Sweep-Out of Tigecycline, Chlortetracycline, Oxytetracycline, and Doxycycline from Water by Carbon Nanoparticles Derived from Tissue Waste

Pharmaceutical pollution has pervaded many water resources all over the globe. The propagation of this health threat drew the researchers’ concern in seeking an efficient solution. This study introduced toilet paper waste as a precursor for carbon nanoparticles (CRNPs). The TEM results showed a particle size range of 30.2 nm to 48.1 nm, the BET surface area was 283 m2 g−1, and the XRD pattern indicated cubical-graphite crystals. The synthesized CRNPs were tested for removing tigecycline (TGCN), chlortetracycline (CTCN), oxytetracycline (OTCN), and doxycycline (DXCN) via the batch process. The adsorption equilibrium time for TGCN, DXCN, CTCN, and OTCN was 60 min, and the concentration influence revealed an adsorption capacity of 172.5, 200.1, 202.4, and 200.0 mg g−1, respectively. The sorption of the four drugs followed the PSFO, and the LFDM models indicated their high sorption affinity to the CRNPs. The adsorption of the four drugs fitted the multilayer FIM that supported the high-affinity claim. The removals of the four drugs were exothermic and spontaneous physisorption. The fabricated CRNPs possessed an excellent remediation efficiency for contaminated SW and GW; therefore, CRNPs are suggested for water remediation as low-cost sorbent.


Materials
Commercial toilet tissue papers were collected from the local market. Phosphoric acid 85% (H 3 PO 4 ) was from Merk, Darmstadt, Germany. The TGCN was provided by the Triveni Interchem in Mumbai, India. The DXCN, CTCN, and OTCN were brought from Fluka, Bushes, Switzerland. The TTPW was collected from the trash container beside a bathroom-hand-wash site as a common source for such waste.

Preparation of CRNPs
For this study, 20 g of toilet paper waste TPW were carbonized at 750 • C for 2.0 h under a nitrogen gas stream (50 mL min −1 ) as inert-environment. The CRMs were placed in an autoclave, macerated in concentrated phosphoric acid, and heated at 150 • C for 2.0 h. The CRMs were washed with distilled water via vacuum filtration to remove the excess acid, then dried at 110 • C for 4.0 h. Then, 5.0 g of the CRMs were transferred to a 50 mL stainless-steel crucible with 7.0 stainless-steel balls (1 cm diameter), and the ball mill machine was set at 500 rounds per minute for 10.0 h.

Adsorption Studies
The CRNPs have been tested for removing TGCN, DXCN, CTCN, and OTCN from the aquatic environment via the batch-experiment technique. The influence of contact time and the kinetic studies were performed by stirring 120 mL of drug solution (100 mg L −1 ) with 50 mg of the CRNPs. The picked portions were filtered via a 0.22 µm syringe filter; then, the absorbance was measured in triplicate using a Shimadzu-2600i UV-Vis spectrophotometer. The adsorption capacity (q t , mg g −1 ) was computed via Equation (5). The solution pH effect on the adsorption of TGCN, DXCN, CTCN, and OTCN on CRNPs was investigated. Additionally, 100 mg L −1 of each drug was adjusted to different pH from 2.0 to 10.0. Due to the crucial role of the solution's pH in the sorption process, recent studies investigated the zero charge point (pH pzc ) for sorbents. The sorbent surface is positively charged at a pH lower than the pH ZPC value and negatively charged above the pH ZPC . The solid addition method was employed for investigating the CRNP's pH pzc , and the point in which ∆pH = 0 is the pH pzc [67].
Furthermore, the contact time findings were employed to inspect the drug's adsorption kinetics. The pseudo-first-order and pseudo-second-order models (PFOM and PSOM) illustrated in Equations (6) and (7) were used to investigate the rate of sorption. In addition, the step controlling the adsorption was explored by the intra-particle and liquid film diffusion models (IDM and LDM, Equations (8) and (9)).
where: C o , C t (mg L −1 ) resents the drug concentrations at time zero, and t; V (L) and M (g) present the volume of drug solution and mass of CRNPs, respectively; k 1 (min −1 ): the PFOM rate constant; k 2 (g mg −1 min −1 ): rate-constant of the PSOM; K IDM (mg g −1 min −0.5 ) and K LDM (min −1 ): the IDM and LDM constants; the q e (mg g −1 ) is the equilibrium adsorption capacity; C i (mg g −1 ): the boundary-layer-thickness parameter.
The initial drug concentration effect on its sorption was tested utilizing 10, 25, 50, and 100 mg L −1 drug solutions. Additionally, the temperature influence investigation for the adsorption of TGCN, DXCN, CTCN, and OTCN on CRNPs was performed using the different concentrations at 298, 308, and 318 • K. The obtained results were employed to study the sorption's isotherms and thermodynamics.

Application of CRNPs to Natural Water Treatment
A groundwater sample (GS) was collected from Sudair industrial city (150 km north of Riyadh, Saudi Arabia), while the seawater sample (SS) was brought from the coast of Aldamam city, Saudi Arabia. A concentration of 5.0 and 10 mg L −1 from each drug was prepared in GS and SS, then 50 mg of the CRNPs were stirred with 120 mL of polluted water sample for 2.0 h. In order to avoid the impact of GS and SS, the standard solutions were prepared in the same matrix, and the water samples were employed as blank. The treated GS and SS samples were filtered via a membrane syringe filter (0.22 µm), and the absorbances of unabsorbed pollutants were determined by a UV-vis spectrophotometer.

Characterization of CRNPs
The SEM results in Figure 1a revealed the CRNPs were clustered due to the ballmilling process. In addition, the CRNPs appeared with a particle size range of 72.2 to 90.6 nm, indicating a successful fabrication of nanoscale carbon particles via the suggested method. Figure 1b,c show the fabricated nanomaterial's electronic image and the elemental composition results, respectively. The CRNPs were mainly composed of 90.6% carbon and 9.0% oxygen, while the rest of 0.4% comprised potassium and calcium traces. Additionally, the TEM analysis was employed to examine the detailed morphology, and the disunity of the CRNPs clusters revealed a particle size range of 30.2 to 48.1 nm (Figure 1d). These results were in line with the SEM results since the second may show clusters as one particle.
The XRD was utilized to analyze the crystallinity and phase purity of the synthesized CRNPs ( Figure 2a). The obtained results revealed diffraction peaks at 26.02 and 43.14 2θ • that can be assigned to the planes of cubical-lattice-graphite-phase of (002) and (100) [68]. Moreover, the baseline elevation around 2θ • of 10.0 indicates the occurrence of some amorphous carbon in the CRNPs. The most intense peak at 2θ • of 26.02 was employed for computing D, a, c, and ε values which were found to be 22.13 nm, 0.20 nm, 6.84 nm, and 0.38, respectively.
The surface characteristics of the synthesized CRNPs were determined via the N 2 adsorption-desorption method ( Figure 2b). The CRNPs possessed a type 2b hysteresis loop that corresponds to mesoporous materials of slit-like pores with narrow pore-neck ranges [69][70][71]. The surface area (SA) of the CRNPs was determined via Brunauer-Emmett-Teller's (BET) procedure, whereas the pore diameter and volume (PD and PV) were estimated by Barrett-Joyner-Halenda (BJH) way. The CRNPs showed PD, PV, and SA values of 79.14Ǻ, 0.58 cm 3 g −1 , and 283.04 m 2 g −1 , respectively; the obtained PD and PV seem suitable for the entrapment of pollutants. Additionally, these parameters were comparable to recent literature findings about CRMs [50,[72][73][74]. It is worth mentioning that almost 90% of these uptakes were acquired during the first 20 min of contact. Figure 3b illustrates the evaluation of pH impact on TGCN, DXCN, CTCN, and OTCN adsorptions, while Figure 3c shows the pH ZCH investigation for the fabricated CRNPs. The pH pzc was equal to 7.2 explains the attainment of higher q t values within the range of 6.0 to 8.0.    In addition, the impact of the concentrations of the drugs on their adsorption by CRNPs was tested. Figure 4 illustrates the increase of q t values as the drug concentration raised and reveals the capability of CRNPs to possess higher q t values. Typically, the adsorption from 100 mg L −1 TGCN, DXCN, CTCN, and OTCN solutions showed q t values of 172. 5, 200.1, 202.4, and 200.0 mg g −1, respectively. These findings inferred the applicability of the prepared CRNPs for treating industrial effluents with high pollutant concentrations. In addition, it can be concluded that the 4:5 sorbent: solution ratio is suitable for removing up to 100 mg L −1 pollutant concentration. The CRNPs also adsorbed the pollutants entirely from the 10 mg L −1 solutions, demonstrating their effectiveness in treating contaminated water resources.

Possible Adsorption Mechanism
Figure 5 monitored the chemical structure of TGCN, DXCN, CTCN, and OTCN. The heteroatoms on the drugs may contribute significantly to the adsorption through their lone pairs of electrons. The oxygen group on drugs withdraws electrons, leading to electron-deficient carbons acting as π-electron-acceptors; meanwhile, the high electron density on CRNPs oxygenated groups will be the π-electron-donor. The multi-carbonyl groups of these drugs would be protonated in acidic media (pH < 6.0), and the cationicformed sites may repel each other and prevent multilayer sorption. In contrast, in an alkaline media of (pH > 8), the hydroxyl groups may compete with the drugs on the sorbent sites, and the deprotonate hydroxyl groups of a drug may increase the repulsion [75,76].  Figure 5 monitored the chemical structure of TGCN, DXCN, CTCN, and OTCN. The heteroatoms on the drugs may contribute significantly to the adsorption through their lone pairs of electrons. The oxygen group on drugs withdraws electrons, leading to electrondeficient carbons acting as π-electron-acceptors; meanwhile, the high electron density on CRNPs oxygenated groups will be the π-electron-donor. The multi-carbonyl groups of these drugs would be protonated in acidic media (pH < 6.0), and the cationic-formed sites may repel each other and prevent multilayer sorption. In contrast, in an alkaline media of (pH > 8), the hydroxyl groups may compete with the drugs on the sorbent sites, and the deprotonate hydroxyl groups of a drug may increase the repulsion [75,76].

Adsorption Kinetics
The linear regression plots of the PSFOM, PSSOM, IDM, and LDM for TGCN, DXCN, CTCN, and OTCN sorption on the CRNPs are illustrated in Figure 6. The k 1 , k 2 , K IDM, and K LDM values gathered in Table 1 were computed utilizing the extracted regression parameters (slope and intercept) [77]. The results revealed that TGCN, DXCN, CTCN, and OTCN sorption fitted the PSFOM model, which may justify their relatively rapid equilibrium. Furthermore, the investigation of the rate-control step revealed that the intraparticle diffusion step controlled the adsorption of DXCN, CTCN, and OTCN on the CRNPs. Conversely, the film-diffusion step was the slowest step during TGCN adsorption. These findings indicated that TGCN has a higher affinity toward the CRNPs surface than

Adsorption Kinetics
The linear regression plots of the PSFOM, PSSOM, IDM, and LDM for TGCN, DXCN, CTCN, and OTCN sorption on the CRNPs are illustrated in Figure 6. The k1, k2, KIDM, and KLDM values gathered in Table 1 were computed utilizing the extracted regression parameters (slope and intercept) [77]. The results revealed that TGCN, DXCN, CTCN, and OTCN sorption fitted the PSFOM model, which may justify their relatively rapid equilibrium. Furthermore, the investigation of the rate-control step revealed that the intraparticle diffusion step controlled the adsorption of DXCN, CTCN, and OTCN on the CRNPs. Conversely, the film-diffusion step was the slowest step during TGCN adsorption. These findings indicated that TGCN has a higher affinity toward the CRNPs surface than the other three drugs and implied the significance of pore-diffusion in removing DXCN, CTCN, and OTCN [78].
ln q e = ln K f + 1 n ln C e (11) where K l (L mg −1 ) is LIM constant, K f (L mg −1 ) is FIM constant; C e (mg L −1 ) is the equilibrium solution concentration, q m is the computed-maximum-sorption capacity, while the 1/n is Freundlich-adsorption-intensity [79]. Figure 7

Adsorption Thermodynamics
The thermodynamics of TGCN, DXCN, CTCN, and OTCN removal by the CRNPs were inspected. The slope and intercept extracted from the plot of Equation (12) (Figure 8) were utilized in computing the enthalpy and entropy (ΔS° and ΔH°). The Gibbs free energy (ΔG°) was obtained by applying the ΔS° and ΔH° values in Equation (13). The ideal gas constant (R) was used as 0.0081345 kJ mol −1 for calculating these parameters, and the findings are in Table 2 The findings monitored in Table 2 revealed better fitting to LIM. The 1/n values of less than unity for the four drugs indicated that their sorption were favorable [80]. Furthermore, the fitting of TGCN, DXCN, CTCN, and OTCN sorption to the LIM was in line with their PFOM agreement [81,82].

Adsorption Thermodynamics
The thermodynamics of TGCN, DXCN, CTCN, and OTCN removal by the CRNPs were inspected. The slope and intercept extracted from the plot of Equation (12) (Figure 8) were utilized in computing the enthalpy and entropy (∆S • and ∆H • ). The Gibbs free energy (∆G • ) was obtained by applying the ∆S • and ∆H • values in Equation (13). The ideal gas constant (R) was used as 0.0081345 kJ mol −1 for calculating these parameters, and the findings are in Table 2.

Natural Water Treatment and Regeneration Investigations
The CRNPs were tested for removing TGCN, DXCN, CTCN, and OTCN from synthetically polluted GS and SS. The average removal percentage for TGCN, DXCN, CTCN, and OTCN were 98.9%, 94.3%, 98.4%, and 99.1%, respectively, with RSD values of 0.894%, 7.192%, 1.413%, and 0.245%, respectively. Although the fabricated CRNPs were excellent in removing these pollutants from both water types (Figure 9a), it can be seen that the remediation of GS was better than the SS, which indicates that the CRNPs can be employed for the sorption of metal ions. This result can be attributed to salt concentration (mainly saline) in the SS, which may affect the pollutant's migration to the CRNPs surface.
Additionally, the reusability of the CRNPs in removing TGCN, DXCN, CTCN, and OTCN was investigated. According to the pH study, the CRNPs used with DXCN and OTCN were regenerated by sonication for 20 min with 20 mL of 1.0 M sodium hydroxide (NaOH). The sonication was repeated with 10 mL ethanol, then filtered, rinsed with distilled water, and dried at 105 • C for 2.0 h. On the other hand, CRNPs used with TGCN and CTCN were regenerated in the same manner but substituting the 1M NaOH with 1M hydrochloric acid. The performance of virgin CRNPs was considered 100% efficient, and the subsequent cycles were relatively determined (Figure 9b) [51,88]. The regenerated CRNPs showed mean efficiency of 93.40%, 95.31%, 95.49%, and 95.94% for TGCN, DXCN, CTCN, and OTCN, respectively, with RSD% values of 5. 71, 3.78, 4.42, and 4.46. According to the kinetic studies, the LFDM was the slowest sorption step indicating the ease of drug penetration into the CRNP's inner shells. The surface area reduction to 200 m 2 g −1 for the regenerated CRNPs (Figure 9c) can be attributed to the entrapment of some molecules in the sorbent's pores, which may explain the efficiency decrease inversely with the regeneration round number. = − (13) The spontaneity and exothermic nature can be noticed for TGCN, DXCN, CTCN, and OTCN adsorptions from their negative ΔG° values (Table 2), and their negative ΔH° values supported the exothermic nature claim. Additionally, the average ΔH° values for the different concentrations were below 80.0 kJ mol −1 , indicating that CRNPs removed TGCN, DXCN, CTCN, and OTCN via physisorption [83][84][85][86][87].

Natural Water Treatment and Regeneration Investigations
The CRNPs were tested for removing TGCN, DXCN, CTCN, and OTCN from synthetically polluted GS and SS. The average removal percentage for TGCN, DXCN, CTCN, and OTCN were 98.9%, 94.3%, 98.4%, and 99.1%, respectively, with RSD values of 0.894%, 7.192%, 1.413%, and 0.245%, respectively. Although the fabricated CRNPs were excellent in removing these pollutants from both water types (Figure 9a), it can be seen that the remediation of GS was better than the SS, which indicates that the CRNPs can be employed for the sorption of metal ions. This result can be attributed to salt concentration (mainly saline) in the SS, which may affect the pollutant's migration to the CRNPs surface.
Additionally, the reusability of the CRNPs in removing TGCN, DXCN, CTCN, and OTCN was investigated. According to the pH study, the CRNPs used with DXCN and OTCN were regenerated by sonication for 20 min with 20 mL of 1.0 M sodium hydroxide (NaOH). The sonication was repeated with 10 mL ethanol, then filtered, rinsed with distilled water, and dried at 105 °C for 2.0 h. On the other hand, CRNPs used with TGCN and CTCN were regenerated in the same manner but substituting the 1M NaOH with 1M hydrochloric acid. The performance of virgin CRNPs was considered 100% efficient, and the subsequent cycles were relatively determined (Figure 9b) [51,88]. The regenerated CRNPs showed mean efficiency of 93.40%, 95.31%, 95.49%, and 95.94% for TGCN, DXCN,

Conclusions
CRMs have been derived from TPW via thermal carbonization. A concentrated phosphoric acid was employed for pores opening, and the ball-milling process was utilized to down-size the CRMs and fabricate CRNPs. The batch adsorption experiments were followed to investigate the usability of CRNPs for removing TGCN, DXCN, CTCN, and OTCN from water. The agreement of TGCN, DXCN, CTCN, and OTCN sorption to the PSFO and the LFDM models indicated high adsorption affinity to the CRNPs. Furthermore, the equilibrium results for the four drugs being fitted to the multilayer FIM supported the high-affinity claim. The removal appeared to be exothermic and spontaneous physisorption for the four drugs. Although its efficiency decreased slightly during the consecutive reuses, the CRNPs possessed an excellent remediation efficiency for SS and

Conclusions
CRMs have been derived from TPW via thermal carbonization. A concentrated phosphoric acid was employed for pores opening, and the ball-milling process was utilized to down-size the CRMs and fabricate CRNPs. The batch adsorption experiments were followed to investigate the usability of CRNPs for removing TGCN, DXCN, CTCN, and OTCN from water. The agreement of TGCN, DXCN, CTCN, and OTCN sorption to the PSFO and the LFDM models indicated high adsorption affinity to the CRNPs. Furthermore, the equilibrium results for the four drugs being fitted to the multilayer FIM supported the high-affinity claim. The removal appeared to be exothermic and spontaneous physisorption for the four drugs. Although its efficiency decreased slightly during the consecutive reuses, the CRNPs possessed an excellent remediation efficiency for SS and GS contaminated by TGCN, DXCN, CTCN, and OTCN; Therefore, the prepared CRNPs are recommended for groundwater and seawater remediation as a low-cost sorbent.

Conflicts of Interest:
The authors declare that they have no known competing financial interests or personal relationships that could have influenced the work reported in this paper.