Humic acid and nano-zeolite NaX as low cost and eco-friendly adsorbents for removal of Pb (II) and Cd (II) from water: characterization, kinetics, isotherms and thermodynamic studies

The effect of pH on the sorption capacity of metal ions was determined experimentally in the range of pH 2 to 7 at room temperature. A 10 ppm as an initial Pb²⁺ and Cd²⁺ concentration was prepared in a 50 mL measuring flask and the pH was adjusted using standard buffer solutions. The pH-adjusted metal ion solution was then added to 20 mg of green-synthesized sorbents and shacked automatically at a rate of 150 rpm for 30 min. After equilibrium was achieved, the samples were filtered and heavy metal concentrations were analyzed.

The effect of the green-sorbent dosage on the metal sorption capacity has been investigated to determine the optimal dose. Various weights of 10, 20, 30, 40, 60, 80 and 100 mg of eco-sorbents for Pb²⁺ removal, and 20, 40, 60, 80, 100, 120, 140, 160, 180 and 200 mg of eco-sorbents for Cd²⁺ removal have been added to a different series of 25 mL of 10 ppm of selected metal ions solution while maintaining the pH value of selected metal ions and specific sorbent constant. The different series were shacked at 150 rpm for 30 min at room temperature.

The optimal contact time for sorption of Pb²⁺ and Cd²⁺ was assessed at room temperature and 150 rpm stirring rate over a range of time intervals; 10, 20, 30, 40, 50 and 60 min, on 25 mL of working metal ions solution with a concentration 10 ppm for each metal ion at certain green-sorbent dosage and pH.

The impact of initial metal concentration (Cd²⁺ and Pb²⁺) on adsorption onto the sorbents at optimal pH, doses, contact time and metal concentrations of 5, 10, 15, 20, 25, 30, 40, 60, 80 and 100 ppm were examined in 25 mL deionized water. The remaining heavy metal concentrations at equilibrium were analyzed. The results were fitted to different isotherm and kinetic models.

Detailed fundamental FT-IR bands of the synthesized sorbents are given in Figs. 2 and S1. The spectrum of free nano-zeolites NaX showed six main absorption bands (Fig. 2a). The stretching vibration of hydroxyl groups (Si-OH) which are present in the faujasite super and sodalite cages as the building blocks of the zeolite structures causes the band at 3449 cm⁻¹ [26]. Also, the well-defined band at 1635 cm⁻¹ is attributed to OH bending vibration (M-OH-M, where M = Si, Al) since it coexists with OH stretching vibration positioned at 3449 cm⁻¹ [27]. Moreover, the sharp IR band at 1036 cm⁻¹ and the less intense band at 789 cm⁻¹ are assigned to the asymmetrical and symmetrical stretching of (M-O-M) group. Besides, the band in the spectral region 538–695 cm⁻¹ is due to external linkage vibrations, which indicates the association of rings that connect the sodalite cages (Si-O-Na). The sharpness of this peak implies the crystallinity of zeolite X [27, 28]. Furthermore, the band at 469 cm⁻¹ is characteristic of O-M bending vibration of the SiO₄ and AlO₄ in the internal tetrahedral structure [29]. Remarkably, the band at 1036 cm⁻¹ of nano-zeolite (Fig. 2a) exhibited changes in the intensity and shifts in position in the spectra of the metal exchanged samples (to 915 and 917 cm⁻¹ for NaX-Cd²⁺ and NaX-Pb²⁺, respectively, Fig. 2b, c). This is probably due to the Ḿ-O (Ḿ = Pb²⁺ or Cd²⁺) bond formation which alters the vibrational frequencies depending on the mass, charge, ion size and cation environment [30]. As well, the disappearance of some bands in the metal exchanged spectra can provide evidence for structural substitution of the Al atoms or dealumination. Similarly, the diverse functional groups present in the free HA were assessed by FT-IR (Fig. S1a). The spectrum of HA showed bands at 3425 cm⁻¹ corresponds to the stretching vibration of the O–H group in –COOH as well as alcoholic and phenolic –OH groups [19]. The absorption bands in the range 2920–2851 cm⁻¹ could be attributed to aliphatic C–H stretching vibrational mode in methyl and/or methylene groups [31]. Also, the band at 1583 cm⁻¹ is ascribable to aromatic COO⁻, C=O of carbonyl and quinone stretching vibration [32]. The sharpness of the bands approved the good extraction and purification of humic acid. Additionally, the FT-IR spectra of the M(II)–HA complex give information about the binding modes of HA with the metal ions. For instance, the band at 1583 cm⁻¹ in HA was shifted to 1573 cm⁻¹ and 1560 cm⁻¹ after M(II) treatment by Cd²⁺ and Pb²⁺ ions, respectively, indicating the involvement of oxygen in COOH groups in the interaction with the M(II) ions (Fig. S1b, c). The new IR bands emerged at 1381 cm⁻¹ and 1361 cm⁻¹ in the Cd²⁺ and Pb²⁺ spectra are assigned to the M-OH stretching vibrations which indicated the interaction between M(II) and OH groups of the surface [33]. Also, the band at 1041 cm⁻¹ is attributed to –CO stretch of polysaccharides and Si–O from silicate impurities [34]. Besides, the peak at 536 cm⁻¹ (assigned to C–O–C bonding in the free HA) was slightly shifted to 539 cm⁻¹ and 527 cm⁻¹ and exhibited notable intensity change after loading the sorbent surface with Cd²⁺ and Pb²⁺, respectively. Therefore, the observations of the current FT-IR study indicated that -COOH, -OH and -CO functional groups present in HA are primarily involved in M(II) binding via ionic interactions on the sorbent's surface as reported by previous studies on HA [35].

Qualitative mineralogy of the synthesized sorbents was determined with the standard interpretation procedures of XRD, where the position of a diffraction peak that represents a lattice plane can be identified by using a database of diffraction patterns. The XRD spectra measured in a range of 2θ from 5° to 70° (Fig. 3). The green synthesized nano-zeolite NaX structure has achieved a crystallinity percentage of about 88.12% after 4 days of ageing at room temperature, indicating amorphous aggregated cubic crystals. The XRD peaks at 2θ = 6.15°; 9.64°; 17.97°, 20.15°, 22.88°, 26.79°, 28.77° and 37.86° indicated the formation of nano-zeolite NaX crystals [17, 36]. These peaks may be indexed as the diffraction of the planes (111) (220) (511) (440) (620) (733) (822) (1022), respectively (JCPDF File No. 39-0218) [37] which indicate the cubic crystals of nano-zeolite. However, the peaks at 2θ = 20.76° and 35.57° were attributed to kaolin which suggests the incomplete dissolution of kaolin with sodium hydroxide during the preparation of gel mixture. As zeolites are thermodynamically metastable phases, the crystallization time allowed the formation of a mixture of Na-X and NaP types. The peaks at 2θ = 28.1°, 38.1° and 46.1° are the characteristic peaks of NaP zeolite [36, 38]. The formation of NaP in the synthesis of Na–X zeolite was also reported by [39]. On the other hand, the X-ray diffraction pattern of the extracted HA exhibited wide bands at 2θ = 20.5° which is attributed to the planes of aliphatic carbon chains and at 2θ = 25.5°, attributed to the structure of condensed aromatic carbon rings (Fig. S2). The diffractogram also showed many sharp peaks which represent the crystalline structures of some minerals such as silicon oxides [40‐42].

EDX is a semi-quantitative technique, where the elemental composition in weight and atomic percentage is obtained along with the spectra and peak area for all sample components. EDX analysis of green-synthesized NaX indicating the presence of Si, Al, Na, C and O (Fig. S3; Table S1) with silicon ion and oxygen as the main elements and sodium as stabilizer cation for zeolite framework [27]. Typically, nano-zeolite NaX is synthesized with Si/Al ratio of 1 [43]. The Si/Al and Na/Al ratios in the current study are 1.16 and 0.08, respectively. Thus, the formation of zeolites NaX was confirmed by the EDX techniques [27, 44].

Likewise, the elemental distribution in the extracted HA was obtained by EDX, as shown in Fig. 4 and Table S1. The major elements observed were C, O, Si, Al, K, Ca and Na, with some other minor elements such as S, Cl and Mg [45]. The presence of Si in HA may be the possible site for the interaction with the metal ions. The combination use of EDX with SEM techniques provides an accurate spatial resolution that is commonly used for surface analysis. In the present study, the morphologies of the synthesized nano-zeolite NaX were characterized by SEM analysis, as presented in Fig. 5. Nano-zeolite NaX appeared as cubic crystals with average sizes of 34–41 nm (Fig. 5a). Additionally, SEM images of HA demonstrated that Pb²⁺ and Cd²⁺ are adsorbed on the surface by replacing aluminium ions (Fig. 5e, f) [33]. Previous reports also suggested that metal ions adsorb to the surface of clay minerals by replacing Ca and Mg ions by ion exchange mechanism [46].

The pH is a critical criterion for determining the adsorption capacity of materials. Metal ion solutions were prepared at different pH levels ranging from pH 2 to 7 [16]. Table 1 shows the change in Cd²⁺ and Pb²⁺ uptake by nano-zeolite NaX and HA at different pH levels. Clearly, the adsorption capacity of nano-zeolite NaX to metal ions increased with the pH of the solution. Whereas, in the case of HA, the adsorption capacity of Cd²⁺ and Pb²⁺ ions increased to its maximum when the initial pH of the solution raised from pH 2 to 4 after that the adsorption performance of HA decreased significantly. Although the strong polarization of heavy metal cations could increase the sorption ability, this force cannot change the charge density of the sorbents [47] and hence, the strong polarization could not explain the large adsorption capacity of the heavy metal cations. Normally, variation of pH leads to physisorption and chemisorption on the adsorbent surface [48]. Nano-zeolite NaX has a great number of terminal silica tetrahedrons in their external surfaces and consistently a great number of negative silanol groups which enables it to have a substantial attraction with inorganic cationic contaminants [16, 49]. For nano-zeolite NaX, the lowest adsorption of all metals was obtained at the lowest pH values (Table 1). This may be due to the increase in the competition for adsorption sites by hydrogen (H⁺) and the existence of aluminium (Al³⁺) ions which is dissolved from the aluminosilicate layers [50]. Natural zeolite preferentially adsorbs H⁺ ions from solution forming a positively charged surface [51], and thus in more acidic conditions more H⁺ ions are adsorbed from the solution. The increase in positive charge on the surface sites of zeolite causes electrostatic repulsion with the metal ions. The process can be described by the following reactions [52].

Humic acids are composed of combinations of weak aliphatic carbon chains and aromatic carbon rings that are not soluble in acidic medium but are soluble in alkaline medium [52]. The pH value of the solution is often the most significant parameter affecting metal adsorption on the HA surface. Data showed that Cd²⁺ and Pb²⁺ adsorption capacities on HA peaked at pH 4 and 3, respectively, and subsequently began to drop (Table 1). Excess H⁺ compete with heavy metals for superficial sites of HA when the pH of the solution is lower than the optimum value [53]. Deprotonation of humic acid particles causes the surface to be negatively charged and increases the electrostatic attraction with the surrounding metal ions [11]. The deprotonation of HA surface sites at low pH can be described by the reaction: HA(OH) + H⁺ → HA(O^-) + H₂O

The contact time effect on the metal sorption capacity of the synthesized green sorbents was examined under the optimum buffering condition for each metal ion at different time intervals of 10, 20, 30, 40, 50 and 60 min. The results are illustrated in Fig. 6a. The data showed that nano-zeolite NaX exhibited the maximum sorption capacity of Cd²⁺ (8.13 mg/g), and Pb²⁺ (24.91 mg/g) upon reaching a contact time of 10 and 60 min, respectively. As for humic acid, the maximum adsorption capacities at 40 min were 12. mg/g and 23.11 mg/g for Cd²⁺ and Pb²⁺, respectively. The adsorption of Pb²⁺ on nano-zeolite proceeded in two stages (Fig. 6a). The first stage is characterized by a rapid adsorption behaviour due to the large available number of active sites on the sorbents and the possibility of the good orientation of the functional groups which enables reasonable interactions with the ions in solution. This stage is followed by a slower one due to sorbent surface saturation. Similar behaviour was discerned for the adsorption of Cd²⁺ on humic acid with a slight downward shift after 40 min probably due to repulsion interaction [37]. It is worthy to mention that the saturation of binding sites on nano-zeolite with Cd²⁺ was determined early, at 10 min of contact time. Achieving maximum efficiency of adsorption in a short contact time is an important factor in practical wastewater treatment systems which indicate less energy consumption in the shaking process and hence more economic feasibility [54].

The amount of adsorbent is an important factor, which determines the sorption capacity at a given initial concentration of adsorbate [50]. The influence of nano-zeolite NaX and humic acid amount on the removal of Pb²⁺and Cd²⁺ ions from aqueous solutions for initial metal concentrations of 20 ppm, pH of 7, contact time of 30 min and with different adsorbent doses from 10–100 mg for Pb²⁺ and 10–200 mg for Cd²⁺ is shown in Fig. 6b. As shown, the adsorption capacities of Pb²⁺and Cd²⁺ ions were increased by increasing the eco-sorbents amounts. The maximum adsorption capacities were achieved using 10 mg and 200 mg of nano-zeolite NaX dose for Pb²⁺ and Cd²⁺ ions, respectively. However, in the case of humic acid, the maximum adsorption capacities were determined at 40 mg and 160 mg for Pb²⁺ and Cd⁺² ions, respectively. This increase is due to the availability of more binding sites as the adsorbent dosage increased [55]. The observed decrease in Cd⁺² adsorption with increasing humic acid dosage may be attributed to the possibility of particle aggregations, which causes a decrease in the total surface area of the sorbents and an increase in the divisional path length. Also, a large adsorbent amount effectively reduces the saturation of the adsorption sites per unit mass, resulting in a reduction in capacity [37, 56]. The order of removal efficiency was Pb²⁺- nano-zeolite NaX (99.89%) > Pb²⁺- HA (97.12%) > Cd²⁺- HA (91.22%) > Cd²⁺- nano-zeolite NaX (44.80%). For both green sorbents, the removal of lead from aqueous solutions at any value of dosage was greater than that of cadmium. This may be attributed to the affinity values of each element toward the reactive sorbent material [10].

Variation of temperature greatly affects the adsorption of metal ions on sorbents. The thermal effect was studied over a temperature range of 10–60 °C under optimum experimental conditions. The data revealed that the adsorption capacities of Pb²⁺ on both sorbents and the adsorption of Cd²⁺ on humic acid decreased with the increase in temperature, indicating that the adsorption process in these cases was exothermic in nature. In contrast, the adsorption capacity increased with the elevated temperature in the case of Cd²⁺ on nano-zeolite NaX, indicating endothermic adsorption. Increased adsorption with temperature can also be ascribed to an increase in the number of active adsorption sites, which is caused by the thermal breakdown of internal bonds at the adsorbent's surface [57]. The effect of temperature is further discussed in the modelling section.

Ions that are co-existing in solutions with the determined heavy metals significantly interfere with their adsorption by competing on the active sites of the sorbents. The effect of selected ions, Na(I), K(I), Mg (II) and Ca (II) as main coexisting ions in seawater, on the metal sorption capacity of the synthesized sorbents, was studied under the current optimal experimental conditions of Cd²⁺ and Pb²⁺. As a representative example, the percentage removal of Cd²⁺ by the HA sorbent in the presence of interfering ions is displayed in Fig. S4. There is a clear decline in the adsorption capacity of nano-zeolite and HA in the presence of the selected interfering ions, as expected. This observation is pointing to the competitive adsorption mechanism on the binding sites which depends on the ionic size, solvation layers and effective nuclear charge of the ions in solution. Moreover, the type of the surface loaded, as well as the functional groups present on the green sorbents, may also affect this behaviour [58].

The influence of initial metal ion concentration on adsorption was explored in the concentration range of 5 to 100 ppm (Fig. 6c). The results showed a gradual increase in the adsorption capacity (q_e) of sorbents with the metal ion in the studied concentration range. It is noticed that at an equal concentration of Cd²⁺ and Pb²⁺ in solution, 100 ppm, nano-zeolite NaX exhibited identical adsorption capacity toward both metals. This could be explained by the effect of the nanoscale sorbents with high surface energy in quick attainment of the maximum adsorption capacity. Nevertheless, dissimilar metal adsorption behaviour was observed on the surface of HA. Obviously, adsorption is independent of the initial concentration at low metal ions concentrations where the ratio of metal ions to available adsorption sites is small. However, the case was different when the metal concentration raised up (Fig. 6c). At this point, many factors may govern the driving force of metal ions toward the active sites on the adsorbents.

The adsorption results were evaluated by a number of linear isotherm models including, Langmuir, Freundlich, Tempkin, Brunauer-Emmett-Teller (BET) and Dubinin Radushkevich. For all isotherms, the correlation coefficients (R²) were used to index the extent of agreement between the experimental data and the isotherm model. Best fitting is achieved when the correlation coefficient equals ~ 1.

The Langmuir adsorption quantifies the development of a monolayer adsorbate on the surface of the adsorbents with an assumption that no further adsorption occurs after this layer [58, 59]. The equation of the model with definitions of all parameters is stated in Table 2. Langmuir constants K_L (Eq. 2) was extracted from the plot of 1/q_e versus 1/C_e (Fig.7a), where q_e is the adsorption capacity at equilibrium (mg/g) and C_e is the concentration of ions at equilibrium (mg/L). According to the calculated q_e, the order of adsorption appears to be Pb²⁺ > Cd²⁺ for the two green-sorbents (Table 3). This trend refers to differences in binding energies and the stability of the resulted metal complexes [37, 64]. Langmuir constant, K_L, indicates the amount of the bonding energy. The multilayer adsorption isotherm for heterogeneous (rough and multisite) surfaces is expressed using the Freundlich model [65]. In Freundlich isotherm (Eq. 3; Table 2), the constants K_f and n_f denote the adsorption efficiency and the surface dissimilarity, respectively. K_f values (Table 3) pointed to a nearly similar removal feature for the studied ions on the green sorbents with much less efficiency for the removal of Cd²⁺ by nano-zeolite NaX under the same experimental conditions. Also, the n_f values for all the studied cases are greater than 1 which infers heterogeneity of the adsorbent surface and easy separation of ions from solution and hence capable adsorption process [24, 37].

Moreover, the data were fitted to the linear Tempkin’s isotherm which accounts for the adsorbent–adsorbate interactions (Eq. 4; Table 2). A_T in Tempkin’s equation is the equilibrium binding constant (L/min) that corresponds to the maximum binding energy [66]. The graphs of q_e versus ln C_e showed the viability of Tempkin’s isotherm (Fig. S5). The experimental data were also fitted to Brunauer–Emmett–Teller’s isotherm (BET, Eq. 5 in Table 2) which is a model used for the multilayer adsorption assuming that a layer does not need to be completed before the next layer starts to seal up [67]. The calculated R² values of adsorbents confirm the applicability of the BET isotherm. Dubinin–Radushkevich’s isotherm (D-R) can explain the physicochemical characteristics of the adsorption process. This model was applied to estimate the porosity characteristics of the sorbent and the apparent energy of adsorption. The liner form of this model is presented by Eq. 6 in Table 2. In addition, the apparent energy E_D in kJ/mol of adsorption from the Dubinin–Radushkevich isotherm can be computed using Eq. 7. The E_D is a parameter used to predict the type of adsorption. The physical adsorption is represented by E_D in the range (1–8) kJ/mol [68], whereas the chemical ion-exchange process is indicated by E_D > 8 kJ/mol [69]. The computed E_D values (Table 3) reflected chemical adsorption on the investigated adsorbents.

Kinetic models are used to examine adsorption efficiency, adsorption mechanism and rate control step that includes mass transfer of the process. Pseudo-first-order, second-order, intraparticle diffusion and film diffusion models were fitted to the data at different contact time (t) [70]. Lagergren proposed a pseudo-first-order kinetic model that was successfully applied to explain the kinetics of many adsorption systems. The integral form of the model (Eq. 8) is expressed in Table 4, where k₁ and q_e are the slope and intercept of the graph of ln(q_e-q_t) vs. (t). The kinetic parameters for pseudo-first and second-orders for nano-zeolite NaX and HA with Pb²⁺ and Cd²⁺ were calculated and presented in Table 5. The low correlation coefficient (R²) values for the pseudo-first-order model (Table 5) imply less linearity and that the model did not fit the whole range of contact time [73]. On the other hand, the pseudo-second-order kinetic model (Eq. 9; Table 4) exhibited superior linearity over the whole range of contact time with correlation coefficients values higher than 0.99 (Fig. 8 and Table 5). The comparison of the experimentally obtained q_e and the estimated q_e is the true assessment of the kinetic model’s validity [37]. It can be noticed that in the pseudo-second-order kinetic model, the calculated q_e are in excellent agreement with the experimental values. Moreover, a trend was observed in the rate constants of the pseudo-second-order model, such that k₂ generally decreased due to the occupation of active sites as the initial concentration increased. No trend was observed for the rate constant k₁ [74]. Usually, the pseudo-second-order model provides information about the overall reaction kinetics. However, it does not give evidence about the rate-limiting step. Metal ions adsorption kinetics from a solution into a solid always constructed on the basis of three consecutive steps: (1) film diffusion (i.e. mass is transferred from bulk liquid phase through boundary film to solid external surface), (2) intraparticle diffusion (i.e. diffusion in the liquid contained in the pores and/or along the pore walls) and (3) mass action (i.e. physical adsorption and desorption between the adsorbate and active sites). Since the adsorption step is very rapid, it is assumed that it does not influence the overall kinetics. The overall rate of adsorption process will be controlled by either surface diffusion or intraparticle diffusion. In many cases, intraparticle diffusion can be considered much slower step. Weber and Morris (1963) had cited some researches to establish intraparticle diffusion model, that assumes that the metal ions are transported from the solution through an interface between the solution and the adsorbent which called film diffusion, followed by a rate-limiting intraparticle diffusion step which bring them into the pores of the particles in the adsorbent. This model is correlated between qt as a function of the square root of time t (Eq. 10; Table 4). The slope and intercept values are represented by k_id and C. The constant C value gives an idea about the thickness of the boundary layer; and the larger the constant C value is, the greater is the boundary layer effect [75]. According to these data (Table S2), the constant C values are positive at the period of experiment, suggesting that surface diffusion has a large role in the rate-limiting step. Increasing the intercept values with increase metal ions concentration, due to agitation used in batch adsorption experiments, high enough to minimize external mass transfer limitations. Figure S6 pointed out the weakness of the linearity of the curve and all initial parts of different curves present an approximately linear behaviour, which influenced by the ratio between adsorbent mass and liquid volume. The weak linear plot did not pass through the origin, indicating that intraparticle diffusion kinetic model was not the only rate-limiting step in the adsorption mechanism. Adsorption kinetics may thus be controlled by liquid film diffusion and intraparticle diffusion simultaneously. The liquid film diffusion kinetic model can be expressed as shown in Eq. 11 (Table 4). For all adsorbents, the plot of –ln (1 – F) as a function of time (t) (Fig. S7) gave straight lines that did not pass through the origin. It is clearly known that several steps are involved in the sorption of sorbate by a sorbent. However, when the transport of the solute molecules from the liquid phase up to the solid phase takes place, the boundary plays the most significant role in the adsorption process.

As a result, it is concluded that the adsorption in the present contribution is a complex mechanism where both surface adsorption and intraparticle diffusion have a role in determining the rate.

The temperature dependence of the adsorption of metal ions on adsorbent was predicted by estimating the thermodynamic parameters such as free energy (ΔG°), enthalpy (ΔH°) and entropy (ΔS°) using the Van’t Hoff equations (Table 4) [76, 77]. The thermodynamic distribution coefficient K_d for the adsorption was calculated from Eq. 12, Table 4, at different temperatures of 283, 293, 303, 313, 323 and 333 K. ∆G° was then calculated from Eq. 13. The values of ∆H° and ∆S° were calculated from the slope and intercept of the plot of ln K_d against 1/T (Fig. 9) using Eq. 14 [78]. The degree of the spontaneity of the adsorption process is indicated by the free energy change (ΔG°), where a greater negative value implies more energetically favourable adsorption (Table S3).

The ΔG° values are negative at all temperatures indicating that the adsorption process is spontaneous and signifying physisorption rather than chemisorption [79]. In the case of Pb²⁺ adsorption on nano-zeolite NaX, ΔG° becomes more negative as the temperature increase which demonstrates the increase in the adsorption capability of Pb²⁺ at higher temperatures and the feasibility of the adsorption process [80]. On the other hand, the positive ΔG° values for the Cd²⁺ adsorption on nano-zeolite Nax indicates a non-spontaneous adsorption process that required a small amount of energy and hence is favourable at higher temperatures [56]. The positive values of ΔH° reveal the endothermic nature of adsorption of Cd²⁺ on nano-zeolite in the studied temperature range. One reason for the positive enthalpy is that metal ions dissolve well in water and must lose a portion of their hydration sheaths before being adsorbed, where the endothermicity of ion de-solvation requires more energy than the exothermicity of ion attachment to the sorbent surface [81]. Nevertheless, the negative ΔH° values of metal ion’s adsorption on humic acid point to exothermic adsorption. Also, the positive values of ΔS° (Table S3) confirm the randomness at the solid/liquid interface during the adsorption process on nano-zeolite [82].

Based on the present measurements, the affinity of adsorption of Pb²⁺ on eco-sorbents is greater than that of Cd²⁺ as reported by previous studies [83]. Several reasons control adsorption processes including the type and distribution of active groups on the adsorbent, the way metal ions interact with the adsorbent and the metal ionic radii. Pb²⁺, with the largest ionic size (the Pauling ionic radii is 1.20 Å), has the maximum adsorption capacity. However, smaller metal ions as in the case of Cd²⁺ (the Pauling ionic radii is 0.97 Å) interact less with the adsorbent because they have a higher hydration enthalpy and hence energy is required to separate H₂O molecules from the metal ions surface [84]. Also, the bigger the unhydrated ions, the lower the charge density, and because the charge is more distributed, the hydration water is less retained, resulting in a weaker metal ion-water phase interaction [85].

The comparison from the literature between the performance of nano-zeolite NaX and HA adsorbents with other clays adsorbents for cadmium and lead ions sorption is shown in Table 6.