Surface modification of carbon-based adsorbents derived from coal gasification slag for the efficient adsorption of trimethoprim

Figures Abstract Given that pristine water bodies are irreplaceable for maintaining ecosystem homeostasis, the remediation of antibiotic-contaminated aquatic systems has emerged as a pivotal challenge. This research used flotation residual carbon (FRC) obtained from coal gasification coarse slag (CGCS) as a precursor and adopted the H2O2 liquid-phase oxidation method to prepare a low-cost and efficient adsorbent by optimizing process parameters such as modification temperature, time, and oxidant concentration. The H2O2 treated flotation residual carbon was denoted as HFRC. Batch adsorption assays revealed that, under their respective optimal conditions, equilibrium uptakes of Trimethoprim (TMP) by FRC and HFRC reached 54.02 mg/g and 80.01 mg/g, with equilibrium times of 270 min and 360 min. Comprehensive characterization, molecular simulations, and model fitting collectively elucidated the adsorption mechanism and the kinetic intensification of TMP. H2O2 modification markedly increased the abundance of oxygen-containing functionalities. On this basis, the van der Waals and electrostatic interactions between HFRC and TMP were significantly enhanced, accompanied by a pronounced increase in hydrogen-bond density, which collectively accelerated the TMP adsorption kinetics. HFRC also exhibited exceptional dynamic adsorption performance in the fixed-bed column. Overall, HFRC integrates rapid removal performance with resource-cycling benefits and superior environmental adaptability, offering an economically and ecologically synergistic paradigm for the valorization of CGCS and the remediation of antibiotic-impaired waters. Citation: Yang Z, Liu J, Li Q (2026) Surface modification of carbon-based adsorbents derived from coal gasification slag for the efficient adsorption of trimethoprim. PLoS One 21(6): e0351221. https://doi.org/10.1371/journal.pone.0351221 Editor: Mahdi Gharabaghi, University of Tehran, IRAN, ISLAMIC REPUBLIC OF Received: December 5, 2025; Accepted: May 25, 2026; Published: June 10, 2026 Copyright: © 2026 Yang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: All relevant data are within the paper. Funding: The author(s) received no specific funding for this work. Competing interests: The authors have declared that no competing interests exist. 1 Introduction Population growth, increased life expectancy, and improved healthcare have led to an exponential rise in the clinical and agricultural usage of antibiotic emerging contaminants of concern (CECs) such as trimethoprim (TMP) [1]. Following excretion, these compounds are continuously discharged into water bodies via municipal wastewater and agricultural runoff. Owing to their structural stability and high biological activity, they accumulate at the sediment-water interface, with concentrations potentially exceeding the μg/L threshold [2,3]. Trophic magnification disrupts the gut microbiome, facilitates horizontal gene transfer of antibiotic-resistance determinants, and imposes oxidative stress on plant roots, ultimately impairing carbon fixation by primary producers [4,5]. Consequently, proactive “source-to-receptor” control strategies that achieve efficient and sustainable elimination of pharmaceutical residues before they reach aquatic environments are urgently required to safeguard ecosystem health [6]. Among available technologies, adsorption is widely recognized as a core strategy for remediating contaminated waters because of its operational simplicity, low energy demand, and high removal efficiency [7,8]. Consequently, a range of advanced sorbents-activated carbons, magnetic chitosan-Fe3O4 nanocomposites, functionalized silica chelators, graphene oxides, natural porous minerals, and biochar-have been developed. However, it has been observed that each material exhibits intrinsic structure-performance bottlenecks under field conditions [9]. Magnetic chitosan-Fe3O4 is highly pH-sensitive; acidic or alkaline media markedly reduce its capacity, and the chitosan backbone is susceptible to microbial degradation, necessitating costly preservatives for long-term storage [10]. Functionalized silicas offer excellent selectivity, but multi-step syntheses escalate costs and their surface protonation diminishes at pH > 8, demanding tight pH control around 8 to maintain performance [11]. Activated carbons suffer from rate-limiting micropore blockage by natural organic matter; thermal or chemical regeneration is energy-intensive and carbon-depletive, undermining economic viability [12]. Graphene oxides readily aggregate in real waters and require magnetic dopants or polymeric stabilizers; their fabrication involves strong oxidants that pose secondary environmental risks [13]. Natural porous adsorbents are constrained by high acid, energy, and water consumption together with operational safety issues, limiting their scalability [14]. Collectively, current adsorbents fail to reconcile capacity, selectivity, stability, cost-effectiveness, and ecological safety. Biochar is typically produced through pyrolysis, involving thermal treatment at elevated temperatures (commonly ranging from 350 °C to 700 °C) under oxygen-limited or anoxic conditions [15]. There is therefore an urgent need for next-generation materials that efficiently remove TMP while meeting the dual imperatives of environmental protection and human-health security [3,16]. Moreover, current simulations of antibiotic adsorption have predominantly focused on the natural population analysis charges and surface electrostatic potential distributions of adsorbents and antibiotics [17], the total adsorption energy variations associated with different functional groups on the adsorbent [18], the preferential occupation of adsorption sites by antibiotic molecules [19], the adsorption trajectories within idealized pore models of hierarchical porous carbons [20], and the adsorption behaviors on adsorbents with varied interlayer spacings [21]. Notably, dynamic simulations that capture the time-resolved adsorption pathway of antibiotic molecules onto chemically modified adsorbent surfaces remain scarce; consequently, the temporal evolution of adsorption routes, conformational transitions, and intermolecular interactions is poorly understood. This absence of atomistic, time-dependent insight constrains the formulation of realistic theoretical frameworks necessary for elucidating the underlying adsorption mechanisms. The annual production of coal gasification slag (CGS) is substantial, exceeding 50 million tons [22,23]. After classification and separation, the residual carbon (RC) obtained is a by-product characterized by both high carbon content and a well-developed porous structure. This renders it highly advantageous for resource utilization in adsorption and separation applications [24,25]. RC inherits a rich meso- and microporous system formed during the gasification process, which facilitates the diffusion of large organic molecules while ensuring high-capacity adsorption of trace heavy metal ions via micropores [26,27]. The surface of RC contains abundant functional groups such as carboxyl, phenolic hydroxyl, and aromatic rings, enabling stable complexation with organic pollutants through multiple mechanisms including electrostatic attraction, complexation, and hydrogen bonding [28,29]. Due to the partial graphitization that occurs during the high-temperature (>1000 °C) gasification and quenching process, RC exhibits superior thermal shock resistance and compressive strength compared to traditional biomass-derived adsorbents [25,30]. Previous studies have demonstrated that CGS-based adsorbents show promising performance in CO2 capture [31], organic pollutant removal [32], and metal ion recovery [33]. However, systematic investigations into the adsorption behavior of TMP remain scarce. Conventional modification strategies such as acid leaching, calcination, and alkaline treatment are often constrained by harsh reaction conditions, high energy consumption, and severe equipment corrosion, making them incompatible with the principles of green chemistry [29]. Therefore, there is an urgent need to develop efficient, low-energy, and environmentally friendly surface functionalization techniques to enhance the TMP adsorption performance of CGS. H2O2 modification offers a promising alternative, enabling the preservation of adsorbent structural integrity under mild conditions while achieving high-density introduction of surface oxygen-containing functional groups, controllable optimization of pore structures, and precise regulation of surface charge [34,35]. Legocka et al. [34] employed H2O2-oxidized carbon black for the elimination of 2,4-dichlorophenoxyacetic acid and 2-methyl-4-chlorophenoxyacetic acid. Peña et al. [35] utilized H2O2 to tailor the surface chemistry and pore architecture of glycerol-derived porous carbons, thereby enhancing CO2 uptake capacity. Zhang et al. [36] synthesized H2O2-modified peanut shell biochar/sodium alginate microspheres for the high-efficiency sequestration of Cu(II). Thue et al. [37] functionalized natural montmorillonite clay with H2O2 to enrich surface moieties and achieve superior removal of Acid Red 114. Badawy et al. [38] incorporated H2O2 during the fabrication of black clay/rice flour/Fe3O4 (BC/RF/MNPs) adsorbents, resulting in advanced performance in ibuprofen abatement. However, research on the surface modification of RC from CGS using hydrogen peroxide remains limited. Its potential for the removal of antibiotic pollutants such as TMP warrants further investigation, and molecular simulation methods for the adsorption process have been underutilized. CGS can be taxonomically differentiated into two distinct fractions: coal gasification coarse slag (CGCS) and coal gasification fine slag (CGFS) [22]. In this study, flotation-enriched residual carbon (FRC) derived from CGCS was employed as a substrate and subjected to mild oxidative modification using H2O2. The effects of modification temperature, reaction time, and H2O2 concentration on the adsorption performance of FRC were systematically investigated. Comprehensive characterization, molecular simulations, and model fitting were employed to elucidate the differences in TMP adsorption mechanisms before and after modification. This approach enabled the low-cost, low-energy, and clean preparation of an efficient TMP adsorbent, offering dual benefits in environmental protection and public health. 2 Materials and methods 2.1 Materials In this study, the CGCS was sourced from Xinjiang Province, China. The schematic illustration of both the preparation protocol and the experimental adsorption procedure is presented in Fig 1. The fraction below 0.5 mm (ash content 60.12%) was employed as feed for foam flotation recovery of RC (Tables 1 and 2). A 100 g aliquot of the prepared CGCS was slurried with 1 L tap water in a flotation cell; the impeller was rotated at 1800 rpm for 2 min to condition the pulp. Subsequently, 14 kg/t kerosene was introduced as collector and agitated for 2 min, followed by 10 kg kg/t sec-octyl alcohol as frother with an additional 2 min of conditioning. The air valve was then opened to initiate flotation, which was performed twice [39]. During the concentrating stage, the dosages of collector and frother were reduced to 2.8 and 2 kg/t, respectively. The concentrate (i.e., flotation residual carbon, designated as FRC) had an ash content of 11.24% (Table 2) and was subsequently subjected to surface modification using H2O2. TMP (≥98.0%) used to prepare the synthetic pharmaceutical pollutant solution was purchased from Shanghai Aladdin Biochemical Technology Co., Ltd.; H2O2 (analytical grade) was supplied by Sinopharm Chemical Reagent Co., Ltd. The following definitions are provided on an air-dried basis: Mad denotes the moisture content, i.e., the mass fraction of inherent water present in the sample under air-dried conditions; Aad is the ash content, representing the mass fraction of inorganic residue remaining after complete combustion of the sample; Vad refers to the volatile matter, i.e., the mass fraction of gaseous substances released when the sample is heated in an oxygen-free atmosphere; FCad is the fixed carbon, corresponding to the combustible solid carbon in the sample; and Cad, Had, Nad, and Oad represent the mass fractions of carbon, hydrogen, nitrogen, and oxygen elements, respectively. Finally, St,d is the total sulfur on a dry basis (free from moisture), defined as the total mass fraction of all sulfur species (organic sulfur, pyritic sulfur, and sulfate sulfur) in the sample. 2.2 Adsorbent preparation A precise quantity of 10g of FRC was transferred to a 100-mL Erlenmeyer flask. The flask was then treated with 60 mL of H2O2 solution, the mass fraction of which had been previously determined. The contents of the flask were agitated at 150 rpm for a fixed duration at a set temperature. The solid was recovered by vacuum filtration, washed with distilled water until neutrality, dried at 105 °C, and stored in a sealed vial to yield the modified FRC. Triplicate runs were performed; the adsorption capacities toward TMP were averaged. 2.3 Test characterization Phase identification was performed on a SmartLab X-ray diffractometer (XRD) (Rigaku, Japan). Surface chemical bonds and functional groups were analysed with a Nicolet Summit X FTIR spectrometer (Thermo Fisher Scientific, USA). Elemental compositions and surface stoichiometries were determined by X-ray photoelectron spectroscopy (XPS) (ESCALAB 250Xi, Thermo Fisher Scientific) calibrated to the C 1s peak at 284.8 eV. Morphological features were visualised with a Verios XHR scanning electron microscope (SEM) (Thermo Fisher Scientific). Pore-size distributions and specific surface areas were measured by N2 physisorption at −196 °C using an ASAP 2460 automated analyser (Micromeritics, USA); samples were out-gassed at 200 °C and 100 mmHg for 600 min with a heating ramp of 10 °C/min. 2.4 Molecular simulation Quantum chemical calculations were performed to investigate the fundamental properties of TMP, oxygen-containing functional groups, and benzene rings. Molecular models were constructed using the Visualizer module of Materials Studio 6.0, and subsequent calculations were conducted with the Dmol3 module. The computational parameters were as follows [40]: the exchange-correlation energy was described by the generalized gradient approximation (GGA-PW91), and the double numerical basis set including polarization functions (DNP) was employed. The self-consistent field (SCF) convergence criterion was set to 10−5, while the convergence thresholds for energy, maximum force, and maximum displacement were 2 × 10−5 Ha, 4 × 10−3 Ha/Å, and 5 × 10−3 Å, respectively. To further elucidate the adsorption behavior of TMP on various surface functional groups of the adsorbent from a microscopic perspective, molecular simulations were conducted using the Forcite module of Materials Studio 6.0 [41]. Oxygen-containing functional groups (C-O-C, C = O, -COOH, and -OH) were grafted onto the graphite model surface, and the initial configuration of the adsorbent, TMP, and vacuum layer was constructed with dimensions of 50 Å × 50 Å × 87 Å. The COMPASS force field was adopted for the simulations. Structural optimization of the models was performed using the Smart method in the Geometry Optimization module with 20,000 iterations. Molecular dynamics simulations were then carried out under an NVT ensemble at 298.0 K, with temperature control achieved via the Nose thermostat and a time step of 1.0 fs. Electrostatic interactions were calculated using the Ewald method with an accuracy of 0.0001 kcal/mol, while van der Waals interactions were treated using the atom-based method with a cutoff radius of 15.5 Å. All systems reached equilibrium after 500 ps, and the results were subsequently analyzed [42]. 2.5 Static adsorption In the context of the adsorption assays, a quantity of 0.03g of adsorbent was introduced into 50 mL of TMP solution. The pH was adjusted to the desired value with dilute NaOH or HCl. The suspension was transferred to a thermostatic shaker and equilibrated at 25 °C. After predetermined intervals, 10 mL aliquots were withdrawn with a syringe and immediately filtered through a 0.22 μm membrane. The residual TMP concentration was quantified by high-performance liquid chromatography (HPLC, Shimadzu CBM-20A) equipped with a C18 column (150 mm × 2.1 mm, 5 μm). The mobile phase consisted of acetonitrile containing 0.1% formic acid and deionised water (20: 80, V/V) delivered at 1.0 mL/min. All measurements were performed in triplicate and averaged. The equilibrium and kinetic data were interpreted by fitting to the Langmuir and Freundlich models, as well as the pseudo-first-order, Elovich, and intra-particle diffusion kinetic equations [43–45]. (1)(2)In which, qm is the maximum adsorption capacity, mg/g ce is the residual concentration of TMP in the solution at adsorption equilibrium, mg/L. qe is the adsorption capacity of TMP in adsorption equilibrium, mg/g. KL is Langmuir model parameter, L·mg-1. KF (mg·g-1·L1/n·mg-1/n) and n are Freundlich model parameters. (3)(4)(5)In which, qt is the adsorption amount of TMP at t time, mg/g. qe is the adsorption capacity of TMP in adsorption equilibrium, mg/g. k1 is the rate constant of pseudo first-order kinetics model, min-1. α is the initial adsorption rate of Elovich kinetic model, mg·(g·min)-1. β is the desorption constant (or surface-coverage coefficient) of Elovich kinetic model, g/mg. k2 is the rate constant of Intra-particle diffusion model, mg·g-1·min-05. 2.6 Dynamic adsorption Continuous fixed-bed column experiments were conducted to remove TMP from simulated flowing wastewater. A working solution of TMP was prepared at a concentration of 50 mg/L, maintained at 25 °C and pH 7.0. HFRC was packed into the column at different bed heights (0.5, 1.0, and 2.0 cm) and operated under various flow velocities (1, 2, and 5 mL/min) to investigate TMP removal. Breakthrough was defined as the point at which the TMP concentration in the effluent equaled that in the influent, indicating adsorption saturation of the medium. After adsorption was completed, samples were collected from the mixture using a 10 mL syringe, filtered through 0.2 μm membranes, and analyzed in triplicate. The adsorption kinetics were subsequently evaluated using the Thomas and Yoon-Nelson models [46]. The Thomas model incorporates Langmuir-type reversible kinetics, assumes negligible axial dispersion, and employs a second-order rate constant (KTh) to quantify the adsorption-desorption equilibrium. Its analytical solution is capable of describing the entire breakthrough curve and is commonly applied to estimate the maximum adsorption capacity (qe) under column operating conditions. (6)where C0 denotes the influent concentration (mg·L-1), Ct represents the solution concentration at time t (mg·L-1), KTh (mL·mg-1·min-1) denotes the rate constant, qTh (mg·g-1) represents the total adsorption capacity, t (min) is the time, h (cm) is the bed height, T (K) is the absolute temperature, and F (mL·min-1) is the flow velocity. The Yoon-Nelson model simplifies the microscopic adsorption mechanism by using a single rate constant (KYN) to describe the probability of adsorbate capture by the adsorbent. It assumes that the breakthrough time (t) follows a symmetric logistic distribution centered on the 50% breakthrough time (τ), making it particularly suitable for rapid prediction of the median breakthrough point. (7)where t (min) is the duration of the dynamic adsorption process, KYN is the rate constant, and τ (min) is the time required for 50% breakthrough of the bed. 3 Results and discussion 3.1 Preparation of modified adsorbent Fig 2 presents the single-factor experimental results for H2O2-modified FRC under the following conditions: a TMP concentration of 50 mg/L, pH 7, an adsorbent dosage of 0.030 g, and a contact time of 150 min. When the modification time was fixed at 60 min and the H2O2 concentration at mass fraction 10%, the adsorption capacity and removal rate of TMP by the modified FRC first decreased and then increased with rising modification temperature. During the initial oxidation stage, H2O2 preferentially grafts carboxyl, hydroxyl and other acidic oxygen-containing functional groups onto the FRC surface [47], enhancing surface hydrophilicity and facilitating TMP uptake. At 40–60 °C, accelerated H2O2 decomposition attenuates its oxidizing power. Subsequent to an increase in temperature, the hydrothermal treatment strengthens the modification. Although there is a slight contraction of the specific surface area due to collapse of the pore walls or enlargement of the micropores, there is a marked increase in the density of surface oxygen functionalities [48]. When the modification temperature was maintained at 20 °C and the H2O2 concentration at 10%, prolonging the modification time caused the TMP adsorption capacity and removal rate to increase initially and then decline. During the moderate-oxidation phase (< 80 min), the progressive introduction of electron-donating groups (-COOH and -OH) intensifies hydrogen bonding and electrostatic interactions between TMP and the modified FRC. Meanwhile, the specific surface area and micropore volume have not yet deteriorated significantly, leading to enhanced adsorption. However, for treatments exceeding 80 min, excessive oxidation erodes the layered carbon scaffold, collapses micropores and triggers decarboxylation, reducing the number of effective adsorption sites and shifting the pore size distribution toward larger pores [49]. Consequently, 80 min represents the optimal modification duration. Finally, at 20 °C and 80 min, increasing the H2O2 concentration produced a similar rise-and-fall trend in both TMP adsorption capacity and removal efficiency. At low concentrations (< 10%), ·OH radicals generated by H2O2 incorporate carboxyl, phenolic hydroxyl and carbonyl groups into the carbon matrix. These electrophilic oxygen functionalities establish multipoint cooperative interactions, including hydrogen bonding, Lewis acid-base complexation, and π-π electron donor-acceptor interactions, with the pyrimidine nitrogen and amide carbonyl of TMP. Notably, the microporous structure remains essentially intact, thereby elevating adsorption capacity [50,51]. Once the H2O2 concentration exceeds the optimum, excessive etching of the carbon skeleton occurs: micropore walls collapse and both the specific surface area and micropore volume decrease sharply, ultimately causing a pronounced reduction in TMP adsorption capacity and removal efficiency. 3.2 Characteristics of adsorbents The characteristics of the adsorbent are summarized in Figs 3, 4, 5 and Table 3. The data in Fig 3 demonstrate that H2O2 modification triggers a pronounced reconstruction of the chemical architecture of FRC: the global abundance of surface oxygen-bearing functionalities (-COOH, -OH, etc.) rises markedly, whereas hydrophobic moieties (-CH2-) decline commensurately. This evolution originates from the deep oxidation of the carbon scaffold by hydroxyl radicals (·OH) generated during H2O2 decomposition. FTIR spectroscopic evidence (Fig 3a) reveals a synchronous intensification of absorbance bands at 3400−3000 cm-1 and at 1640−1540, 1500 and 1400 cm-1, corroborating a substantial increase in alcoholic hydroxyls, aromatic rings and carboxylic hydroxyls. Conversely, the attenuation of bands at 1725−1700 cm-1 (C = O stretching) and 1025 cm-1 (C-O-C stretching) indicates that a fraction of carbonyl and ether linkages are further oxidised to carboxyl or higher-valence oxygenated species [49]. This oxidative pathway not only elevates surface polarity but also strengthens the affinity of the adsorbent toward the polar molecule TMP. The XPS full spectrum (Fig 3b) indicates a substantial augmentation in the atomic fraction of oxygen on HFRC, thereby validating the experimental findings. The high-resolution C 1s curve-fitting (Fig 3c and 3d) provides a quantitative analysis of the transformation. The relative decrease in the C-C/C-H and O = C- components, as compared to the FRC, is 17.58% and 5.08%, respectively. Conversely, there is an increase in the C-O-C/C-OH and O = C-O components by 19.63% and 5.02%, respectively. The enrichment of oxygenated functionalities, particularly hydroxyl and carboxyl groups, provides a substantial number of polar sites, thereby facilitating efficient TMP uptake. a. FTIR spectrum, b. XPS full spectrum, Fitting curve of C1s peak of FRC (c) and HFRC (d). XRD patterns (Fig 5) exhibit broad “bread-loaf” humps for both FRC and HFRC within 15–30° 2θ, signifying a predominantly amorphous or micro-crystalline carbon matrix [27]. This carbonaceous phase coexists with crystalline quartz and calcite, implying possible physical encapsulation or chemical bonding between carbon and mineral ash. As a potent oxidant, H2O2 induces calcite dissolution and promotes the amorphisation of quartz, thereby profoundly altering the surface morphology of FRC. Following H2O2 treatment, micro-cracks and localized etching appeared on the HFRC surface (Fig 5). To corroborate this observation, BET analysis was conducted (Fig 5 and Table 3). The Isotherm linear absolute plot shifted only marginally, the BET specific surface area rose from 329.3634 m2/g to 332.0318 m2/g, and the average pore diameter decreased from 2.9384 nm to 2.9324 nm, indicating that the micro-porous architecture of the FRC was altered only slightly. Taken together with the FTIR and XPS results, the enhanced adsorbent performance is ascribed to modifications in surface functional groups rather than to pore diameter changes [52,53]. a. Isotherm linear absolute plot, b. Pore size distribution. With the advancement of chemical theory and computational technology, the interactions between TMP and various functional groups of the adsorbent can now be understood at the molecular scale, thereby elucidating how H2O2 induced modifications in the distribution of surface functional groups influence TMP adsorption behavior. The molecular properties of TMP, oxygen-containing functional groups, and benzene rings are illustrated in Fig 6. Charge analysis, performed by calculating and assigning local atomic charges using the Mulliken method, provides critical insights into the electronic structure and physicochemical characteristics of molecules. This approach quantitatively reveals molecular polarity, reactive sites, and the nature and strength of intermolecular interactions, thus bridging the gap between electronic structure and macroscopic properties. The highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) are not merely static energy labels but serve as dynamic indicators of molecular reactivity and selectivity in electronic structure theory [54]. The electron density distribution of the HOMO identifies the regions most prone to electron loss, with its energy level directly correlated to ionization potential and nucleophilic activity. Conversely, the nodal topology and energy position of the LUMO indicate the capacity of the molecule to accept electrons, serving as microscopic criteria for evaluating electron affinity and electrophilic reactivity (with blue regions denoting higher activity). Structurally, TMP consists of a 2,4-diaminopyrimidine ring and a 3,4,5-trimethoxybenzyl moiety connected via a methylene bridge. Its functional groups include aromatic amines (-NH2), methoxy groups (-OCH3), ether linkages (C-O-C), and conjugated π-systems of both the benzene and pyrimidine rings. Negatively charged nitrogen and oxygen atoms constitute the primary contributors to electrostatic interactions and hydrogen bonding between TMP and oxygen-containing functional groups on the adsorbent surface. In terms of van der Waals forces, dispersion interactions dominate, primarily arising from π-π stacking between TMP’s aromatic rings and the graphene planes of activated carbon, with substantial strength attributed to the large contact area. Secondary contributions stem from possible dipole-dipole and induced-dipole interactions between TMP’s polar substituents (methoxy and amino groups) and surface oxygenated groups (e.g., carboxyl and hydroxyl). Moreover, electron sharing between oxygen/nitrogen atoms on the adsorbent surface and TMP molecules has been identified as a key driving mechanism of chemisorption [5,55]. a. Charge, b. LUMO, c. HOMO. (Black, white, red, and blue represent carbon atoms, hydrogen atoms, oxygen atoms, and nitrogen atoms.). Following H2O2 modification, the abundance of oxygen-containing functional groups, specific surface area, and exposure of aromatic structures on HFRC were all significantly enhanced compared with the pristine FRC, thereby accelerating TMP adsorption. The initial and equilibrium configurations derived from molecular simulations are presented in Fig 7a. In this study, the adsorption energies of TMP on carbon-based models enriched with oxygen-containing functional groups were systematically calculated, together with their corresponding energy components [56]. The calculated adsorption energies of TMP on surfaces functionalized with C-O-C, C = O, -COOH, and -OH were −341.192, −418.339, −458.620, and −394.311 kJ/mol, respectively. All values were markedly negative, indicating the thermodynamic spontaneity of TMP adsorption on these surfaces (Table 4). The adsorption strength followed the order: -COOH > C = O > -OH > C-O-C, a trend further corroborated by relative concentration profiles (Fig 7b). Comparative experiments confirmed that H2O2 oxidation introduced concurrent increases in the surface abundance of -COOH, C-O-C, and OH groups on HFRC (Fig 3), thereby resulting in significantly higher adsorption rates than conventional adsorbents [57–60]. a. Initial and equilibrium conformations, b. Relative concentrations, c. Number of hydrogen bonds, d. Mean square displacement (MSD) and diffusion coefficient. Van der Waals and electrostatic interactions dominated the adsorption process. Van der Waals forces comprise dispersion, dipole-dipole, and induction interactions [61,62]. These forces were notably stronger between TMP and surfaces functionalized with C-O-C and C = O groups than with those featuring -COOH or -OH groups. This effect arose from the synergistic interplay between electron cloud polarizability and interfacial hydrogen-bond competition. C-O-C and C = O groups possess extended π-conjugated frameworks and relatively less electronegative peripheral atoms (C, O), which confer higher static polarizabilities than carboxyl (-COOH) and hydroxyl (-OH) groups. This enhanced polarizability facilitated stronger instantaneous dipole–dipole coupling, thereby contributing additional attractive forces. In contrast, electrostatic interactions were most pronounced for -COOH due to its strong polarity, large dipole moment, and bifunctional capability as both proton donor (-OH) and acceptor (C = O). These properties enabled intense ion-dipole interactions and multiple hydrogen bonds with the electron-rich amino and ether oxygen moieties of TMP [63]. The OH group lacked the dual donor-acceptor synergy of -COOH, while C = O functioned solely as a hydrogen-bond acceptor with weaker interactions. The ether oxygen in C-O-C was delocalized within a conjugated system, exhibiting the lowest polarity and, consequently, the weakest electrostatic interactions [64]. Thus, the order of electrostatic attraction strengths was -COOH > -OH > C = O > C-O-C, with -COOH representing the key group that reinforces interfacial interactions due to its unique acid–base properties and integrated hydrogen-bonding capabilities. Hydrogen bonding provided an additional contribution, with TMP forming diverse configurations with different functional groups [65,66]. -COOH could act as both a donor (O-H···N/O) and an acceptor (C = O···H-N), forming strong bifunctional hydrogen bonds with TMP’s amino and ether oxygen groups. OH could serve as a donor (O-H···N/O) or a weak acceptor (O···H-N), resulting in moderate hydrogen-bonding interactions. CO functioned exclusively as an acceptor (C = O···H-N), engaging in single-point interactions with TMP’s amino group. By contrast, C-O-C, lacking active protons and with limited electron lone-pair exposure, acted only as a weak acceptor (C-O-C···H-N), forming relatively unstable hydrogen bonds (Fig 8). Consequently, the order of hydrogen bond numbers was primarily determined by the donor and acceptor capacities of each group: -COOH provided both strong donor (acidic H) and strong acceptor (carbonyl O) sites, -OH offered donor capacity but weaker acceptor ability, C = O acted solely as an acceptor, while C-O-C, constrained by electron delocalization and steric hindrance, exhibited the weakest acceptor ability and lacked any donor function. The resulting hydrogen-bonding sequence was therefore -COOH > -OH > C = O > C-O-C (Fig 7c). At the microscopic level, the low diffusion coefficients and strong adsorption of TMP on -COOH, -OH, and C = O functionalized surfaces were attributable to the cooperative restrictions imposed by these interactions (Fig 7d). Following H2O2 oxidation, the increased abundance of C-O-C groups enhanced van der Waals forces, while the concurrent enrichment of -COOH and -OH groups significantly strengthened electrostatic interactions and hydrogen bonding. Collectively, these effects accelerated the adsorption kinetics of TMP through a synergistic mechanism. This structure-property coupling endowed the modified HFRC with superior capacity and faster TMP sequestration compared with unmodified FRC, providing both theoretical insights and practical advantages for efficient environmental remediation. Fig 9a illustrates the variation in the adsorption performance of FRC and HFRC toward TMP under different dosages. With increasing adsorbent dosage, the adsorption capacity exhibits a gradual decline, whereas the overall removal efficiency increases markedly. This phenomenon can be ascribed to the substantial enlargement of the total specific surface area at higher dosages, which provides a greater number of adsorption sites and thereby enhances the bulk removal of TMP from solution. A further comparison reveals that HFRC possesses a stronger affinity and selectivity for TMP than FRC, indicating that the H2O2 modification increases the number of active adsorption sites. The dosage corresponding to the intersection of the unit adsorption capacity and removal efficiency curves is commonly regarded as the optimal dosage for the adsorption system [67]. Accordingly, 0.6 g/L was selected for subsequent adsorption experiments. The effect of the initial TMP concentration on adsorption performance is illustrated in Fig 10a. A progressive enhancement in the uptake of both pollutants is observed as the initial TMP concentration rises, because this concentration constitutes the principal driving force that overcomes the solid-liquid mass-transfer resistance [6]. The adsorption capacity of HFRC exceeds that of the unmodified counterpart, demonstrating that HFRC can rapidly and efficiently eliminate TMP in domains where the local concentration is elevated, and underscoring the superiority of H2O2 modification. Adsorption isotherms were fitted with established models, and the results are compiled in Table 5. The adsorption isotherm of TMP onto HFRC is better described by the Freundlich model, indicating that the adsorption process involves multilayer or pore-filling behavior rather than ideal monolayer adsorption. Moreover, its relatively high adsorption rate and capacity are attributed to the modifications in surface functional groups induced by the treatment, as well as its intrinsic pore structure. a. Adsorption models, b. Kinetic models, c. Intra-particle diffusion. Adsorption profiles of TMP uptake by FRC and HFRC, together with the corresponding kinetic fits, are presented in Fig 10b. Equilibrium is attained within 270 min for FRC and 360 min for HFRC, with respective capacities of 54.02 and 80.01 mg/g. The H2O2-treated FRC evidently accelerates and intensifies TMP sequestration from the aqueous phase. Kinetic modelling (Table 5) reveals that the pseudo-first-order equation best fits the experimental data, indicating that the overall process is governed by physical diffusion and that the rate is proportional to the number of unoccupied surface sites [43]. This outcome corroborates the hypothesis that van der Waals forces, hydrophobic interactions, and π-π stacking dominate TMP adsorption, and implies that the ultimate capacity is constrained by pore architecture and specific surface area. The higher k₁ value obtained for HFRC signifies faster migration of TMP molecules from the bulk liquid to the active surface sites, enabling the system to approach equilibrium more rapidly. Furthermore, the non-equilibrium stages of adsorption on both FRC and HFRC conform to the intra-particle diffusion model: the initial step reflects film diffusion, whereas the second step represents intraparticle transport. The non-zero intercept of the fitted lines evidences the simultaneous operation of film and intra-particle diffusion mechanisms [68,69]. The H2O2-modified HFRC exhibits superior adsorption performance relative to the pristine FRC. Consequently, the adaptive characteristics of HFRC were systematically evaluated, and the corresponding results are presented in Fig 11. Investigating sorbent performance across a gradient of ionic strengths furnishes indispensable thermodynamic evidence for full-scale deployment. In typical effluents, NO3-, SO42-, Cl- and HCO3- coexist as background electrolytes, among which NO3- exerts the most pronounced competitive pressure. HFRC buffers the interference of co-existing anions under elevated ionic strength and retains a high affinity for the target contaminant (Fig 11a). Solution pH, the master variable of aquatic chemistry, governs not only the speciation of the contaminant but also the surface-charge density of the adsorbent via protonation/de-protonation reactions [16]. As shown in Fig 11b, the adsorption capacity of HFRC toward TMP remains virtually constant within the pH window 5–10, evidencing its robustness against wide pH fluctuations encountered in complex water matrices. Reusability is a pivotal criterion for assessing the life-cycle economics of an adsorbent. Five consecutive adsorption-desorption cycles were therefore conducted under static conditions (Fig 11c). Although the equilibrium capacity exhibits an exponential decline with cycle number, the loss after the fifth cycle is only 7.19%, corroborating the synergistic reinforcement of structural integrity and the regenerability of surface active sites in HFRC. Notably, HFRC reaches and 80.01–165.03 mg/g within 270 min, substantially outperforming previously reported sorbents: Microplastics (1.11–1.31 mg/g, 60 h) [70], peanut husk modified with potassium permanganate (25 mg/g, 360 min) [71], modified peanut husk (16–22 mg/g, 180 min) [72], mesoporous silica/reduced graphene oxide composite (17.60 mg/g, 30 min) [73], hydrogel-TiO2 (0.10–2.00 mg/g, 18 h) [59], and commercial activated carbon F400 (18.40 mg/g, 10 d) [12]. This comparison underscores the superior sorption capacity and kinetic efficiency of HFRC, attesting to its promise as a next-generation adsorbent. The production cost of ton-scale HFRC fabricated from CGCS via a flotation-H2O2 oxidation coupling route, expressed in Ren Min Bi (CNY), is itemized below. In the flotation stage, operated with 16.8 kg/t kerosene and 12 kg/t sec-octyl alcohol at a concentrate yield of 19.13%, 5.23 t of CGCS can be processed to obtain 1 t of FRC. Based on current industrial procurement prices (kerosene 6.3 CNY/kg, sec-octanol 10.0 CNY/kg), the reagent cost amounts to 1,180 CNY. During the oxidation stage, 1 t of FRC consumes approximately 3,000 kg of 27.5% Industrial-grade H2O2, incurring an expenditure of 1,900 CNY. After adding 10% overhead and profit, the ex-works cost of HFRC is approximately 3,388 CNY/t, which is markedly lower than the prevailing market price of conventional activated carbons (4,000–20,000 CNY/t), demonstrating the pronounced economic advantage of converting CGCS into HFRC. 3.5 Dynamic adsorption and model fitting The experimental data pertaining to the dynamic adsorption performance of HFRC, along with the corresponding model-fitting outcomes, are systematically summarized in in Fig 12 and Table 6. As shown in Fig 12a, increasing the bed height provides a greater number of accessible adsorption sites. Theoretically, extending the bed height can prolong the operational lifespan of the column, thereby reducing the frequency of regeneration and replacement and consequently lowering labor costs. Breakthrough curves obtained at different bed heights were fitted using the Thomas model, with the fitting parameters summarized in the table. The coefficients of determination (R2) for the three experimental sets were 0.9948, 0.9740, and 0.9780, indicating excellent agreement between the model and experimental data. The calculated equilibrium adsorption capacity (q0) increased with bed height, reaching 47.60, 59.91, and 65.18 mg/g, respectively. As bed height increased, the adsorption rate constant (KTh) decreased monotonically, attributed to the extended solid-liquid contact time and the consequent reduction in mass transfer driving force, whereas the equilibrium adsorption capacity (qTh) increased due to longer residence time. Similarly, the Yoon-Nelson model also demonstrated excellent fitting performance (R2 > 0.9700). The model parameter KYN decreased with increasing bed height, while the 50% breakthrough time (τ) was correspondingly prolonged, further confirming the effect of contact time. Fig 12b illustrates that increasing the flow velocity significantly shortens the time required for TMP to reach adsorption equilibrium, which is attributed to limited solid-liquid contact under high flow conditions. Fitting with the Thomas model indicated that KTh increased linearly with flow velocity, consistent with the experimentally observed earlier breakthrough. The corresponding equilibrium adsorption capacities (q0) were 67.69, 59.91, and 52.46 mg/g, showing a decreasing trend. Although higher flow rates enhance the mass transfer driving force, they simultaneously reduce the residence time of solutes within the column, leading to decreased utilization of adsorption sites and, consequently, a lower qTh. Linear fitting using the Yoon-Nelson model yielded correlation coefficients exceeding 0.9700, confirming good agreement. The adsorption rate constant KYN increased gradually from 0.02601 to 0.03792, reflecting a shorter time required to reach adsorption saturation. With increasing influent flow rate, the residence time of the liquid phase in the fixed bed decreased, resulting in a reduction in the 50% breakthrough time (τ). 4 Conclusion This study employed the green oxidant H2O2 to modify the surface of CGCS flotation residual carbon and systematically evaluated the adsorption performance of the resulting adsorbent toward a typical antibiotic pollutant TMP. The underlying enhancement mechnism was also thoroughly elucidated. The main conclusions are as follows: - (1) Experimental results indicated that the optimal modification conditions were 20 ℃, 15% H2O2 concentration, and 80 min treatment time. In these conditions, equilibrium uptakes of TMP by FRC and HFRC reached 54.02 mg/g and 80.01 mg/g, with equilibrium times of 270 min and 360 min, respectively. These results confirm the potential for efficient adsorption. - (2) The adsorption behavior conformed to the Freundlich, pseudo-first-order kinetic, and Intra-particle diffusion models, demonstrating that the adsorption of TMP by HFRC is a rapid process dominated by physical diffusion and jointly governed by film and intra-particle diffusion. H2O2 modification markedly enlarged both the aromatic domains and the number of reactive oxygen sites, thereby strengthening van der Waals, electrostatic and hydrogen-bonding interactions and ultimately enabling efficient TMP adsorption. - (3) HFRC maintained a pronounced affinity for TMP despite the competitive presence of co-existing anions, and accommodated wide pH fluctuations typical of complex aquatic matrices. After multiple operational cycles, the adsorbent retained >90% of its original adsorption capacity, corroborating its environmental robustness and engineering sustainability. - (4) With increasing bed height, the residence time of TMP molecules on the HFRC surface within the fixed-bed column is substantially prolonged, thereby facilitating the progressive establishment of adsorption equilibrium from a kinetic perspective. Concurrently, the higher bed height offers a greater number of accessible adsorption sites, effectively delaying the onset of the breakthrough curve. In contrast, an elevated flow velocity reduces the contact time between solute and adsorbent, constrains the completeness of the mass transfer-diffusion process, and consequently results in a shorter breakthrough time, decreased treated water volume, and ultimately a lower overall adsorption capacity per unit bed. These findings elucidate the coupled mechanisms between mass transfer and adsorption kinetics in fixed-bed systems, while also providing critical insights for the optimization of bed configuration and operational parameters in engineering practice, thereby offering valuable guidance for the large-scale application of HFRC. References - 1. Aryee AA, Mpatani FM, Dovi E, Li Q, Wang J, Han R, et al. A novel antibacterial biocomposite based on magnetic peanut husk for the removal of trimethoprim in solution: adsorption and mechanism study. Journal of Cleaner Production. 2021;329:129722. - 2. Berges J, Moles S, Ormad MP, Mosteo R, Gómez J. Antibiotics removal from aquatic environments: adsorption of enrofloxacin, trimethoprim, sulfadiazine, and amoxicillin on vegetal powdered activated carbon. Environ Sci Pollut Res Int. 2021;28(7):8442–52. pmid:33063209 - 3. Ma Y, Yuan P-K, Wu Y, Meng H, Wang G-X, Xie W-M, et al. Insight into the role of different extracellular polymeric substances components on trimethoprim adsorption by activated sludge. J Environ Manage. 2022;306:114502. pmid:35033891 - 4. Lavecchia R, Zuorro A, Baaloudj O, Brienza M. Trimethoprim removal from aqueous solutions via volcanic ash-soil adsorption: process modeling and optimization. Water. 2024;16(15):2209. - 5. Xue Z, Feng Y, Li H, Xu C, Ju J, Dong L, et al. Adsorption of trimethoprim and sulfamethoxazole using residual carbon from coal gasification slag: behavior, mechanism and cost-benefit analysis. Fuel. 2023;348:128508. - 6. Mpatani FM, Aryee AA, Kani AN, Han R, Li Z, Dovi E, et al. A review of treatment techniques applied for selective removal of emerging pollutant-trimethoprim from aqueous systems. Journal of Cleaner Production. 2021;308:127359. - 7. Plazinski W, Rudzinski W, Plazinska A. Theoretical models of sorption kinetics including a surface reaction mechanism: a review. Adv Colloid Interface Sci. 2009;152(1–2):2–13. pmid:19735907 - 8. Ortiz-Ramos U, Leyva-Ramos R, Mendoza-Mendoza E. Modeling adsorption rate of trimethoprim, tetracycline and chlorphenamine from aqueous solutions onto natural bentonite clay. Elucidating mass transfer mechanisms. Chemical Engineering Journal. 2024;493:152666. - 9. Zamiri MA, Wang J, Niu CH. Competitive adsorption and molecular interaction study of sulfamethoxazole and trimethoprim on biomass-based activated carbon. Chemosphere. 2025;385:144591. pmid:40714540 - 10. Alishiri M, Gonbadi M, Narimani M, Abdollahi SA, Shahsavaripour N. Optimization of process parameters for trimethoprim and sulfamethoxazole removal by magnetite-chitosan nanoparticles using Box-Behnken design. Sci Rep. 2023;13(1):14489. pmid:37660165 - 11. Jodeh S, Jaber A, Hanbali G, Massad Y, Safi ZS, Radi S, et al. Experimental and theoretical study for removal of trimethoprim from wastewater using organically modified silica with pyrazole-3-carbaldehyde bridged to copper ions. BMC Chem. 2022;16(1):17. pmid:35313931 - 12. Chianese S, Fenti A, Blotevogel J, Musmarra D, Iovino P. Trimethoprim removal from wastewater: adsorption and electro-oxidation comparative case study. Case Studies in Chemical and Environmental Engineering. 2023;8:100433. - 13. Carrales-Alvarado DH, Rodríguez-Ramos I, Leyva-Ramos R, Mendoza-Mendoza E, Villela-Martínez DE. Effect of surface area and physical–chemical properties of graphite and graphene-based materials on their adsorption capacity towards metronidazole and trimethoprim antibiotics in aqueous solution. Chemical Engineering Journal. 2020;402:126155. - 14. Coutinho R, Hoshima HY, Vianna MTG, Marques M. Sustainable application of modified Luffa cylindrica biomass for removal of trimethoprim in water by adsorption with process optimization. Environ Sci Pollut Res Int. 2024;31(43):55280–300. pmid:39227535 - 15. Zoroufchi Benis K, Minaei S, Soltan J, McPhedran KN. Adsorption of lincomycin on microwave activated biochar: batch and dynamic adsorption. Chemical Engineering Research and Design. 2022;187:140–50. - 16. Xu S, Bao Y, et al. Isotherm, kinetics, and adsorption mechanism studies of coal gasification coarse slag as highly efficient phosphate adsorbents. Separations. 2024;11(6):182. - 17. Cheng Y, Yang J, Shen J, Yan P, Liu S, Kang J, et al. Preparation of P-doped biochar and its high-efficient removal of sulfamethoxazole from water: adsorption mechanism, fixed-bed column and DFT study. Chemical Engineering Journal. 2023;468:143748. - 18. Xu X, Dong H, Liu J, Zhuang J, Huang T, Gan Y, et al. Phosphate-ligand engineered biochar with electron-donor and nanoconfinement effects for selective antibiotic adsorption. Chemical Engineering Journal. 2025;522:167713. - 19. Zhao H, Wang Z, Liang Y, Wu T, Chen Y, Yan J, et al. Adsorptive decontamination of antibiotics from livestock wastewater by using alkaline-modified biochar. Environ Res. 2023;226:115676. pmid:36907344 - 20. Xu H, Xu L, Wang Z, Wei F, Zhang M. Diffusion-adsorption synergistic mechanism in the adsorption of hierarchical porous carbon for antibiotics removal from water. Colloids and Surfaces A: Physicochemical and Engineering Aspects. 2025;711:136327. - 21. Wang H, Yang Q, Li Z, Xie Y, Kong F, Liang D, et al. Adsorption of antibiotics on montmorillonite and site energy distribution analysis. Journal of Molecular Liquids. 2023;383:122078. - 22. Liu X, Jin Z, Jing Y, Fan P, Qi Z, Bao W, et al. Review of the characteristics and graded utilisation of coal gasification slag. Chinese Journal of Chemical Engineering. 2021;35:92–106. - 23. Xue Z, Yang C, Dong L, Bao W, Wang J, Fan P. Recent advances and conceptualizations in process intensification of coal gasification fine slag flotation. Separation and Purification Technology. 2023;304:122394. - 24. Dong S, Qiao J, Kang C, Sun M, Yang S, Zhao Y, et al. A review of coal solid waste-based catalytic materials in wastewater treatment: preparation and application. Separation and Purification Technology. 2025;367:132800. - 25. Yan S, Xuan W, Cao C, Zhang J. A review of sustainable utilization and prospect of coal gasification slag. Environ Res. 2023;238(Pt 2):117186. pmid:37741569 - 26. Guo F, Guo Y, Chen L, Jia W, Zhu Y, Li Y, et al. Multitudinous components recovery, heavy metals evolution and environmental impact of coal gasification slag: a review. Chemosphere. 2023;338:139473. pmid:37451637 - 27. Ren L, Ding L, Guo Q, Gong Y, Yu G, Wang F. Characterization, carbon-ash separation and resource utilization of coal gasification fine slag: a comprehensive review. Journal of Cleaner Production. 2023;398:136554. - 28. Lv B, Deng X, Jiao F, Dong B, Fang C, Xing B. Enrichment and utilization of residual carbon from coal gasification slag: a review. Process Safety and Environmental Protection. 2023;171:859–73. - 29. Yuan N, Zhao A, Hu Z, Tan K, Zhang J. Preparation and application of porous materials from coal gasification slag for wastewater treatment: a review. Chemosphere. 2022;287(Pt 2):132227. pmid:34826920 - 30. Su S, Tahir MH, Cheng X, Zhang J. Modification and resource utilization of coal gasification slag-based material: a review. Journal of Environmental Chemical Engineering. 2024;12(2):112112. - 31. Wang Y, Fu N, Hu X, Wen X, Yu Q, Guo Q, et al. A hierarchical porous carbon/silicon composites from coal gasification fine slag for CO2 capture and CO2/H2 catalysis. Carbon Lett. 2025;35(5):2093–109. - 32. Wang Y, Liu R, Wang J, Hao F. Optimization and application of acid-modified coal gasification fine slag for effective removal of chemical oxygen demand and volatile phenols in semi-coking wastewater. Results in Engineering. 2025;26:104697. - 33. Peng D, Zhang S, Wang K, Dong T, Zhang M, Dong G. Preparation of a chitosan/coal gasification slag composite membrane and its adsorption and removal of Cr (VI) and RhB in water. Molecules. 2022;27(21):7173. pmid:36363999 - 34. Legocka I, Kuśmierek K, Świątkowski A, Wierzbicka E. Adsorption of 2,4-D and MCPA herbicides on carbon black modified with hydrogen peroxide and aminopropyltriethoxysilane. Materials (Basel). 2022;15(23):8433. pmid:36499931 - 35. Guerrero Peña GDJ, Varghese AM, Kuppireddy S, Hart P, Zakari RSB, Alamoodi N, et al. Hydrogen peroxide-treated glycerol sourced porous carbon with elemental sulfur-based sulfur-phosphorus co-doping for CO2 capture. Sci Total Environ. 2025;969:178967. pmid:40020584 - 36. Zhang L, Li Q, Zhu J, Liu H, Liu X, Wang Y, et al. H2O2 modified peanut shell-derived biochar/alginate composite beads as a green adsorbent for removal of Cu(II) from aqueous solution. Int J Biol Macromol. 2023;240:124466. pmid:37062377 - 37. Thue PS, Teixeira RA, Lima EC, Mello BL, dos Reis GS, Machado FM, et al. Surface modification of natural clay with H2O2 for high adsorption of acid Red 114: experimental and modelling studies. Journal of Molecular Liquids. 2023;388:122740. - 38. Badawy AM, Farghali AA, Bonilla-Petriciolet A, Seliem MK, Selim AQ, Ali MA, et al. Facile synthesis of a recyclable multifunctional magnetic adsorbent prepared from H2O2-modified carbon clay/rice flour polymer/Fe3O4 nanoparticles interface for effective removal of ibuprofen. Journal of the Taiwan Institute of Chemical Engineers. 2023;152:105177. - 39. Guo F, Miao Z, Guo Z, Li J, Zhang Y, Wu J. Properties of flotation residual carbon from gasification fine slag. Fuel. 2020;267:117043. - 40. Fuerstenau DW, Harris GH, Jia R. Nonionic Surfactants as Collectors for the Flotation of Oxidized and/or Low-Rank Coal. ACS Symposium Series. American Chemical Society. 1999. p. 230–47. - 41. Wang B, Tian J, Fan B, Wang X, Li E. Enhancement on KCl flotation at low temperature by a novel amine-alcohol compound collector: experiment and molecular dynamic simulation. Minerals. 2025;15(8):862. - 42. Xiao M, Xu S, Huang H, Ren S. Release behavior of Pb(II) Ions on the galena surface: dissolution experiment, DFT calculation, and md simulation. Minerals. 2024;14(11):1075. - 43. Revellame ED, Fortela DL, Sharp W, Hernandez R, Zappi ME. Adsorption kinetic modeling using pseudo-first order and pseudo-second order rate laws: a review. Cleaner Engineering and Technology. 2020;1:100032. - 44. Mabuza M, Premlall K, Daramola MO. Modelling and thermodynamic properties of pure CO2 and flue gas sorption data on South African coals using Langmuir, Freundlich, Temkin, and extended Langmuir isotherm models. Int J Coal Sci Technol. 2022;9(1). - 45. Wang J, Guo X. Rethinking of the intraparticle diffusion adsorption kinetics model: interpretation, solving methods and applications. Chemosphere. 2022;309(Pt 2):136732. pmid:36223824 - 46. Meina L, Qiao M, Zhang Q, Xu S, Wang D. Study on the dynamic adsorption and recycling of phosphorus by Fe-Mn oxide/mulberry branch biochar composite adsorbent. Sci Rep. 2024;14(1):1235. pmid:38216644 - 47. Zhang Y, Qu Y, Wang J, Wang D. Adsorption of bromate in water using hydrogen peroxide-modified activated carbon. CyTA - Journal of Food. 2025;23(1). - 48. Jain A, Balasubramanian R, Srinivasan MP. Production of high surface area mesoporous activated carbons from waste biomass using hydrogen peroxide-mediated hydrothermal treatment for adsorption applications. Chemical Engineering Journal. 2015;273:622–9. - 49. Li Y, Liu Y, Yang W, Liu L, Pan J. Adsorption of elemental mercury in flue gas using biomass porous carbons modified by microwave/hydrogen peroxide. Fuel. 2021;291:120152. - 50. Jin X, Che R, Yang J, Liu Y, Chen X, Jiang Y, et al. Activated carbon and carbon quantum dots/titanium dioxide composite based on waste rice noodles: simultaneous synthesis and application in water pollution control. Nanomaterials (Basel). 2022;12(3):472. pmid:35159817 - 51. Danish M, Ansari KB, Danish M, Khatoon A, Ali Khan Rao R, Zaidi S, et al. A comprehensive investigation of external mass transfer and intraparticle diffusion for batch and continuous adsorption of heavy metals using pore volume and surface diffusion model. Separation and Purification Technology. 2022;292:120996. - 52. Rijfkogel LS, Ghanbarian B, Hu Q, Liu H-H. Clarifying pore diameter, pore width, and their relationship through pressure measurements: a critical study. Marine and Petroleum Geology. 2019;107:142–8. - 53. Kamseu E, NGouloure ZNM, Ali BN, Zekeng S, Melo UC, Rossignol S, et al. Cumulative pore volume, pore size distribution and phases percolation in porous inorganic polymer composites: relation microstructure and effective thermal conductivity. Energy and Buildings. 2015;88:45–56. - 54. Zhu Q, Li Z. Hydrogel-supported nanosized hydrous manganese dioxide: synthesis, characterization, and adsorption behavior study for Pb2+, Cu2+, Cd2+ and Ni2+ removal from water. Chemical Engineering Journal. 2015;281:69–80. - 55. Williams NE, Aydinlik NP. KOH ratio effect, characterization, and kinetic modeling of methylene blue from aqueous medium using activated carbon fromThevetia peruvianashell. Chemical Engineering Communications. 2020;208(8):1189–208. - 56. Xia Y, Zhang R, Cao Y, Xing Y, Gui X. Role of molecular simulation in understanding the mechanism of low-rank coal flotation: a review. Fuel. 2020;262:116535. - 57. Martínez-Costa JI, Leyva-Ramos R, Padilla-Ortega E, Aragón-Piña A, Carrales-Alvarado DH. Antagonistic, synergistic and non-interactive competitive sorption of sulfamethoxazole-trimethoprim and sulfamethoxazole‑cadmium (ii) on a hybrid clay nanosorbent. Sci Total Environ. 2018;640–641:1241–50. pmid:30021289 - 58. Nielsen L, Bandosz TJ. Analysis of the competitive adsorption of pharmaceuticals on waste derived materials. Chemical Engineering Journal. 2016;287:139–47. - 59. Villanueva ME, Puca M, Pérez Bravo J, Bafico J, Campo Dall Orto V, Copello GJ. Dual adsorbent-photocatalytic keratin–TiO2 nanocomposite for trimethoprim removal from wastewater. New J Chem. 2020;44(26):10964–72. - 60. Li J, Zhang H. Factors influencing adsorption and desorption of trimethoprim on marine sediments: mechanisms and kinetics. Environ Sci Pollut Res Int. 2017;24(27):21929–37. pmid:28780692 - 61. Israelachvili JN, Tabor D. Van der Waals forces: theory and experiment. Progress in surface and membrane science. Elsevier. 1973. p. 1–55. - 62. Hermann J, DiStasio Jr RA, Tkatchenko A. First-principles models for van der Waals interactions in molecules and materials: concepts, theory, and applications. Chem Rev. 2017;117(6):4714–58. pmid:28272886 - 63. Nakatsuji H. Electrostatic force theory for a molecule and interacting molecules. I. Concept and illustrative applications. J Am Chem Soc. 1973;95(2):345–54. - 64. Besley E. Recent developments in the methods and applications of electrostatic theory. Acc Chem Res. 2023;56(17):2267–77. pmid:37585560 - 65. Dannenberg JJ. The nature of the hydrogen bond: outline of a comprehensive hydrogen bond theory. J Am Chem Soc. 2010;132(9):3229–3229. - 66. Buckingham AD, Del Bene JE, McDowell SAC. The hydrogen bond. Chemical Physics Letters. 2008;463(1–3):1–10. - 67. Yao Y, Feng Y, Li H, Liu M, Cui Y, Xu C, et al. Investigation of the adsorption performance and mechanism of multi-source mineral composite calcination materials on heavy metal ions. Desalination. 2024;586:117847. - 68. Dharmarathna SP, Priyantha N. Investigation of boundary layer effect of intra-particle diffusion on methylene blue adsorption on activated carbon. Energy Nexus. 2024;14:100294. - 69. Auton LC, Aguareles M, Valverde A, Myers TG, Calvo-Schwarzwalder M. An analytical investigation into solute transport and sorption via intra-particle diffusion in the dual-porosity limit. Applied Mathematical Modelling. 2024;130:827–51. - 70. Li Z, Meng X, Shi X, Li C, Zhang C. Adsorption–desorption behaviors of enrofloxacin and trimethoprim and their interactions with typical microplastics in aqueous systems. Sustainability. 2025;17(2):516. - 71. Aryee AA, Han R, Qu L. Investigation into the adsorption of methylene blue and trimethoprim onto modified peanut husk in single and binary systems. Desalination and Water Treatment. 2024;317:100248. - 72. Aryee AA, Han R. A novel biocomposite based on peanut husk with antibacterial properties for the efficient sequestration of trimethoprim in solution: batch and column adsorption studies. Colloids and Surfaces A: Physicochemical and Engineering Aspects. 2022;635:128051. - 73. Dogan AD, Sari Yilmaz M, Sari Erkan H. Mesoporous silica/reduced graphene oxide composite for fluoxetine and trimethoprim adsorption. Microporous and Mesoporous Materials. 2025;396:113713.