eq . The extracted parameters are shown in Table . On the basis of Table , ΔG0 is below zero at all temperatures, suggesting that the adsorption process of KBC on TC is spontaneous. As ΔG0 decreases with increasing temperatures, this suggests that heat enhances the spontaneity of the adsorption. It can also be seen that ΔH0 is greater than zero, implying that the adsorption process can be an endothermic reaction. Apart from that, the positive ΔS0 value is also favorable.

Figure 8

Thermodynamics of the adsorption of TC on KBC.

Table 6

Extracted Adsorption Thermodynamic Parameters for TC Adsorption on KBC

sample	T (K)	ΔG0 (kJ·mol–1)	ΔH0 (kJ·mol–1)	ΔS0 (J·mol–1·K–1)
KBC	298	–4.751	35.864	0.136
	303	–5.568
	308	–6.114
	318	–7.477

Recyclability of the Adsorbent

The regeneration performance of the prepared adsorbent was assessed using adsorption–desorption experiments across several cycles. Figure shows the adsorption performance of the adsorbents across four adsorption cycles. With the increase of cycle times, the adsorption capacity of KBC for TC decreased from 196.67 mg/g to 156.99 mg/g, BC for TC decreased from 76.01 mg/g to 63.29 mg/g, which could be ascribed to the residual TC on biochar that could not be desorbed by NaOH and deionized water. TC adsorption capacity and removal rate remained high after four cycles, demonstrating the great regeneration performance of KBC on TC adsorption.

Figure 9

Regenerative properties of BC and KBC.

Adsorption Mechanism(s)

Both the functional groups and the physical structure of the adsorbent surface generate obvious impacts on the adsorption of TC. The analysis of the impact of different pH values on TC adsorption by KBC demonstrated that the electrostatic effect was negligible, which indicates that TC adsorption does not proceed via an electrostatic effect. Instead, we propose that a combination of three alternative mechanisms is responsible for TC adsorption by KBC—the detailed mechanism of action is shown in Figure . The first contributing mechanism is pore adsorption—it has been reported[56] that carbonaceous materials with high specific surface area and high pore volume stimulate the adsorption of organic matter because of their obvious pore filling effect. Thus, the carbonized, high-temperature KOH-activated KBC with its well-developed micro- and mesoporous structures and relatively high surface area could remove TC molecules by physical diffusion. The second contributing mechanism is hydrogen bonding. On the basis of the analysis of FTIR spectra, the 3427 cm–1 peak of KBC was enhanced after adsorption, and the position of the C–H peak at 882 cm–1 shifted, both of which are an indication of the removal of TC molecules by hydrogen bonding between KBC and functional groups in the TC molecules. The final contributing mechanism is the π–πEDA stacking interaction. The XRD analysis showed that KBC has a graphitized structure rich in π electrons, which may interact with the aromatic and amino structures in TC molecules by π–π EDA.[57]

Figure 10

Schematic diagram of the combination of three mechanisms responsible for the adsorption of TC on KBC.

Conclusions

In the present study, Caulis spatholobi residue was used to prepare biochar. High-temperature KOH activation was conducted to improve biochar with more developed pore structures and greater specific surface area, which could be verified by BET analysis, SEM images, EDS elementary analysis, FTIR, and XRD. pH displayed a slight influence on tetracycline adsorption, which signified the slight influence of the existence of tetracycline and the charge potential of biochar. Accordingly, TC adsorption on biochar possibly originated from pore filling, hydrogen bonding, and π–π EDA stacking interaction. TC adsorption kinetics was better fitted by pseudo-second-order and Elovich kinetic models, demonstrating the involvement of chemisorption and physisorption in TC adsorption. Relative to the Freundlich isotherm model, the Langmuir isotherm model better matched TC adsorption, demonstrating the relation between TC adsorption and the specific surface areas of biochar, as well as the developed pore structure. Besides, the maximum adsorption capacity of KBC on TC was 830.78 mg/g at 318 K. The optimal pH for TC adsorption was 2. As displayed by thermodynamic analysis, TC adsorption on KBC was an endothermic reaction process. KBC also showed good thermal stability and regeneration performance.