FT-IR spectroscopyIn this study, the chemical structure of each component involved in the preparation of the adsorbent and the adsorption process were investigated using FT-IR spectroscopy, as depicted in Fig. 1. The broad peak observed at 3439.45 cm⁻1 in spectrum 1a corresponds to the stretching vibration of OH bonds. This phenomenon is attributed to the presence of carboxylic acid and alcohol groups in cellulose or residual water50. Additionally, the peak observed in the 1614 cm⁻1 region is associated with the stretching vibration of C=O bonds, specifically for carbonyl and carboxylic acid groups. Furthermore, the band observed at 1449.5 cm⁻1 is related to the stretching vibration of C–C bonds due to the presence of aromatic rings50. After the pyrolysis process and subsequent activation, certain bonds were rearranged, leading to the formation of new functional groups on the surface of the PFRC-A adsorbent. Notably, the stretching vibration of S=O bonds at 1043 cm⁻1 provides clear evidence of sulfuric acid modification51. Finally, upon adsorption of TC and PC onto the adsorbent, the initial peak positions and intensities in the FTIR spectrum underwent changes, confirming the successful adsorption process50.Figure 1The FT-IR spectra each part of the preparation of adsorbent and adsorption process.BET measurementsThe BET method is employed to determine the specific surface area of materials. This technique relies on measuring the amount of nitrogen gas adsorbed within a relative pressure range of 0.1 to 1. Additionally, the pore size distribution is determined through the use of adsorption isotherms. In Fig. 2, the adsorption and desorption isotherms for two adsorbents are depicted. According to the IUPAC classification, these isotherms fall into Type IV (adsorption) and Type IV (desorption). Based on this classification, the pores exhibit a layered and sheet-like structure. Giving that Table 1, the specific surface area for carbon and m carbon adsorbents is equal to 195.5,182.5, and 224.6 m2/g, respectively. Also, in Table 1, the specifications of the absorbent surface are given. As can be seen, with chemical activation of the absorbent, the active sites have been improved and the available surface has increased.Figure 2The N2 gas adsorption–desorption isotherm of (a) carbon, (b) m-carbon, and (c) carbon acid.Table 1 Results of the BET measurements of adsorbent.SEM analysisAs seen in the SEM images of MPFRC-A before adsorption (Fig. 3), the adsorbent’s surface appears relatively smooth, with varying pore sizes. However, after the adsorption of TC and PC, the surface becomes rougher, and the pore dimensions decrease. This reduction in pore size suggests the successful adsorption of the pharmaceutical compounds onto the adsorbent material’s surface.Figure 3SEM images of (a,b) MPFRC-A, (c,d) MPFR-A/TC, (e,f) MPFRC-A/PC.XRD analysisThe observed patterns in several prior studies indicate the amorphous nature of MPFRC-A (2θ = 15.8° to 22.8°). This amorphous character is attributed to the presence of organic materials, such as hydroxyl and carboxyl groups50. Upon activation and thermal decomposition, the amorphous nature transforms due to the presence of elements like graphite, calcium, and silica, resulting in a semi-crystalline structure50,51. Typically, the activation of the adsorbent involves acid treatment and high temperatures, leading to the breakdown of functional groups within cellulose and hemicellulose, ultimately forming graphite. The sharp peaks observed at 2θ = 29.38° and 43.12° in Fig. 4a correspond to the presence of graphite, silica, and the amorphous carbon nature, primarily due to the low graphite content50. Also, XRD pattern of the synthesized MPFRC-A shows significant reflection peaks are found at 2θ = 30.09, 35.28, 43.02, 53.62, 56.91, and 62.73. When compared to the reference (JCPDS Card No. 19-0629), the peaks were confirmed to be magnetite nanoparticles. The peaks correspond to respective phase planes of (220), (311), (400), (422), (511), and (440)42. Also, the proposed structure for MPFRC-A was also confirmed by EDX analysis. As expected, the EDX measurement proves the presence of C, O, Fe, and S elements (Fig. 4b).Figure 4The (a) XRD patterns and (b) EDX analyses of MPFRC-A.VSM analysisThe VSM analysis of MPFRC-A was performed in order to demonstrate the magnetic property (Fig. 5). As can be seen, the compound has magnetic property and shows a nice increase (35, emu/g), and probable reduction in the dipolar–dipolar interactions between the magnetic nanoparticles after mixing with activated carbon.Figure 5pHThe pH of the point where the desired molecule or particle has no surface electric charge is called the pHpzc or isoelectric point. Identifying the isoelectric point is essential for carrying out the adsorption process, because the creation of attractive or repulsive force between particles depends on this point. The determination of the isoelectric point of the adsorbents was determined by the drift method52. For this purpose, the pH of each of the four erlens, in which 20 mL of distilled water was poured into each one, and with the use of 0.1 M NaOH and 0.01 M HCl, was determined in one of pH 2, 4, 6 and 8 are adjusted with a pH meter. Then 0.5 g of adsorbent was added to them and the solution was stirred in a shaker for 24 h, the pH of the solution was measured again and the graph of secondary pH changes of the solutions was drawn according to the initial pH. The intersection of this graph with the line y = x shows the isoelectric point and was determined according to it. Figure 6 shows the results of secondary pH in terms of initial pH. According to the obtained results, the isoelectric point of the adsorbent was found to be equal to 5. The surface of the adsorbent at a pH higher than this value has a negative charge, and the surface of the adsorbent at a pH lower than this value has a positive charge.Figure 6Isoelectric point of adsorbent by DRIFT method.The pH of a solution is one of the most influential factors in the adsorption process and the mechanism of electrostatic forces. To determine the optimal pH for the adsorption of PC and TC, each drug was individually tested at a concentration of 20 mg/L, using 0.4 g/L of the adsorbent within the pH range of 3 to 8. Considering the pka values of the drugs and the isoelectric point of the investigated adsorbent (which is 5), we observe that at pH values below the isoelectric point, the surface charge of the adsorbent becomes positive, while at pH values above the isoelectric point, the surface charge becomes negative. The results of the pH effect on adsorption capacity are illustrated in Fig. 7. For PC pka is equal to 9.51. Also, the pHpzc of the adsorbent is equal to 5, as a result, at pHs greater than 9, it is negative with both pollutant and adsorbent, and at pH less than 5, it is positive with both pollutant and adsorbent, as a result, in this pH pollutant and adsorbent repel each other and the adsorption capacity is low. At pH between 5 and 9 pollutant and adsorbent are not synonymous and attract each other. As a result, at pH equal to 6, the highest adsorption capacity is observed. TC has three pka of 3.30, 7.68 and 9.68. TC is an organic molecule whose protonated or deprotonated form depends on the pH of the solution. TC appears at pH less than 3.3 due to the protonation of amine groups in a cationic form with two positive charges, which is why According to the surface charge of the adsorbent, the pollutant and the adsorbent both have a positive charge and repel each other. At pH 3–7, one end of TC has a positive charge and the other end has a negative charge, and the highest amount of removal occurs at this pH. And finally, at a pH higher than 9, due to the loss of a proton from the carboxylic group in the structure of the antibiotic, it appears in a completely anionic form with two negative charges, at this pH, according to the surface charge of the adsorbent, pollutant and adsorbent of each the two have a negative charge and repel each other. As a result, the highest adsorption capacity occurs at pH 6 and 5 for PC and TC drugs. Considering that pH change affects the adsorption capacity of two drugs, it can be said that electrostatic forces are involved in the adsorption process.Figure 7The effect of pH on the adsorption capacity of PC and TC (Initial concentration 20 mg/L, m = 0.4 g, V = 35 ml and T = 25 °C).Dosage of adsorbentTo determine the optimal amount of adsorbent, different amounts of adsorbent 0.4, 0.6, 0.8, 1, 1.5 and 2 g/L of solution with an initial concentration of 20 mg/L, pH equal to 4 and 5, and a duration of 120 min were investigated. As can be seen in Fig. 8a, the adsorption capacity raises with an increase in the adsorbent amount from 0 to 0.4 g/L due to the availability of unsaturated adsorption sites. However, by increasing the adsorbent amount from 0.4 to 2 g/L, the adsorption capacity for both drugs decreases due to the agglomeration and clumping of the adsorbent, reducing the available active surface area. Therefore, an adsorbent amount of 0.4 g/L was considered optimal for further experiments.Figure 8(a) The effect of adsorbent dosage on absorption capacity of PC and TC (initial concentration of 20 mg/l, pH equal to 4 and 5, duration of 120 min and ambient temperature), (b) The effect of temperature on absorption rate of PC and TC (initial concentration 20 mg/l, pH equal to 4 and 5, duration of 120 min and ambient temperature, amount of adsorbent 0.4 g).Also, with the increase in temperature, the speed of movement of drug molecules in the solution increases (Fig. 8b). As a result, the energy of the random collision speed increases and causes the absorption capacity to decrease. Therefore, an increase in temperature breaks the bonds created between the absorbent and the pollutant and disposal occurs. According to the Fig. 8, the maximum absorption capacity is related to the ambient temperature, so the temperature of 25 degrees Celsius was chosen to continue the experiments.Thermodynamic of adsorptionEnthalpy and entropy may be calculated from the Van’t Hoff curve by researching how temperature affects adsorption capacity (Fig. 9)53,54.Figure 9Table 2 lists the outcomes of the adsorption thermodynamics drawn from the Van’t Hoff diagram. Table 2 shows that ∆G° was positive, indicates the non-spontaneity of the reaction. Throughout the procedure, irregularities in terms of the positive values of ∆S° increased. Based on the value of ∆H°, which was positive, the adsorption process by all adsorbents was endothermic. Moreover, based on the references, and values of ∆H°, which were elder than − 20 kJ, it can be concluded that the adsorption band was physical55,56.Table 2 Thermodynamic parameters of the adsorption process.Effect of contact time and kineticsIn our study, we investigated the adsorption kinetics of TC and PC to understand the impact of contact time on their adsorption behavior. The initial solutions were prepared with a concentration of 20 mg/L, and 0.4 g/L of each adsorbent was added. To ensure that the adsorption process reached equilibrium, the samples were agitated for 2 h in a shaker operating at 250 rpm and maintained at a temperature of 25 °C. The results of this investigation shed light on the dynamic behavior of TC and PC adsorption, providing valuable insights for environmental and pharmaceutical applications.The obtained data were fitted to pseudo-first-order, pseudo-second-order, and intraparticle diffusion models, with the results summarized in Table 3. Based on the correlation coefficient (R2) for each model, the pseudo-second-order model is deemed acceptable for the adsorption of TC and PC. The pseudo-second-order kinetic model operates under the assumption that the rate-limiting step involves ion exchange. Additionally, the rate constants for the adsorption of TC and PC are estimated to be 0.116 g/mg.min and 0.096 g/mg.min, respectively. The investigation of intraparticle diffusion kinetics provides valuable insights into the transport of solute species within porous adsorbents. Notably, the rate constant for intraparticle diffusion falls within the range of 3.21 to 3.78, while the constant C signifies the boundary layer thickness and characterizes the external mass transfer process. Furthermore, the pronounced increase in adsorption capacity during the initial stages of the process is attributed to the phenomenon of rapid mass transfer.Table 3 Constants and coefficients related to adsorption kinetic models.As shown in Fig. 10, in the early times, the slope of the graph is very steep (the rapid phase of external mass transfer) and as the equilibrium conditions are approached, the speed of the adsorption process slows down (the slope of the graph decreases and reaches zero) until the adsorbent does not have more pollutant adsorption power and the adsorption graph is fixed57,58. Also, after about 30 min, the adsorption graph stabilizes and the adsorption reaches equilibrium, however, to ensure the achievement of complete equilibrium, the studied time period was continued up to 120 min.Figure 10The pseudo-second-order kinetic fitting for adsorption of PC and TC (Initial concentration 20 mg/L, pH = 5 and 6, T = 25 °C).Adsorption isothermIn the context of adsorption processes, quantifying the removal efficiency of pollutants from aqueous solutions is of paramount importance. To achieve this, we employ adsorption isotherm models, which explore the surface properties and affinity of the adsorbent toward the adsorption process. In this study, we investigated the adsorption behavior of TC and PC using the Langmuir, Freundlich, and Temkin isotherm models, as well as the three-parameter Redlich–Peterson model. The Langmuir model had the most overlap with the experimental results of TC and PC adsorption. This model describes the adsorption process in terms of a monolayer at equilibrium. The maximum adsorption capacity (qm) represents the adsorption capacity of a single layer. For TC and PC, the Langmuir model yielded high R2 = 0.99. The Langmuir constants KL characterizes the adsorbent’s affinity for the pollutant. The calculated values of qm for TC and PC were 43.75 mg/g and 41.70 mg/g, respectively. By comparing the experimental data with the Langmuir isotherm, we observed excellent agreement between the model predictions and the measured adsorption capacities (as shown in Table 4). In summary, the Langmuir isotherm provides a robust framework for understanding the adsorption behavior of TC and PC, offering valuable insights for environmental remediation and pharmaceutical wastewater treatment.Table 4 Constants and coefficients of adsorption isotherm models for PC and TC.The Freundlich isotherm, which characterizes heterogeneous surface adsorption, has been employed to investigate the interaction between the adsorbent and the pharmaceutical compounds. As shown in Table 3, the R2 for TC and PC adsorption using the Freundlich model is 0.95 and 0.91, respectively. Notably, these values are lower than those obtained from the Langmuir isotherm. This discrepancy underscores the limitations of the Freundlich model in describing the adsorption behavior of the aforementioned drugs. The parameter n in the Freundlich isotherm reflects the favorability of the adsorption process. For the range of n values between 0 and 10, the experimental data in Table 4 confirm favorable adsorption behavior. In summary, while the Freundlich isotherm provides insights into surface heterogeneity, the Langmuir model remains a more robust choice for describing the adsorption performance of TC and PC.The Temkin isotherm model postulates that the heat of adsorption decreases linearly with coverage. Additionally, it assumes that the adsorption process is characterized by a uniform distribution of cohesive binding energies. According to the data presented in Table 4, the heat of adsorption follows a linear trend with decreasing surface coverage. The parameter AT represents the equilibrium binding energy, corresponding to the maximum bond energy, while BT pertains to the heat of adsorption. A low value of BT indicates weak interactions between the adsorbate molecules and the adsorbent surface.Hill’s isotherm model, in this model, it is a cooperative phenomenon where the ligand is attached in one place on the macromolecules. In this case, it may affect different binding sites on the same macromolecules. qH is the adsorption capacity of the Hill isotherm, nH is the Hill binding interaction coefficient, and KH is the Hill constant.Among the three-parameter adsorption isotherm models, the Redlich–Peterson model is frequently employed for liquid-phase adsorption of organic compounds. Based on the obtained correlation coefficients, both the Langmuir isotherm model and the Redlich–Peterson model outperformed the two-parameter models. This observation suggests that the adsorbent surface becomes more closed during surface adsorption, leading to a reduction in the influence of external factors on drug adsorption. According to the results, the adsorption of pharmaceutical pollutants on the researched adsorbent is of single layer type. The adsorbent structure is homogeneous and has the same adsorption energy. Also, adsorption has been done physically. Considering that the value of β in the Redlich–Peterson equation is close to 1, the Langmuir model is more consistent with the experimental equilibrium data. Figure 11 shows the fit of the Langmuir model curve.Figure 11The Langmuir isotherm model fitting for the adsorption of PC and TC (Time = 120 min, pH = 5 and 6, T = 25 °C).MechanismThe adsorption mechanism is evaluated according to the adsorbent properties, such as pores, functional groups, surface area, and also pH value. It is suggested dominant interactions could be electrostatic attraction, and Ion exchange between functional groups of Tetracycline and Paracetamol, and the protonated surface area of adsorbents in acidic media. Also, bonding between pharmaceutical compounds and carbonaceous material on the surface could be occurred in electron donor–acceptor hydrogen, and π-hydrogen interaction59.RecoveryAdsorbent recovery for reuse is important from an economic perspective (Fig. 12). Although the price of synthesized adsorbents is higher, but due to their high absorption capacity, they can be used in smaller amounts than other mentioned adsorbents, which will bring a higher economic value. To recover the adsorbents, after each use, we add 5 cc of ethanol to the adsorbent, it is placed on the shaker for 20 min, then it is separated with a magnet, and in the next step, 5 cc of the buffer prepared in the ratio (1:10) is added to it and placed on the shaker for 20 min and finally separated by a magnet. According to Fig. 12, the synthesized adsorbent for PC and TC drugs can be used 6 times with a removal efficiency of 71.6 and 66.4% (Fig. 13).Figure 12The adsorption mechanism of TC and PC by MPFRC-A.Figure 13Plant collection statementWe confirm that our experimental research and field studies on plants, both cultivated and wild, strictly adhere to the relevant institutional, national, and international guidelines and legislation. The collection of plant material was conducted in accordance with these guidelines, ensuring ethical and responsible practices throughout our study.
