Two–three parameters isotherm modeling, kinetics with statistical validity, desorption and thermodynamic studies of adsorption of Cu(II) ions onto zerovalent iron nanoparticles

Adsorption of problematic copper ions as one of the endocrine disruptive substances from aqueous solution onto nanoscale zerovalent iron (nZVI) was studied. The high pore size 186.9268 Å, pore diameter 240.753 Å, and BET surface area 20.8643 m2 g−1 and pH(pzc) enlisted nZVI as an efficient nano-adsorbent for treatment of heavy metals from synthetic wastewater. SEM and EDX revealed the morphology and elemental distribution before and after adsorption. 98.31% removal efficiency was achieved at optimum adsorption operational parameters. Of all the thirteen isotherm models, equilibrium data were well fitted to Langmuir. Kinetics and mechanism data across the concentrations from 10 to 200 mg L−1 were analyzed by ten models. PSO best described kinetics data as confirmed by various statistical error validity models. The intraparticle diffusion model described that the intraparticle diffusion was not the only rate-limiting step. The adsorption mechanism was diffusion governed established by Bangham and Boyd models. Feasible, spontaneous, endothermic, and degree of randomness were reveal by the thermodynamic studies. Better desorption index and efficiency were obtained using HCl suggesting multiple mechanism processes. The performance of ZVI suggested it has a great potential for effective removal of endocrine disruptive cationic contaminant from wastewater.

Surface charge (pH pzc ), BET surface area, surface morphology and elemental distribution. Following the salt addition and pH variation method, the point of zero charge was determined as presented in the supplementary materials 18 . Surface area by BET, pore width, and volume were determined using Micrometritics AutoChem II Chemisorption Analyzer. The surface morphological characterization and elemental analysis were carried out using a Scanning Electron Microscopy (SEM) integrated with Energy Dispersive X-ray (EDX) analyzer. SEM images and EDX spectra were obtained using a TESCAN Vega TS 5136LM typically at 20 kV at a working distance of 20 mm. Samples for SEM analysis were prepared by coating them in gold using a Balzers' Spluttering device.
Effect of stirring speed, pH, and co-existing ions. In order to optimize the stirring speed, 160-240 rpm speed was studied at optimum conditions. Effect of pH was studied by regulating the solution to the desired pH value using 0.1 M NaOH and 0.1 M HNO 3 solutions. Effect of Co-existing ions/Ionic strength varying the concentration of NaCl introduced into Cu 2+ solution from 0.001 to 1.0 M. Batch isotherm, kinetics, and thermodynamic studies. A typical batch adsorption study was carried out following procedure reported in our previous study 19,20 . 1000 ppm Cu 2+ stock solution was prepared by dissolving 2.5 g of CuSO 4 .5H 2 O in 1000 mL of distilled-deionized water. Study on initial Cu 2+ ion concentration was examined by adding 100 mg nZVI at different Cu (II) ions concentrations (10-200 ppm) and residual concentration determined by using AAS model AA320N. The quantity adsorbed and percentage removal efficiency were calculated utilizing Eqs. (2) and (3) 21-23 : The characteristics and mechanism of the adsorption process were investigated from the study of the Cedependent changes of Qe applied to thirteen isotherm models. Similarly, the batch adsorption kinetic experiments were conducted at optimum conditions for contact time ranging from 10 to 120 min. Adsorption capacities at contact time were obtained using Eq. (4) 5,23 : Kinetic data were fitted to ten kinetics and mechanism models. From the thermodynamics studies, the effect of temperature at optimum conditions was investigated at five different temperatures (298 K, 303 K, 318 K, 328 K, 333 K) for adsorption of endocrine disruptive Cu 2+ onto nZVI following our previously reported procedure 24,25 . The study was carried out in a temperature-controlled water bath. Data obtained were fitted to Van't Hoff equation depicted in Eqs. (5) and (6).
R is the gas constant (8.314 J mol −1 K) −1 ), T the absolute temperature (K), K c (q e /Ce) an equilibrium constant at various temperature. Standard enthalpy change ∆H° (kJ mol −1 ) and standard entropy change ∆S° (J mol −1 K − 1) were determined from the slope and intercept of the Van't Hoff plot of log K c versus 1/T 26,27 . Desorption studies. The desorption is a means of regenerating the adsorbent capacity for reusability and cost effectiveness determination. Desorption studies were investigated using the following eluents: deionized (DI) water, 0.2 M HCl and 0.2 M CH 3 COOH, at pre-determined optimum conditions. The desorption capacity, percentage desorbed, desorption efficiency and desorption index were determined using Eqs. (7)-(10) 28 : where q des is the quantity of metal ion desorbed (mg g −1 ), q e is the quantity of metal adsorbed after sorption (mg g −1 ), C des is the concentration of metal ion left after desorption (mg L −1 ), V is the volume of the metal ion solution (mL) while W is the weight of adsorbents (mg).

Results and discussion
nZVI physicochemical characterization: pH(pzc), BET surface area, pore volume and pore size. Summarized in Table 1 are pH, point of zero charge (PZC), BET surface area, and other physicochemical parameters describing the core-shell. The point of zero charge finds relevance in surface and nanoscience. Figure S1 (from the supplementary document associated with this study) shows the pH(pzc) of nZVI. It revealed that adsorption of Cu 2+ would take place at a pH > pH pzc as a result of more active binding being available due to deprotonation and low electrostatic repulsion. This finding shows that nZVI was positive at pH < pH pzc and negative at pH > pH pzc . Thus this signposts the suitability of nZVI for effective adsorption. The BET surface area 20.86 m 2 g −1 and the external surface area 16.4503 m 2 g −1 being greater than their corresponding micropores further support the suitability of nZVI for adsorption. Higher surface area enhances the adsorption process as supported. Therefore, it can be deduced that nZVI nano-adsorbent would utilize its external surfaces for heavy metal uptake than its micropore areas 29 . SEM/EDX characterization. Percolation of EDC-Cu 2+ into the pores ad matrix of nZVI was proved by the SEM/EDX depicted in Fig. 1A-D. Figure 1A revealed the SEM image before adsorption while Fig. 1B depicted the EDX with an intense peak of a zerovalent iron nanoparticle. Before adsorption spherical, chain-like aggregated morphology of nZVI was revealed SEM. The core-shell nature of zerovalent iron with intense peaks between 0.6-6.4 and 7.0 keV was revealed from the EDX result in Fig. 1B. Presented in Fig. 1C is the SEM micrograph showing swollen and robust nature of the surface of the nZVI nano-adsorbent after adsorption suggesting that the nZVI surface had been Cu-loaded up. More so, corroborating the result from SEM, Fig. 1D revealed the EDX spectrum showing the presence of Cu(II) as evidence of Cu adsorption onto core-shell nZVI. This is supported by finding in the literature 30 .  Figure S2 This showed that at optimum conditions, 85.04% RE and 81.04 mgg −1 quantity of Cu 2+ was adsorbed. The extent of removal of Cu(II) ion cation increased based on the availability of more active sites at lower concentrations until the pore sizes were saturated at an advanced concentration (150-200 mg L −1 ).  www.nature.com/scientificreports/ Concentration gradient was built up in the Cu-nZVI system due to intensification of drive force as concentration increased from 10 to 200 mg L −1 . This is supported by the findings in the literature 31 .
Effect of contact time. The Build-up of Cu(II) ions at the solid-liquid interfaces are controlled by the contact time. From this study, optimization of the contact time was investigated from 10 to 120 min. Fast kinetics from the bulk to the outer and inner surface of the nano-material (nZVI) identified by a short contact time to reach equilibrium was observed in the supplementary document as seen in Figure S3 Quantity of Cu(II) adsorbed increase from 4.96 to 82.82 mg g −1 as the initial Cu 2+ concentration increased from 10 to 200 ppm. A similar trend was observed by Baby et al. 32 on the adsorption of heavy metals.
Effect of initial solution pH. The key to the adsorption of heavy metal ions is the solution pH because it affects the surface chemistry of the system. A plot of the effect of initial concentration is presented in the supplementary document as seen in Figure S4 portrayed the effect of pH at optimum conditions. Coined from the understanding of the isoelectric point of the pH(pzc), nZVI is suitable for the uptake of cationic pollutants such as Cu 2+ since the pH > pH(pzc). At low solution pH, the system is protonated leading to electrostatic competition among Cu 2+ and other cationic species such as H + , Cu(OH) + , Cu(OH) 2 . However, at solution pH > pH(pzc), the system is negative, deprotonation occurs, there is less competition between Cu 2+ and other anionic species (Cu(OH) 3 − and Cu(OH) 4

2−
). Effective adsorption occurs at pH > pH(pzc). Optimum adsorption was achieved at pH 6 with 98.31% removal efficiency and quantity adsorbed 73.73 mg g −1 indicating effective binding of Cu 2+ onto nZVI surface. This is corroborated by the findings of other researchers 33 .
Effect of ionic strength. Analysis of Figure S5 (presented in the supplementary document associated with this study) showed the effect of ionic strength on Cu 2+ adsorption. Pollution of the water system is not limited to heavy metal ions only, some co-existing ions increase the salinity and ionic strength of the water body as investigated in this study. Co-existing ions polluted waste system increases the salinity and background electrolyte of the water body. A decrease in the percentage of Cu 2+ removed from 81.99 to 79.73% with a reduction in quantity adsorbed from 61.49 to 59.79 mg g −1 was observed in Figure S5. The decrease in Cu(II) ions uptake may also be due to a decrease in the electrostatic attraction arising from compressed electrical diffuse double layer supporting the findings of Advantageously, the removal efficiency of 81.99% shows that nZVI is an effective nano-sorbent in treatment industrial discharge containing co-existing ions. This is supported by findings in the literature 34 .
Effect of stirring speed. This study demonstrated as shown in Figure S6 (supplementary document) that at 200 rpm maximum adsorption of Cu 2+ onto nZVI was attained. Stirring speed is also one of the important parameters in adsorption studies because it promotes turbulence, frequency of collision and improves mass transfer in the medium between the two phases. At 200 rpm, the percentage Cu 2+ removal efficiency and quantities adsorbed are 96.98% and 72.73 mg g −1 for nZVI. Stirring speed increases the retention of Cu 2+ and it encourages a better transfer of Cu 2+ between solid-liquid interfaces (Cu 2+ -nZVI system) 18 . No appreciable percentage removal efficiency was observed after 200 rpm and all other study was carried out at this stirring speed. Two-three parameters adsorption isotherm modelings. One of the important and significant aspects of adsorption studies is mathematical isotherm modeling. Isotherm modeling is important in order to observe the relationship between nZVI and Endocrine disruptive Cu(II) ions at equilibrium conditions. A good understanding of this would significantly enhance the design of the adsorption system, effluent treatment reactor, and the pattern describing adsorbate-adsorbent interaction. Equilibrium data obtained from initial concentration were analyzed using thirteen mathematical isotherm models. All mathematical isotherm models used in this study were presented in Table 2 together with their non-linear, linear equations and parameters' description. The estimated parameters are portrayed in Table 3. The plots in the isotherm studies are presented in the supplementary document associated with this article from Figure S7A-S7M.
Langmuir isotherm model ( Figure S7A) assumes no interaction of the neighboring sites, monolayer surface, identical active sites, uniformity in adsorption energy 14 . The non-linear and linear Langmuir equations are presented in Eq. (11). In this study, the Langmuir isotherm model has the highest correlation coefficient (R 2 > 0.97) indicating the appropriateness and best fitting of equilibrium data to the Langmuir model. The Langmuir essential feature, as well as the separation factor or dimensionless constant (R L ), was calculated using Eq. (12) 20 . Values of calculated characteristics parameters are presented in Table 3. The values of R L (1 > R L > 0) portrayed in Table 3 supported favorable adsorption process 35 .
Freundlich isotherm model is presented in Eq. (13) ( Table 2) and the plot is as depicted in Figure S7B. The characteristic parameters are represented in Table 3. The values 12.54 and 1.83 indicated Freundlich capacity (K F ) and intensity (n F ) of adsorption respectively. The values of n F also measure whether the adsorption is favorable or not. The value of 1/n F (0.5457) less than unity and n F greater than unity and less than 10 indicated a normal and favorable adsorption 36,37 . Temkin model (Eq. 14) fits the experimental data (R 2 = 0.95) as depicted in Figure S7C. The positive value of B (14.678) and high b T (168.794 J mol −1 ) revealed the binding of Cu 2+ onto nZVI as well as the endothermic nature of the system. A report from other researchers corroborated this 36 .
Equations (15)-(17) defined the Dubinin-Kaganer-Raduskevich (DKR) model, Polanyi potential, and adsorption free energy of DRK. DRK plot is presented in Figure S7D and Table 3 shows evaluated parameters. The DKR free energy (E = 1581.14 J mol −1 ) lower than 8 kJ mol −1 supported that electrostatic interaction between Cu 2+ -nZVI system is Physisorption mechanism 38 . Halsey isotherm model (Eq. 18 and plot in Figure S7E) with parameters of K H and n H (0.0097 and -1.8325) further supported normal and favorable adsorption indicated by  Figure S7F) showed that the adsorption nature of the nZVI surface is not multilayer and heterogeneous 39 . Combination of both the Langmuir and Freundlich isotherm attribute could be assessed in Redlich-Peterson Isotherm model (Eq. 20, Figure S7G). Redlich-Peterson correlation value (R 2 = 0.9475) shows its versatility and fitting to equilibrium data 40,41 . Figure S7H depicts the Jovanovic isotherm (Eq. 21) model plot. Elovich isotherm model (Eq. 22, Figure S7I) has a foundation on the kinetic principle with the assumption of an increase in the adsorption sites exponentially 42 . It also takes into consideration the maximum monolayer capacity (Q max ). Based on R 2 value, the equilibrium data were fitted to the Elovich model but it was not as better described as compared to the Langmuir model. Also, its estimated Q max = 32.573 mg g −1 (Table 3) being less than that of Langmuir signposted Langmuir as a better model. Jossen's isotherm model (Eq. 23, Figure S7J) is based on a distribution of the energy of interactions between the system solid-liquid system 43 . Jossen's fit equilibrium data with R 2 > 0.97 (Table 3). Both Kiselev Isotherm Model (Eq. 24, Figure S7K) and the Flory-Huggins isotherm model (Eq. 25, Figure S7L) take into consideration the surface coverage (θ) of the Cu 2+ adsorbate on the nZVI. Jovanovic isotherm model corresponds to another approximation for monolayer localized adsorption without Table 2. Adsorption isotherm and kinetics models 35,43,47 .

Types of adsorption models Non-linear expression Linear expression Parameters nomenclature and description
Langmuir K L is the Langmuir isotherm constant (L mg −1 ) related to the binding energy of adsorption.Q max is the maximum monolayer coverage capacity (mg g −1 ) R L dimensionless separation factor indicating the nature and favourability of adsorption process. From slope and intercept of linear plot of Ce/Qe versus Ce, K L and Q max were determined C e equilibrium concentration of the MG dye adsorbate (mg L −1 ); Q e amount of MG dye adsorbed at equilibrium per unit weight of nZVI (mg g −1 ); K F Freundlich indicator of adsorption capacity 1/n F Intensity of the adsorption indicating the surface heterogeneity and favourability of the adsorption process. 1/n F and K F were determined from slope and intercept of linear plot of log Qe versus log Ce Temkin b T is the Temkin isotherm constant related to the heat of adsorption and A T is the Temkin isotherm equilibrium binding constant (L g −1 ) R = universal gas constant (8.314 J mol −1 K −1 ) T = absolute Temperature in Kelvin B = RT/b T = constant related to heat of sorption (J mol −1 ) obtained either from intercept or slope Q DKR is the theoretical adsorption isotherm saturation capacity (mg g −1 ) obtained from intercept. A DkR is the D-R isotherm constant (mol 2 kJ −2 ) related to free sorption energy obtained from the slope. Ɛ is Polanyi potential determined by the expression = RT ln(1 + 1/C e ). E is the mean adsorption free energy helpful in determining the adsorption nature (physisorption or chemisorption of the adsorption process  Kiselev K i is Kiselev equilibrium constant (Lmg −1 ) and K n is equilibrium Constant of the formation of complex between adsorbed molecules. Kiselev constants were determined from the plot of 1/C e (1 − Ө) versus 1/Ө Flory-Huggins (26) θ is degree of surface coverage, n H is number of adsorbates occupying adsorption sites, and K FH is Flory-Huggins equilibrium constant (L mol −1 ). n FH and K FH were determined from the linear plot of Log(θ/C o ) versus log (1 − θ) K FG is the Fowler-Guggenheim (F-G) equilibrium constant (L mg −1 ), θ the fractional coverage, R the universal gas constant (kJ mol −1 K −1 ), T the temperature (K), and W is the interaction energy between adsorbed molecules (kJ mol −1  www.nature.com/scientificreports/ lateral interactions which ought to be similar to the Langmuir isotherm model. However, a lower correlation coefficient (R 2 = 0.6105) obtained in this study indicated that there is a lateral interaction and thus this model lower approach towards saturated compared to Langmuir adsorption isotherm as reported by Al-Ghouti et al. 44 . This is supported by the fit and parameters obtained from Fowler-Guggenheim (F-G) isotherm model (Eq. 27, Figure S7M). Taken into consideration is the lateral interaction of adsorption of EDC-Cu 2+ onto nZVI by the FG-isotherm model 45 . As reported by the literature, the interaction between the adsorbed molecule is attractive, if W is positive; repulsive interaction if W is negative and no interaction between the adsorbed molecules will be observed if W = 0 42,46 . In this study, the fit of the Fowler-Guggenheim isotherm model (R 2 = 0.9487) and positive value of W (W = 881.02 J mol −1 ) indicated that there is positive contact in the Cu 2+ -nZVI system, hence the adsorption heat increased with loading confirming endothermic adsorption process as observed in the thermodynamics studies. The screening and arrangement are based on the understanding of the important parameters (Q max and R 2 ). With regards to Q max (in descending order) Langmuir > DKR > Elovich > Jovanovic. With respect to R 2 (in descending order): Langmuir > Jossen = Elovich > Freundlich = Halsey > Temkin > Fowler-Guggenheim > DKR > Flory-Huggins Kiselev > Harkin-Jura > Jovanovic. Presented in Table 4 is the comparison of maximum monolayer adsorption capacities of adsorption of Cu 2+ onto various nano-adsorbents and nZVI used in this study. It is obvious that nZVI exceedingly surpassed other existing adsorbents reported. This indicated that nZVI is an excellent potential nano-adsorbent for effective removal of endocrine disruptive heavy metal ions.
Adsorption kinetic with statistical error validity modeling. A kinetic study was undertaken to understand the controlling pathway, the rate of surface adsorption of the contaminant to the adsorbent, and the Table 3. Isotherm models and their various evaluated parameters for the adsorption of Cu 2+ onto nZVI.  www.nature.com/scientificreports/ quantity of the adsorption capacity. The kinetics equations vis-à-vis pseudo-first-order (PFO), pseudo-secondorder (PSO), Elovich, Avrami, and Power Function (Fractional power) are represented on Eqs. (28)-(34) 47 .
The kinetic plots are presented in Fig. 2A-E with error bars indicating the application of error models and the evaluated parameters are presented in Table 5. The kinetic constant k 1 of Pseudo first order (PFO), its adsorption rate constant h 1 , disagreement between q e , exp and q e , cal and low correlation coefficient, R 2 < 0.90, demonstrated that PFO is not applicable in this study. A similar low trend in the R 2 value was observed in the Avrami model demonstrating that it is not applicable in this study. A good agreement between the experimental quantity adsorbed (q e , exp ) and the calculated quantity adsorbed (q e , cal ) was observed in PSO, Elovich, and Power function. From the Elovich model, the values of ∝ (adsorption rate) increased with an increase in concentration as a result of an increase in the number of sites. The values of 1/β at 10 ppm, 50 ppm, 100 ppm, and 150 ppm are 5.882, 11.764, and 17.123 respectively. These values reflect the number of sites available for adsorption 30 . Kinetic parameters from Power Function in Table 5 indicated time-dependent of Cu(II) onto nZVI with the value of constant v less than 1 across all the concentrations. Of all these kinetic models, PSO best described the Cu(II) adsorption process and this was supported by the statistical error validity model presented in Table 5. The PSO initial adsorption rate (h 2 ) increases with increase in concentration from 33.67 to 238.095 mg g −1 min −1 . R 2 values range from 0.99 to unity demonstrating the best fitting by PSO suggesting chemisorption mechanism.
Statistical validity of the kinetic models. Assessment on the best kinetic fitting model that is always based on linear regression coefficient could be biased inherent, hence the need for statistical validity model. The suitability, agreement, and best fit among the kinetic models are judged not only by regression coefficient (R 2 ) but also with the use of statistical error validity models. Validity of kinetic data was fitted to statistical error models namely; Average relative error (ARE), Normalized Standard Deviation Δq t (%), Hybrid fractional error function (HYBRD), Derivative of Marquardt's percent standard deviation (MPSD), Standard deviation of relative Error (S RE ). The various statistical functions are presented in Table 6. Presented in Table 7 are the statistical error validity data of the kinetic models. Five statistical tools were used for the validity of these kinetic models. It is observed that the closer the agreement between the experimental quantity adsorbed (qe, exp) and calculated quantity adsorbed (qe, cal), the lower the values of these statistical tools, the better the model. In order to justify and juxtapose the best model, a reference was made to the coefficient of regression (R 2 ). The higher the R 2 values, the closer the values of qe, exp, and qe, cal, the lower the values of ∆q, HYBRID, MPSD, ARE, and S RE, the better the kinetic models in describing the sorption process [48][49][50] . The values in Table 4 vividly show that pseudo-second-order at various initial Cu 2+ concentrations (10 ppm, 50 ppm, 100 ppm, 150 ppm, and 200 ppm) best describe the sorption process. the model can be arranged in descending order with respect to R 2 : pseudosecond-order > Elovich > fractional power > Avrami > pseudo-first-order. Figure 3A-E show the linear plots of intraparticle diffusion, liquid diffusion, external diffusion, Bangham and Boyd models. Adequate understanding of the adsorption mechanism is enhanced by the determination of the ratecontrolling/determining step. The three definite steps that could be used to describe the adsorption rate are 51 : (1) Intraparticle or pore diffusion, where adsorbate molecules percolate into the interior of adsorbent particles, (2) Liquid film or surface diffusion where the adsorbate is transported from the bulk solution to the external surface of the adsorbent, and (3) adsorption on the interior sites of the sorbent. Since the plot of Intraparticle diffusion (Fig. 3A) did not pass through the origin, it is demonstrated that it is not the only rate-determining step 52 . Other mechanisms such as surface diffusion and external diffusion also participated in the mechanism of Cu(II) removal. However, the higher R 2 values of intraparticle diffusion from the evaluated parameters presented in Table 8 demonstrated that the mechanism is pore diffusion dependent which was confirmed by Bangham and scattered plot of Boyd models 53,54 . The intercept of intraparticle diffusion which is the thickness of the (28) Pseudo first-order Lagergren's rate equation ℓog(q e − q t ) = ℓog q e − k 1 t 2.303 (29) h 1 initial pseudo first-order adsorption rate mg g −1 min −1 h 1 = k 1 q e (30) Pseudo second-order rate equation:

Adsorption mechanisms for sorption of Cu 2+ onto nanoscaled zerovalent iron (nZVI).
1 q e t (31) h 2 is the initial pseudo second-order adsorption rate: h 2 = k 2 q 2 e (32) Elovich model: Thermodynamics analysis. Thermodynamics analysis is imperative to determine the (enthalpy change) heat content (ΔH); entropy change (degree of randomness, ΔS), possibility, and spontaneity (Gibbs free energy change, ΔG) in every adsorption process. The plots in thermodynamics studies are presented in the supplementary document associated with this article. As observed in Figure S8, intensification in the percentage removal efficiency was attained with an increase in temperature of the system supporting the endothermic process. This is due to a decrease in the mass transfer resistance and boundary layer thickness of nZVI 55 . Van't Hoff 's linear plot of log Kc against 1/T was portrayed in Figure S9 and the result obtained was presented in Table 9. The positive value of ΔH (+ 50.6059 kJ mol −1 ) confirmed that the adsorption process is endothermic in nature 56  Desorption mechanism. Figure 4 shows the comparative effect of different eluents in the desorption of Cu 2+ from Cu 2+ -loaded nZVI. The opportunity to investigate regeneration and reusability of loaded adsorbent is enhanced by desorption studies. The effectiveness of three different eluents and desorbing agents (HCl, CH 3 COOH, and H 2 O) was investigated. The basic desorption mechanisms are ion exchange, complexation, and  50 Acetic acid also performed averagely while distilled-deionized water was a poor desorbing agent in the desorption of Cu(II) from Cu(II)-loaded-nZVI. Thus, the adsorption of Cu(II) onto nZVI  Table 6. Adsorption statistical error validity models (ASEVM) 24,43,44,46 .

Conclusion
This study revealed the effectiveness of nZVI as an auspicious nano sorbent for the efficient elimination of endocrine disruptive heavy metal ions. The quality physicochemical properties of nZVI gave it an edge among the list of other nano-adsorbents compared. Evidence of the adsorption of Cu 2+ onto nZVI was revealed by a change in morphology and elemental distribution by SEM and EDX respectively from post adsorption characterization.   www.nature.com/scientificreports/ The adsorption of Cu 2+ onto nZVI was well influenced by operational parameters. Optimum adsorption was achieved at pH 6 with 98.31% removal efficiency, 73.73 mg g −1 quantity adsorbed and 200 rpm stirring speed. Thermodynamics parameters ΔH° (+ 50.6059 kJ mol −1 ), ΔS° (174.6790 J mol −1 K −1 ), ΔG° (− 1.6765 kJ mol −1 to − 7.9602 kJ mol −1 ). Indicated random, feasible, spontaneous, and endothermic nature of the adsorption process. The adsorption behavior was well explained by the Langmuir isotherm model and it followed the following order: Langmuir > Jossen/Elovich > Freundlich/Halsey > Temkin > Fowler-Guggenheim > Redlich-Peterson > DKR > Flory-Huggins > Kiselev > Harkin-Jura > Jovanovic. Langmuir best described equilibrium data. The Langmuir monolayer adsorption capacity (90.09 mg g −1 ) surpassed other nano-adsorbents utilized for the adsorption of Cu(II) ion. The Pseudo-second-order (PSO) best described the kinetics model based on R 2 values greater than 0.99, close agreement between qe, exp and qe, cal and lower values of the five rigorous statistical validity models (Δq t, ARE, HYBRD, MPSD, and S RE ). The mechanism model was pore diffusion dependent. Best desorption capacity and the index was portrayed by HCl indicating that ion-exchange, electrostatic, and physisorption mechanism. Based on the capacity displayed by nZVI in adsorption of EDC Cu 2+ , it could be recommended for effective industrial treatment of heavy metal ions. Table 9. Thermodynamic parameters for adsorption of Cu 2+ onto nZVI.  Figure 4. Comparative effect of different eluents in the desorption of Cu 2+ from Cu 2+ -loaded nZVI.