Adsorption of heavy metal ions and organic pollutants from refining wastewater by magnetically synthesized silver nanoparticles coated with graphene oxide

Abstract

The release of heavy metals and organic pollutants from industrial, agricultural, and domestic sources presents a serious threat to water bodies and ecosystems. Removing contaminants from refinery wastewater is essential for protecting water quality and the environment. This study developed silver nanoparticles using Saccharomyces cerevisiae in a magnetic field, which were subsequently coated with graphene oxide (GO) to create an effective adsorbent. The beads have been tested for their effectiveness in removing heavy metals such as lead, mercury, and cadmium, as well as organic contaminants like naphthalene, phenol, and fluorene from wastewater. At an optimal pH of 7, a contact time of 6 h, and a temperature of 298 K, the as-developed Ag-GONA beads exhibited maximum adsorption capacities of 326.77 mg/g for lead Pb, 300.37 mg/g for mercury Hg, 219.13 mg/g for cadmium Cd, 71.93 mg/g for naphthalene, 67.77 mg/g for fluorene, and 58.11 mg/g for phenol. Kinetic analysis suggested that the process involved chemisorption following a pseudo-second order (PSO) mechanism, while equilibrium data were well described by the Langmuir isotherm. Adsorption took place via hydrogen bonding, pore filling, and electrostatic attraction. The Ag-GONA beads showed remarkable reusability, with the removal rate of Pb(II) decreasing from 99.9% to 78% after six cycles. In contrast, the removal rates for Hg(II) and Cd(II) remained consistent, staying between 75% and 80%. The removal of naphthalene, phenol, and fluorene demonstrated a reduction of about 20% over six consecutive adsorption and desorption cycles. This novel adsorbent presents a promising, sustainable approach for refinery wastewater treatment, advancing environmental conservation strategies.

Introduction

Water is essential for human existence, sustaining essential processes and daily activities¹. However, the surge in worldwide water consumption, driven by factors such as rising population, urbanization, and industrialization, has significantly increased wastewater generation². This expansion has substantially contributed to the environment, leading to the presence of detrimental compounds such as chemical pesticides, dyes, phenolic compounds, and poisonous heavy metals in water bodies such as lakes and rivers³. Petroleum refinery wastewater (PRW) is produced at a rate of approximately 30–50 cubic meters per metric ton of refined crude oil⁴. The PRW comprises harmful pollutants, including chemical oxygen demand (COD) ranging from 310 to 600 mg/L, grease from 100 to 400 mg/L, phenols from 200 to 1600 mg/L, and a large number of organic contaminants and heavy metals such as Pb and Cr in between 10 and 150 mg/L^4–6. The serious consequences of these contaminants underline the crucial need for appropriate refinery effluent management and treatment strategies to protect human health and the environment.

Recently, numerous technologies have been used to treat petroleum wastewater, including biodegradation⁷, membrane ultrafiltration⁵, ion exchange⁸, photocatalysis⁹, electrocoagulation¹⁰, and adsorption^11,12. Adsorption is a common wastewater treatment method because it is simple, cost-effective, and efficient. However, the properties of the adsorbent and its regeneration capacity are directly related to adsorption efficiency. Ongoing research focuses on developing efficient and sustainable adsorbents with enhanced regeneration and reusability^13,14. Due to their small size and large surface area-to-volume ratio, microorganisms have a high interaction with water impurities. Waste microbial biomass is a promising biotechnological approach for addressing water pollution due to its high contaminant removal efficiency, environmental adaptability, and compatibility with established ion exchange technology¹⁵. Saccharomyces cerevisiae, a major byproduct of beer wineries, stands out for its applicability in biosorption because of its low production cost, high adsorption capability, safety, and large-scale growth¹⁶. Saccharomyces cerevisiae biomass is widely available and frequently utilized, making it a promising solution for environmental engineering, especially in wastewater treatment¹⁷.

GO finds extensive application in various fields due to its remarkable physicochemical properties, including its high specific surface area, electrical and thermal conductivity, and exceptional adsorption capacity^18,19. Its sustained hydrophilicity enhances GO’s effectiveness in water purification applications^20,21. Due to its large surface area, GO functions as a matrix for developing diverse nanocomposites for various applications²². However, even after the contaminants have been adsorbed, the high dispersibility of GO makes it difficult to separate them from aqueous solutions. Magnetizing GO has proven to be a valuable solution for addressing this issue. Magnetized GO can be easily separated when subjected to an external magnetic field. This method ensures quick and easy separation from aqueous solutions and improves the contaminants adsorption capacity. The success of this technology depends on selecting an adequate magnetic adsorbent material that ensures both efficient separation and pollutant selectivity²³. It has been observed that adding GO layers to a cross-linked alginate matrix can reduce adsorbent leaching. During this technique, the GO layers are trapped within the beads²⁴. However, it is interesting to note that no chemical interactions between GO and the polymer matrix exist, and no other features are introduced to enhance the stability and adsorption capacity of the GO-based adsorbent.

Silver nanoparticles (AgNPs) stand out among metal nanoparticles due to their unique morphology, nanoscale size, high thermal conductivity, and strong reactivity, all while being relatively abundant and cost-effective^25,26. These characteristics render AgNPs exceptionally adaptable for water purification applications²⁷., with minimal toxicity risks for humans. For instance, Issac et al.²⁸ demonstrated the time-dependent adsorption efficiency of AgNPs, achieving up to 92.92% removal of Pb(II) and 53.34% of Co(II) over 14 days, confirmed via ICP-OES. Similarly, Sumesh et al.²⁷. synthesized AgNPs stabilized with mercaptosuccinic acid (MSA) and supported on activated alumina, showing excellent Hg(II) removal (up to 0.8 g/g) at room temperature, particularly in the pH range of 5–6. These results underscore the potential of AgNPs as effective adsorbents for remediating polluted water.

According to the literature, magnetic and silver-based materials are frequently used as adsorbents in wastewater treatment. For example, Luo et al.²⁹ developed magnetic reduced graphene oxide (M-rGO) using a co-precipitation method and examined its adsorption properties for mixtures of silver ions and silver nanoparticles (AgNPs). Ankita et al.³⁰ similarly synthesized a TiO₂-rGO-Ag hybrid nanocomposite using a two-step process that involved solvothermal treatment of TiO₂-rGO in a water-ethanol mixture, followed by microwave-assisted deposition of AgNPs. Muradiye et al.³¹ reported the green synthesis of silver, magnetic iron/copper, and iron oxide nanoparticles using Lathyrus brachypterus extract, which were subsequently incorporated into alginate- and chitosan-based magnetic nanocomposite beads. In this study, silver nanoparticles were reduced in a magnetic field using Saccharomyces cerevisiae and later functionalized with GO to create Ag-GONA beads, which serve as effective adsorbents for removing heavy metal ions (Pb, Hg, and Cd) and organic contaminants (naphthalene, phenol, and fluorine) from refinery wastewater. The product obtained was subjected to comprehensive characterization through a series of methods, such as TEM, XRD, and SEM, to investigate the size, shape, crystallinity, and morphology of the Ag-GONA beads synthesized. Adsorption behavior was assessed through variation of initial concentration, contact time, pH values, and temperature concerning the magneto-synthesized beads. Adsorbent regeneration was investigated for cycling performance and long-term feasibility, resulting in a proposed practical adsorption process. Thus, this novel and eco-friendly adsorbent provides a good alternative for the effective treatment of refinery effluents, hence supporting environmental conservation and water quality management.

Results and discussion

Physicochemical characterizations

The textural and morphological properties of the as-synthesized Ag-GONA adsorbent were investigated using SEM. As shown in Fig. 1a, the micrograph reveals a uniform distribution of silver nanoparticles embedded within the sodium alginate matrix, indicating the successful incorporation of AgNPs. Notably, graphene nanosheets are densely interwoven with both the silver nanoparticles and the alginate framework, forming a compact and interconnected structure that can provide abundant active sites for adsorption. Moreover, the presence of SA promotes the development of irregular porous structures within the cross-sectional regions of the Ag-GONA beads. These irregularities, along with the interwoven network of graphene nanosheets, enhance surface roughness and pore accessibility, which are favorable for efficient interaction between the adsorbent surface and pollutant molecules. Fig. 1b presents the TEM image of the as-synthesized AgNPs, which exhibit predominantly spherical morphology with well-defined geometric boundaries and a narrow size distribution, averaging around 10 nm. Importantly, no significant aggregation was observed, highlighting the effectiveness of the Baker’s yeast-mediated synthesis in producing highly dispersed and stable nanoparticles. The uniformity in particle size and shape further confirms the reliability of this green synthesis approach for generating AgNPs with desirable physicochemical characteristics.

Fig. 1.

(a) SEM image of silver nanoparticles, (b) TEM image of graphene sheets and silver nanoparticles embedded within a sodium alginate matrix, (c) XRD analysis of Ag-GONA material, and (d) FTIR analysis of Ag-GONA material.

To investigate the crystalline structure of the composite adsorbent, XRD analysis was carried out in the 2θ range of 5°–70°, as shown in Fig. 1c. The diffraction pattern exhibited a distinctive (002) peak at 2θ = 9.98°, characteristic of GO, which corresponds to the (001) plane and reflects the presence of interlayer spacing between the graphene oxide sheets. A minor peak was also detected at ~ 42.5°, suggesting partial structural ordering within the composite. In addition, four well-defined diffraction peaks were observed at 2θ = 38.1°, 44.3°, 64.5°, and 77.7°, corresponding to the (111), (200), (220), and (311) planes of silver nanocrystals, respectively. These peaks are consistent with the standard Bragg reflections of a face-centered cubic (FCC) crystal structure, thereby confirming the successful formation of crystalline Ag nanoparticles within the Ag-GONA adsorbent^32–34.

Furthermore, the Ag-GONAs’ FTIR results were consistent with recent findings for sodium alginate GO composites (Fig. 1d). The spectrum showed minor dips caused by OH stretching vibrations at 3347 cm^− 1, stretching vibrations of C = O groups at 2500 cm^− 1, and OH vibrations at 1500 cm^− 1. Furthermore, the presence of C-O stretching vibrations at 1043 cm^− 1 validated the existence of reactive groups inside the sodium alginate matrix³⁵. The complete results reveal that silver and graphene nanoparticles are consistently intercalated into the sodium alginate matrix. This finding validated the durability and reliability of Baker’s yeast method for synthesizing AgNPs and the successful integration of these nanoparticles with graphene within a sodium alginate matrix.

Adsorption studies for heavy metal removal

pH plays a crucial role in the sorption process, as it governs both the speciation of metal ions in solutions and their interactions with the sorbent surface. Fig. 2a and Fig. 2e illustrate the effect of pH (2–13) on the removal efficiencies of Pb, Hg, and Cd using GONA and Ag-GONA beads. A significant increase in adsorption was observed as the pH increased from 2 to 7. For GONA beads, Pb removal increased from 25% to 70%, Hg from 15% to 52.9%, and Cd from 32.6% to a maximum of 79.8% before slightly decreasing to 72.5%. This trend can be explained by metal speciation: at low pH (pH 2–4), metal ions (Pb²⁺, Hg²⁺, Cd²⁺) exist predominantly in their free ionic forms, but strong competition with abundant H⁺ ions limits their interaction with the sorbent functional groups. As the pH approaches neutral, deprotonation of surface functional groups (–OH, –COOH, –NH₂) occurs, enhancing electrostatic attraction and complexation with metal cations, which explains the rise in removal efficiency. For Ag-GONA beads, further increasing the pH above 7 resulted in a moderate enhancement of Pb removal, reaching up to 90%. However, the removal efficiencies for Hg and Cd declined, dropping to 75% and 90%, respectively, at pH 13. This decline can be attributed to changes in metal ion speciation under alkaline conditions. At higher pH, Pb²⁺ may partially hydrolyze to Pb(OH)⁺, which still retains affinity for functional groups, whereas Hg²⁺ and Cd²⁺ readily form hydroxide precipitates, such as Hg(OH)₂, Cd(OH)₂, reducing their availability for sorption. Overall, between pH 2 and 7, the removal efficiencies improved markedly for all three metals: Pb from 35% to 89%, Hg from 20% to 80%, and Cd from 45% to 99.7%. These observations confirm that the adsorption mechanism is strongly dependent on metal ion speciation: at low pH, competition with protons dominates, while at near-neutral pH, electrostatic attraction and complexation with deprotonated surface groups prevail. At high pH, reduced affinity due to hydroxide complexation or precipitation explains the observed decline for certain metals.

Fig. 2.

Effect of (a) pH, (b) contact time, (c) initial metal ion concentration, and (d) temperature on the percentage removal of metal ions using GONA bead adsorbent; and (e) pH, (f) contact time, (g) initial metal ion concentration, and (h) temperature on the percentage removal of metal ions using Ag-GONA bead adsorbent.

Fig. 2b shows the influence of contact time on the removal efficiency of heavy metal ions by GONA and Ag-GONA beads (Fig. 2f). For GONA beads, the removal efficiency of lead and mercury increased continuously with time and reached 78% and 75.6%, respectively, after 24 h. In contrast, Cd showed a different trend in its removal efficiency; the equilibrium was achieved at 6 h (70.9%) and then declined to 62.7%, likely due to desorption or site saturation over prolonged contact. In the case of Ag-GONA beads, adsorption occurred more rapidly. Within the first three hours, removal efficiencies reached 81.3% for Pb, 85.7% for Hg, and 68.3% for Cd, indicating a strong affinity between the metal ions and the sorbent surface (Fig. 2f). After 6 h, adsorption equilibrium was achieved, with nearly complete removal: 99.8% for Pb, 96.11% for Hg, and 98% for Cd. The performance enhancement seen can be due to the increased number of active sites and the enhanced surface reactivity of Ag-GONA beads.

The concentration of the initial metal ions greatly influenced the adsorption efficiency of both GONA and Ag-GONA beads, as indicated by Fig. 2c,g. Typically, the concentration of the adsorbate in the aqueous solution influences the removal of metal ions. For GONA beads, raising the initial concentration from 10 mg/L to 110 mg/L resulted in a reduction of the removal efficiency for all the metal ions studied, namely Pb, Hg, and Cd, indicating a saturation of the available active sites. Specifically, Pb removal decreased from 75.9% to 68%, Hg from 68% to 62.1%, and Cd from 70.8% to 65.9%. In contrast, Ag-GONA beads demonstrated superior performance and stability across wider concentration ranges. For Pb ions, removal remained remarkably high, decreasing only slightly from 99% to 98.56% as concentration rose from 10 to 110 mg/L, indicating strong adsorption affinity and sufficient active sites. Similarly, Hg removal initially declined from 94% to 90.89% between 10 and 56 mg/L but improved to 94.7% at 109 mg/L, likely due to the exposure of additional active sites at higher concentrations. A comparable trend was observed for Cd, with removal decreasing from 90% to 84% between 10 and 25 mg/L, then increasing to 88% at 109 mg/L.

Temperature influences on the adsorption efficiency of GONA and Ag-GONA beads were explored systematically in the temperature range from 298 K to 328 K, as presented in Fig. 2d,h. Both types of beads displayed the same trend of reduced metal ion removal efficiency at elevated temperature, reflecting the exothermic nature of the adsorption process. For GONA beads, the removal efficiency of Pb declined from 74.5% to 70.2%, Hg from 68% to 64.2%, and Cd from 62% to 56.6%. Similarly, Ag-GONA beads showed a slight but consistent decrease in removal efficiency: Pb decreased from 99.7% to 98%, Hg from 95.8% to 94%, and Cd from 90.66% to 88%. This decline suggests that higher temperatures negatively impact adsorption performance, likely due to increased kinetic energy of metal ions, which disrupts the interaction between ions and active adsorption sites. The overall decrease in removal efficiency with temperature confirms the exothermic and physiosorption-driven nature of the process, making lower temperatures more favorable for optimal metal ion removal.

Adsorption studies for organic contaminants

The solution pH has a great impact on the adsorption process of the organic pollutants, controlling an immense number of parameters like the surface charge of the adsorbent material, the ionization state of the pollutants, the dissociation of functional groups at the active sites of the adsorbent, and the adsorbate structure. Fig. 3(a,e) illustrate the pronounced effect of pH (ranging from 2 to 13) on the adsorption of naphthalene, phenol, and fluorene organic pollutants by GONA and Ag-GONA beads. A remarkable enhancement in the removal efficiency was noted with a change in pH from acidic to basic. In the case of GONA beads, naphthalene removal rises from 20% at pH 2 to 52.7% at pH 7 and then decreases slightly to 50% at pH 13. Phenol and fluorene, on the other hand, exhibited a steady rise in the removal efficiency, from 8% to 62.8% and 5% to 61.7%, respectively, over the same pH range. Ag-GONA beads outperformed GONA beads, exhibiting improved removal across all pH levels. Specifically, naphthalene removal increased from 35% to 80%, phenol from 14% to 76%, and fluorene from 9% to 75% as pH rose from 2 to 7. At pH 13, naphthalene removal slightly declined to 79%, while phenol and fluorene reached 80% removal. These results suggest that increasing pH enhances the availability of active sites and favors electrostatic interactions between the adsorbent surface and deprotonated organic molecules. Additionally, the improved adsorption under alkaline conditions may be attributed to changes in pollutant speciation, enhancing the affinity between organic molecules and the sorbent.

Fig. 3.

Effect of (a) pH, (b) contact time, (c) initial metal ion concentration, and (d) temperature on the percentage removal of organic containments using GONA bead adsorbent; and (e) pH, (f) contact time, (g) initial metal ion concentration, and (h) temperature on the percentage removal of organic containments using Ag-GONA bead adsorbent.

Contact time significantly influences the adsorption effectiveness of organic pollutants on Ag-GONA beads, highlighting the dynamic relationship between the duration of adsorption and the level of contaminant uptake. Fig. 3b demonstrates how contact time affects the adsorption of organic contaminants with GONA beads. The removal rate of naphthalene increased consistently from 0% to 67% as contact time progressed from 0 to 24 h, indicating a gradual filling of adsorption sites. In contrast, phenol and fluorene reached their maximum removal efficiencies within the first 6 h, rising from 0% to 51% and 0% to 30%, respectively, after which their adsorption efficiency declined, dropping to 41% for phenol and 30% for fluorene by 24 h. This suggests a potential desorption or equilibrium shift at prolonged exposure. A similar trend was observed for Ag-GONA beads, where rapid adsorption occurred within the first 6 h, achieving removal efficiencies of 86% for naphthalene, 69.5% for phenol, and 59% for fluorene (Fig. 3f). However, further increase in contact time beyond 6 h led to a slight decline in removal, indicating saturation of available active sites. This behavior reflects a typical adsorption equilibrium pattern, where prolonged contact may lead to reduced driving force for adsorption or partial desorption of previously bound molecules.

The influence of initial organic pollutant concentrations, ranging from 5 to 30.78 mg/L, on adsorption efficiency was examined for both GONA and Ag-GONA beads, as shown in Fig. 3(c,g). A consistent decline in removal efficiency was observed with increasing contaminant concentration, indicating a possible saturation of available adsorption sites. For GONA beads, the removal of naphthalene decreased from 62.8% to 52.7%, phenol from 46.8% to 35.8%, and fluorene from 40.8% to 30.2%. Similarly, Ag-GONA beads exhibited higher overall performance but followed the same downward trend: naphthalene removal decreased from 84% to 74%, phenol from 64.7% to 49%, and fluorene from 53.1% to 42.36%. This reduction in efficiency at higher concentrations is because the limited number of active sites becomes saturated as more pollutant molecules are added. In addition, the effect of temperature on the treatment of organic pollutants using GONA and Ag-GONA beads was investigated systematically in a temperature range of 298–328 K. Fig. 3d illustrates the influence of temperature on the efficiency of organic contaminant removal using GONA beads, and it decreased gradually with increasing temperature in the range from 298 K to 328 K. Specifically, naphthalene removal dropped from 65% to 60.5%, phenol from 45.2% to 40.2%, and fluorene from 39% to 31.9%. A similar trend is observed in Fig. 3h for Ag-GONA beads, where naphthalene removal decreased slightly from 83% to 81%, phenol from 64% to 63%, and fluorene from 53% to 50%. This downward trend validates the exothermic nature of the adsorption process, in which rising temperature decreases the efficiency of adsorption by loosening the interaction between the adsorbent and impurities. The decrease can also be explained by the higher kinetic energy at higher temperatures, which disrupts the equilibrium between adsorbate and absorbent and leads to fewer surface bindings. Understanding this temperature dependency is essential for optimizing operational conditions and maximizing the adsorption capacity of Ag-GONA beads in practical applications.

Adsorption isotherm

Adsorption isotherms are widely applied to evaluate sorbent efficiency and interpret pollutant interactions³⁶. These models provide information on adsorbent capacity and surface properties. Two well-known adsorption isotherm models, the Freundlich and the Langmuir, were fitted to the equilibrium data. These mathematical models were utilized to illustrate adsorption patterns and assess their effectiveness in interpreting experimental data, providing a comprehensive understanding of the process.

Langmuir isotherm

The Langmuir isotherm explains that the adsorption occurs on a homogeneous surface via monolayer sorption without considering any interactions among the adsorbed molecules³⁷. In this case, the adsorption sites have equal affinities for the adsorbate, and any movement of the adsorbate on the surface of the adsorbent is not allowed³⁸. The Langmuir equation is represented in linearized form as follows:

1

Here, C_e stands for the solution’s equilibrium concentration, and q_m and q_e, respectively, denote the maximum and equilibrium adsorption capacities. Plotting C_e/q_e against C_e using Eq. 1 allows for a linear regression analysis and determines q_m and the Langmuir constant K_L. The separation factor (K_L), despite being dimensionless, is an essential element in the Langmuir isotherm. The following formula is used to determine the Langmuir constant (K_L),

2

Freundlich isotherm

The Freundlich isotherm indicates multilayer adsorption on a heterogeneous surface³⁹. The linearized expression for the Freundlich equation is as follows:

3

The Freundlich constants “n” and “K_F” represent the adsorption intensity and distribution coefficient, respectively. These constants provide information on the bonding energy related to the adsorption process.

Isotherm studies

The adsorption between metal ions and organic pollutants on the Ag-GONA bead surface was well examined from experimental data of the above isotherms, as shown in Fig. 4, 5. The R² value was utilized to evaluate the authenticity of experimental data, and Table 1 summarizes the relative isothermal parameters. In addition, the RL values of the Langmuir isotherm for both the metal ions and organic pollutants ranged from 0 to 1, which is a sign of favorable adsorption conditions and further confirmation of the Langmuir model being superior to the Freundlich model. Furthermore, the Langmuir model’s higher fit resulted in monolayers of adsorbate on the Ag-GONA beads’ homogeneous surfaces. Additionally, under these conditions, the Ag-GONA beads’ n values (1 < n < 10) in the Freundlich model indicated favorable adsorption. Additionally, as Table 1 illustrates, the Langmuir model showed that Ag-GONA beads had the highest adsorption capacity for both metal ions and organic contaminants. Thus, the developed adsorbents exhibited superior adsorption performance compared to previously reported materials, making them suitable for refinery wastewater treatment.

Fig. 4.

Isotherm model for metal adsorption (a) Langmuir (b) Freundlich.

Fig. 5.

Isotherm model for the adsorption of organic contaminants using (a) Langmuir and (b) Freundlich isotherms.

Table 1.

Adsorption of metal ions and organic concentrations using the Langmuir and the Freundlich model parameters.

Isotherm model	Adsorbate	Isotherm model parameters
		qmax (mg/g)	RL	KL	R2
Langmuir	Pb	326	0.02 ≤ RL≤0.17	0.502	0.98633
	Hg	300.37	0.12 ≤ RL≤0.59	0.086	0.96505
	Cd	219.13	0.10 ≤ RL≤0.54	0.070	0.96205
	Naphthalene	71.93	0.20 ≤ RL≤0.59	0.1278	0.99262
	Phenol	58.11	0.37 ≤ RL≤0.78	0.0561	0.99658
	Flourene	67.77	0.39 ≤ RL≤0.76	0.0303	0.98547
Freundlich		KF	n	R2
	Pb	121.18	1.10	0.9875
	Hg	22.11	1.016	0.9084
	Cd	12.81	1.063	0.9578
	Naphthalene	8.366	1.33	0.9846
	Phenol	3.87	1.40	0.9621
	Flourene	2.84	1.36	0.9412

Adsorption kinetics

Kinetic studies investigate the adsorption rate in aqueous media to determine the rate-controlling step as well as the adsorption capacity. Pseudo-second order (PSO) analysis was used to comprehensively assess the experimental data. By improving our knowledge of adsorption dynamics, this model makes it easier to design and optimize processes effectively for a range of applications.

Pseudo 2nd -order kinetics model

This kinetic model was based on the principle of chemisorption, in which contaminants and adsorbents undergo chemical reactions that tend to form strong covalent bonds. A PSO kinetic model accurately simulates the adsorption kinetics and phenomena in the removal of metal ions and organic pollutants⁴⁰. Numerous studies have used the PSO kinetic model for studying adsorption kinetics^41,42. Equation 4 represents the linear form of PSO,

4

Where, k₂ (g/mg/min) is the second-order adsorption rate constant and q_t (mg/g) is the adsorption capacity at time ‘t’.

Kinetic studies

To analyze the effect of various adsorption kinetic parameters on the Ag-GONA bead adsorbent, linear regression analysis was utilized. The parameters of the PSO kinetic model are presented in Table 2 and Fig. 6, with the best data set fitting having a correlation coefficient (R²) value of approximately 0.99. Assuming the rate-determining process, generally chemisorption or electron transfer between adsorbent and adsorbate, permits the adsorption of both metal ions and organic pollutants.

Table 2.

PSO kinetic model parameters for metal ion and organic containment adsorption on Ag-GONA beads.

Kinetic model	Adsorbate	Co (mg/L)	qexp (mg/g)	qe (mg/g)	k2 (g.mg− 1.min− 1)	R 2
PSO	Pb			101.94	0.000187	0.99729
	Hg			84.25	0.000808	0.99962
	Cd			80.32	0.000406	0.99625
	Naphthalene			20.59	0.00126	0.99391
	Phenol			17.63	0.00054	0.9909
	Flourene			2.19	0.02365	0.99714

Fig. 6.

PSO kinetic model for (a) metal ions and (b) organic compounds.

Regeneration studies

For functionalized Ag-GONA beads to be used for an extended period, adsorbed metal ions and organic contaminants must be effectively removed, and their capacity to regenerate is an essential feature of their economic viability. Following the adsorption study, the beads were immersed in a solution of hydrochloric acid (HCl) at pH 5 and 200 mg/L calcium chloride for 3 h. This procedure aims to regenerate the beads by preparing them for additional adsorption cycles. In an acidic environment, the protonation of acetate ions within the alginate polymer may have hindered cross-linking within the beads. Concerns of its mechanical stability were relieved by the stabilizing action of introducing 200 ppm CaCl₂. The measured removal rates were 99.9%, 98%, 93%, 83%, 82%, and 78%, respectively, in six consecutive cycles, indicating that the regeneration efficiency of Pb(II) ions was outstanding in each adsorption/desorption cycle. Remarkably, even after six cycles, the removal efficiency of Pb(II) always fell within 75–80%, as illustrated in Fig. 7. The same trend was noticed for all the other ions tested. The results indicate that the synthesized beads possess an excellent ability to adsorb heavy metal ions from polluted water bodies regularly, thereby facilitating long use and reusability over a series of cycles. Similarly, the regeneration capacity of Ag-GONA beads for organic pollutants was assessed. Over six cycles, the naphthalene, phenol, and fluorene removal efficiencies decreased from 88%, 86%, and 86.1% to 63%, 66%, and 61%, respectively. The findings demonstrated a slight drop in the removal % of Ag-GONA beads, confirming their long-term economic feasibility for the removal of contaminants from refinery wastewater.

Fig. 7.

Regeneration of Ag-GONA beads for metal ions and organic contaminants.

Adsorption mechanism

The Ag-GONA beads exhibit a unique porous architecture with a surface rich in diverse functional groups, making them highly effective for the removal of toxic chemicals from refinery wastewater and related ecosystems. FTIR analysis (Fig. 1d) confirmed the presence of various oxygen-containing functional groups, including hydroxyl, amino, carbonyl, and carboxyl groups, all of which play a vital role in adsorption. These groups act as active binding sites, either by exchanging hydrogen ions in solution with pollutant species or by donating electron pairs to form coordination complexes with metal ions, such as Cd(II). Such interactions provide multiple pathways for contaminant removal, including surface adsorption, chemical precipitation, and ion exchange. The adsorption mechanism of Ag-GONA beads is illustrated in Fig. 8, which shows the interaction between both the metal ions and organic pollutants with the adsorbent surface. The adsorption of organic molecules is facilitated by weak and reversible van der Waals interactions, while metal ion binding involves stronger chemisorption processes, which have been identified as the primary rate-controlling step based on adsorption kinetics. The applicability of both the Langmuir and Freundlich isotherm models further confirms that adsorption occurs through a combination of monolayer and multilayer processes. Morphological analysis by SEM (Fig. 1a) revealed that the incorporation of sodium alginate promoted the development of irregular porous structures throughout the bead cross-sections. These irregularities, coupled with the intertwined network of graphene nanosheets, significantly enhance the surface roughness and pore accessibility, facilitating efficient pollutant-adsorbent interactions. Additionally, chemical precipitation occurs through redox reactions involving surface functional groups, while ion exchange is achieved by replacing bound ions on the adsorbent surface with competing ions from the solution. So, the synergy of porous morphology, abundant functional groups, and the combined mechanisms of adsorption, precipitation, and ion exchange make Ag-GONA beads a robust and versatile material for the elimination of both heavy metal and organic pollutants.

Fig. 8.

Adsorption mechanism of Ag-GONA beads on metal ions and organic contaminants.

Adsorption comparative analysis

The adsorption capacities of the Ag-GONA beads were evaluated and compared to those of other adsorbents, such as GO membranes⁴³, calcium alginate⁴⁴, Saccharomyces cerevisiae³⁷, activated carbon⁴⁶ and other magnetic nanoparticles like rGO/magnetite/silver⁴⁷, Fe₃O₄–CS³¹ and green mCS/GO⁴⁸, as shown in Table 3. Evidence on GO-alginate and double-network alginate hydrogels shows durable integrity over 5–10 regeneration cycles when crosslinking and pH are controlled. Accordingly, the Ag–GONA protocol maintained pH 5.5–7.5, avoided phosphate and strong chelators, applied post-curing in 0.2–0.5 M CaCl₂, operated under moderate shear, with no collapse observed after ≥ 5 cycles and quantified capacity retention, collectively supporting the mechanical integrity and reusability of Ag–GONA beads under the stated recycling conditions. The findings demonstrated that Ag-GONA beads are efficient, cost-effective, and novel adsorbents for removing hazardous metallic ions and organic contaminants from refinery wastewater. This shows a potential use for removing pollutants from wastewater.

Table 3.

Comparative analysis of the adsorption capacity of Ag-GONA beads to that of other adsorbents.

Adsorbent	Adsorbate	Contact time (h)	Temperature (K)	Adsorption capacity (mg/g)	Ref.
GO membranes	Cd	0.25	333	83.8	43
Calcium alginate	Pb	7	308	120	44
SC-HIS-CCBs	Phenol	0.33	313	28.4	49
Saccharomyces cerevisiae	Phenol	4	-	27	45
Biomass derived carbon	Naphthalene	72	298	85	50
Activated carbon	Pb	1.66	298.15	476	46
rGO/magnetite/silver	Cd	30	298	3.441 (mmol/g)	47
Fe3O4–CS	Pb	120	298	9.839	31
Fe3O4–CS	Cd	120	298	5.154
green mCS/GO	Naphthalene	60	303	340.3439	48
Ag-GONA beads	Pb	6	298	326	This study
	Hg	6	298	300.37	This study
	Cd	6	298	219.13	This study
	Naphthalene	6	298	71.93	This study
	Phenol	6	298	58.11	This study
	Flourene	6	298	67.77	This study

Materials and methods

Materials

S. cerevisiae, also called Baker’s yeast, and sodium alginate, calcium chloride (CaCl₂), methanol, tetrahydrofuran (THF), and polyethylene glycol (PEG) were all procured from Sigma-Aldrich. Mercury(II) chloride (HgCl₂) was procured from R.P. Norampac, cadmium(II) chloride (CdCl₂·2.5H₂O) from LABCHEM Products, and lead(II) nitrate (Pb(NO₃)₂) from Fisher Scientific. GO was synthesized via a modified Tour procedure using graphite flakes with a particle size of 10 μm. The real refinery wastewater in the Middle East and North Africa (MENA) region was collected and stored at 278 K.

Synthesis protocols

S. cerevisiae cells were cultivated in a growth medium supplemented with 5 g/L of biomass. Cultivation was performed aerobically at 30 °C and pH 6.5. Following cultivation, the S. cerevisiae cells were collected by centrifugation, and any leftover media or impurities were cleaned with distilled water.

Preparation of silver salt solution

The silver salt precursor was produced using 1 mM silver chloride (AgCl) dissolved in 50 mM NaCl and buffered to pH 6.5 with 20 mM MOPS. S. cerevisiae cells (OD₆₀₀ ≈ 1.5; ~5–10 g wet biomass/L) were washed and mixed with a silver salt solution in a reaction vessel. The solution was magnetically stirred at 300 rpm to ensure a uniform distribution of the yeast cells. The foil-wrapped vessel was maintained at 30 °C and magnetically stirred at 300 rpm. A 500 Gauss magnetic field (with a parallel magnet placed 2 cm from the flask) was applied for 4 h. The combination of yeast cells and a magnetic field helped reduce silver ions, resulting in the development of AgNPs. Yeast cells act as both reducing agents and stabilizers during nanoparticle synthesis. The reaction mixture was analyzed after exposure to a magnetic field, which confirmed the successful synthesis of AgNPs using Saccharomyces cerevisiae under the specified experimental conditions. Fig. 9 demonstrates a schematic of nanoparticle synthesis under a magnetic field, using Saccharomyces cerevisiae as a source of baker’s yeast.

Fig. 9.

Magnetic synthesis of silver nanoparticles from Saccharomyces cerevisiae (Baker’s yeast).

Bead preparation

Batch adsorption studies were carried out in a shaker bath operated at 100 revolutions per minute. Each test run contained 2.5 g of wet functionalized Ag-GONA beads, 0.2 g of alginate, and 0.025 g of graphene oxide, all loaded into 50 mL polyethylene terephthalate tubes. The tubes were totally immersed in 20 mL of adsorbate solution containing either discrete metal ions or target quantities of Hg(II), Cd(II), Pb(II), and organic pollutants. The temperature and shaking time differed according to the experiment, with contact times ranging from 1 to 24 h. Adsorption beads were developed by mixing GO and a sodium alginate matrix. Iterative trials were utilized to determine the amounts of GO, sodium alginate (SA), and polyethylene glycol (PEG). Fig. 10 schematically represents different steps involved in the proposed bead preparation. Briefly, a GO solution in deionized water was made by dispersing 25 mg of GO in 5 mL. Following adequate dispersion, 0.2 g sodium alginate was put into the GO solution, resulting in a thick, light-brown liquid. This mixture was added dropwise to a beaker of 5 wt% calcium chloride solution, forming immediately spherical, light brown beads. The beads were left immersed in a calcium chloride solution for approximately three hours prior to being rinsed extensively with water. During crosslinking, divalent calcium ions substituted for monovalent sodium ions, permitting carboxylate groups to bind. The beads acquired had a dry weight of 0.2 g of GO and alginate (0.025 g) and a wet weight of approximately 2.5 g. Afterwards, the beads were heated for 20 h at 40 °C while immersed in a polyethylene glycol solution (250 mg in 5 mL of water). GONA beads were synthesized without the addition of silver nanoparticles for the control adsorbent study. During this process, PEG functionalizes and chemically reduces the GO. Finally, excess amine was removed from the functionalized beads by Soxhlet extraction with methanol, yielding blackish-brown beads. In conclusion, Ag-GONA beads were prepared by integrating GO into a sodium alginate matrix via crosslinking with calcium chloride, followed by functionalization and chemical reduction using PEG and Soxhlet extraction to eliminate excess amine.

Fig. 10.

Bead preparation: (i) silver nanoparticles and GO dispersed in water, (ii) beads made in CaCl₂ solution, and (iii) beads functionalization in PEG solution.

Characterization of adsorbent beads

FTIR analysis was carried out using a Bruker Tensor II instrument equipped with a diamond attenuated total reflection (ATR) accessory, a deuterated L-alanine-doped triglycine sulfate (DLaTGS) detector with a potassium bromide (KBr) window, and a KBr beam splitter. The level of crystallinity and the structural defects were analyzed through an X-ray diffractometer (XRD) diffractometer (D8 Advance, Bruker, USA). Infrared spectra were recorded in the range of 4000 cm⁻¹ to 600 cm⁻¹ using OPUS software with a resolution of 4 cm^− 1 from the average of 32 scans. Following deposition of Au-Pd coating via plasma at 25 mA current for 40 s, samples were analyzed with an FEI Quanta 250 system. Furthermore, the surface morphology, crystalline structure, and growth of the nanomaterials were examined using high-resolution transmission electron microscopy (HR-TEM: Titan TEM 300 kV).

Batch adsorption experiment

Adsorption studies were conducted by varying the GO concentration in water from 2.5 to 20 mg/mL, and adsorption efficiency was evaluated at pH values of 2, 7, and 13 at 298 K. Following the necessary adsorption time, syringes were used to carefully collect the supernatant, which was then filtered using 0.2 μm PTFE syringe filters. After adsorption, the supernatant was carefully collected using syringes and filtered through 0.2 μm PTFE syringe filters. Adsorption was followed by the quantification of the adsorbate in filtered solutions using inductively coupled plasma optical emission spectroscopy (ICP-OES). The equilibrium adsorption capacity (q_e) and percent removal were determined using the following mathematical equations:

5

6

C_o and C_f, represent the initial and final adsorbate concentrations in mg/L. The parameters M and V represent the adsorbent mass (mg) and adsorbate solution volume (mL), respectively.

Regeneration studies

For the desorption experiments, pre-adsorbed Pb (II), Hg (II), and Cd (II) beads, and organic pollutants such as naphthalene, phenol, and fluorene, were washed with distilled water to desorb any unadsorbed metal ions. The beads were immersed in 20 mL of eluent comprising HCl (pH adjusted to 5) and a 200 ppm CaCl₂ solution. To desorb the toxic components, the solution was mechanically shaken on 1,00 rpm at room temperature for three hours. These beads were collected, rinsed with distilled water, and employed for additional adsorption research.

Conclusion

Silver-modified graphene composite beads effectively removed heavy metal ions and organic contaminants from refinery wastewater. At pH 7 at 398 K, the optimum adsorption capacities for various compounds were as follows: naphthalene (71.93 mg/g), phenol (58.11 mg/g), fluorine (67.77 mg/g), Pb (326.77 mg/g), Hg (300.37 mg/g), and Cd (219.13 mg/g). The adsorption process demonstrates a PSO mechanism, and the Langmuir model explains the physical and chemical interactions that occurred during monolayer adsorption. The presence of polyamines for complexation and functional groups in GO resulted in improved adsorption, and the carboxylate groups on the alginates. Adsorption-desorption experiments conducted over six cycles showed consistent removal effectiveness, especially for Pb(II), suggesting its potential for wider use in water treatment. Therefore, this developed material has long-term applicability and shows promise in effectively treating refinery wastewater.

Author contributions

Conceptualization: SS, AH, FB, Methodology: SS, LJ, Supervision: FB, AH, KA, Validation: SS, LJ, NR, KA, VS, Visualization: SS, LJ, NR, Resources: FB Writing original draft: SS, AH, Writing review and edits: FB, KA, AH.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.