Heavy metal pollution is produced throughout manufacturing and farming doings, and furthermore is realized in vehicular productions. Mercury ion is one of the most severe contaminants of heavy metals in water and sediment, upcoming from chlor-alkali business, medicinal, paper, oil factory, dye, and battery activities. Heavy metals are nonbiodegradable, consequently removal of heavy metal ions from water and wastewater is very important [1-3]. Up to now, the progress of analytical systems for the specification the amount of mercury has been a difficult task. A severe trouble faced in the specification the amount of mercury is that goal types generally exist in small concentrations. A lethal concentration of mercury salts covers from less than 0.1 ng/ml to further than 200.0 ng/ml for aquatic types and river creatures. Diverse types of ways are obtainable for eliminating mercury in water and wastewater containing electrodialysis, conventional coagulation, chemical precipitation, reverses osmosis, line softening, photocatalytic reduction, ion-exchange, adsorption [2-5]. Amongst these procedures, adsorption is an extremely operational, inexpensive, and commonly practical way. Definite adsorption of mercury ion by diverse kinds of low price farming waste ingredients such as tree bark, peanut skin, wool, onion skin, sawdust and coconut husk and manufacturing waste ingredients for example rubber, waste tyre, enricher waste slurry, coal, and photofilm waste slurry has been informed . Most recently, many classes of clay minerals have been used as adsorbents to remove pollutants from wastewater . Clay minerals have drawn greatly consideration because of their characteristic structures; for example its great specific surface area, swelling, ion exchange properties, high sorption capacity, intercalation behavior and their smaller price paralleled to synthesized ingredients (e.g. silica gel, zeolite) . This paper is a report on a research trying to extract and preconcentrate mercury ions from aqueous solutions using batch equilibrium technique by Ghezeljeh montmorillonite nanoclay as natural and native adsorbent. It is worth nothing that Ghezeljeh montmorillonite nanoclay (Geleh-Sar-Shor) was used in ancient Persia to clean the body, hair, and also to bathe dead bodies prior to the funeral (“Geleh-Sar-Shoor” means head-washing clay). The clay is still used in some parts of Iran. This clay is readily obtainable, low-priced and environmentally friendly with chemical and mechanical stability. Ghezeljeh clay is distinguished by using Fourier transform infrared spectroscopy (FT-IR), scanning electron microscopy-energy dispersive spectrometer operating (SEM-EDS), X-ray diffractometry (XRD), X-ray fluorescence (XRF), cation exchange capacity (CEC) measurements, specific surface area and zeta potential. The results of XRD, FT-IR, zeta potential and CEC of Ghezeljeh clay confirm that montmorillonite is the dominant mineral phase. On the basis of SEM images of clay, the distance between the layers is in nm scale. The current study overlaid the use of Dubinin–Radushkevich, Temkin, Langmuir and Freundlich isotherm models to define the adsorption progression onto nanoclay. Kinetic models such as first-order, second-order equations and intra-particle diffusion model and thermodynamics were moreover studied. So far, Ghezeljeh montmorillonite nanoclay (Geleh-Sar-Shoor) has not been studied as an adsorbent for the extraction-preconcentration of heavy metals. Only Soleimani and Hassanzadeh in the Imam Khomeini International University (IKIU) have used Geleh-Sar-Shoor for the extraction of metal ions [7-9].
Clays are hydrous aluminum silicates which are ordered as 1:1 and 2:1 clay minerals. The sheets in these clays are kept up by weak van de Waals forces making it easy for other compounds to take up the interlayer region. Montmorillonite is dioctahedral clay of the smectite type including alumino-silicate layers. The silica tetrahedra (T) (Si4+ in tetrahedral coordination with O2−) and alumina octahedral (O) (Al3+ in octahedral coordination with O2−) are interrelated (through the sharing of O2− at polyhedral corners and edges) in such a fashion that a sheet of alumina octahedral is inserted between two sheets of silica tetrahedral. Consequently, the configuration is T-O-T (2:1) . Most of the surface charges on montmorillonite are created by isomorphous replacement or non-ideal octahedral possession. Adsorption of metal cations in clay minerals is restricted by factors such as charge features of the clay. Charge features contain the amount of the active places and cation exchange aptitude. The exchange progressions depend on several elements, similar the physicochemical features of solid and cation (such as ionic radius, charge extent, hard–soft acid–base characteristics, hydration volume and hydration enthalpy of cation), temperature, ionic power, concentration of ions, existence of challenging ions, and experimental situations involving time of reaction, and pH of the medium [11-18]. The information of the heavy metal binding to the mineral building is of main importance to calculate the motion and long-term behavior of heavy metals in natural structures .
Metal sorption mechanisms onto montmorillonite
Although numerous examinations on metal adsorption on montmorillonite have been successfully exhibited, there is no entire agreement on adsorption locations, surface responses and exhibiting methods. It is consequently challenging to differentiate the adsorption affinity of numerous metal cations by montmorillonite. Through the dissimilar nature of the clays, it is probable that numerous metal sorption mechanisms happen at the same time creating the determination of metal/clay factor interaction problematic on such substance [1,6,19]. Brigatti et al.  have reported that mercury is specially adsorbed as Hg–OH complexes in montmorillonite. Hg–OH2 complexes are fewer powerfully linked to the 2:1 sheets, as verified by EXAFS and thermal study. On the opposing, Hg–O intercalates are further toughly bounded to the layers and mercury is released at upper temperatures. Sajidu et al.  have studied the interactions of copper(II), cadmium(II), zinc(II), mercury(II), chromium(III), and lead(II) on natural basic variegated clays using EXAFS spectroscopy. Mercury(II) is adsorbed as hydrolyzed lined O–Hg–O elements on the clay surface at neutral pH, whereas it is reduced to mercury(I) at low pH and adsorbed as ≡O–Hg–Hg–OHx complexes (x = 1 or 2). The adsorption of metal ions chiefly relies on its hydrated radius [20-22]. The metal ions having lesser hydrated ionic radius have easier contact to the mineral’s surface and can diffuse more simply inside its openings. On the other hand, the existence of ions having a higher hydrated ionic radius outcome an extra quick overburden of the adsorption locations .
MATERIALS AND METHODS
Reagents and solutions
All the chemicals were purchased from the German company of Merck: acids, bases, hydrogen peroxide, sodium acetate, sodium carbonate, sodium citrate, nitrate salts of copper, lead, chromium, cobalt, cadmium, mercury, sulfate salts of aluminum, nickel, manganese, zinc, magnesium, chloride salts of sodium, potassium, iron, calcium, and ammonium. Since the chemicals were at the maximum purity, they were applied without any purification. The element standard solutions were produced by diluting a stock solution of 1000 mg/l of the specified component using doubly distillated water. The Hg(II) stock aqueous solution (1000 mg/l) was obtained from the dissolution of Mercury(II) nitrate monohydrate (Merck, Darmstadt, Germany), acidified with nitric acid to avoid Hg precipitation. The employed solutions of Hg(II) ions were readied day-to-day via suitable dilution of mercury stock solution which was prepared weekly. A citrate-citric acid buffer solution was readied using 0.1 M citric acid solution at pH 2-3. Acetate buffer solution was readied via combining suitable volumes of 0.1 M acetic acid and 0.1 M sodium acetate at pH 4-6. Phosphate buffer solution was readied by 0.1 M phosphoric acid at pH 7. Carbonate buffer solution was readied using 0.1 M sodium carbonate at pH 7. Citrate buffer solution was readied using 0.1 M sodium citrate at pH 7. Ammonium buffer solution was readied by mixing suitable quantities of 0.1 M ammonia and 0.1 M ammonium chloride at pH 8-10. The pH of the buffer solutions was balanced by adding 1 M NaOH or HCl, as needed. The Ghezeljeh montmorillonite clay (adsorbent) was collected from Ghezeljeh, a village 18 km west of the city of Tafresh (Markazi Province) in Iran. Different water samples used in the experiments were collected from Caspian Sea, Karun River (inside and outside the city of Ahvaz), Persian Gulf, Iran, Well water, Haryrood river, and Tap water (from Herat city), Afghanistan. Five different fresh fish samples used in the experiments were obtained from the local markets in Qazvin, Iran.
A model 420A digital Orion pH meter (Gemini, the Netherlands) prepared by a combined glass electrode was applied for pH corrections. An ultrasonic water bath (Bandelin, Berlin, Germany) was used to disperse and disaggregate this clay. Batch experiments were carried out in Incubator Shaker (model 3020 DRS, FSA, Iran) at 200 rpm in Imam Khomeini International University (IKIU). X-Ray diffraction (XRD) data were attained by an Ital Structures diffractometer (GNR, Novara, Italy), with Cu Kα radiation (40 kV/30 mA, λ= 1.542 Aº). Fourier transform infrared (FT-IR) examine was applied by Tensor Bruker MIR-T27 (Germany) taking a standard mid-IR DTGS detector in IKIU. To quantitatively measure the mercury ions in the standard solutions, a GBC 902 flame atomic absorption spectrometry (FAAS), (Dandenong, Victoria, Australia 3175) with deuterium background corrector and an air-acetylene flame was used in IKIU. The working situations in the FAAS spectrometer were adjusted according to the standard procedures of the company. However, the analysis of natural water and fish samples were achieved with a Varian 220Z (Australia) atomic absorption spectrometer (AAS) with vapor generation accessory (VGA) system was used in Iran mineral processing research center (IMPRC). Philips X-ray fluorescence (XRF) of the sample was carried out using XRF analysis instruments (Philips Magix Pro, Netherlands) in IMPRC. A scanning electron microscope (SEM) (LEO 1450 VP, Thornwood, N.Y., USA) by variable pressure secondary electron detector and energy dispersive spectrometer operating (EDS) at 30 kV (Oxford INCA software, High Wycombe, U.K.) were applied for SEM-EDX analysis in IMPRC. Zeta potential measurements were carried out on a Zetameter ZetaCAD (CAD Instruments, France) in Islamic Republic of Iran Ministry of Agriculture-Jahad. The specific surface areas were calculated by the BET method by means of a Belsorp mini II instrument (BelJapan, Japan) in University of Tehran.
Preparation of the adsorbent
The adsorbent was readied using the Galehouse method [7-9]. Natural Ghezeljeh clay was first treated with 0.1 M of acetic acid to eliminate carbonates, and then 30% H2O2 was used to exclude mineral and organic impurities. The clay was carefully rinsed with doubly distilled water to eliminate traces of acetic acid and hydrogen peroxide. The treated clay was dispersed and disaggregated in doubly distilled water through an ultrasonic water bath. The resulting suspension was transferred to a measuring cylinder and permitted to stand for 3 h, 26 min, 6 sec for sedimentation. The fine fraction (< 2 µm) was removed and placed in an electric vacuum oven at 50°C for 72 h to be dried. Then, it was placed in a desiccator for subsequent experiments.
Studies on adsorption process were performed through batch method at room temperature. At first, a 50 ml solution containing mercury was moved into an Erlenmeyer flask. Then, 10 ml of an appropriate buffer solution was added and followed by 0.5 min of agitation. Next, 2 g of the Ghezeljeh clay was added. The mixture was agitated for 10 min using a mechanical shaker. The liquid phase was separated from the solid part via centrifugation at 3500 rpm for 30 min. The supernatant was decanted.
To elute the analytes adsorbed onto the Ghezeljeh clay, 10 ml of 3M HCl was added to the solid part, then it was stirred for less than 0.5 min. The suspension was allowed to stand for 20 min. Then, it was centrifuged at 3500 rpm for 30 min. The supernatant (10 ml) was collected to measure its mercury ion concentration. To optimize the experimental conditions, these steps were repeated three times. The equivalent method was used to the blank solution.
Physicochemical characterization of Ghezeljeh nanoclay
Scanning electron microscopy (SEM) is a powerful technique applied in micro imaging of a variety of surfaces. Clay samples were covered with Au under vacuum in an argon atmosphere. Solid morphology; particle size and texture of the clay surface were determined by scanning electron microscopy (SEM) studies (Fig. 1a). On the base of SEM images of clay, the distance between the layers is in nm level, as shown in Fig. 1.
X-ray diffractograms were obtained for the 2θ angles ranging from 2º to 40º 2θ at room temperature. The clay was treated with ethylene glycol, an organic compound which steadily intercalates itself into the lattice of the clay. The structural possessions of the clay were monitored beforehand and afterward treatment with ethylene glycol. The X-Ray diffraction analysis revealed that the clay sample was chiefly made up montmorillonite minerals (Fig. 1b) [10,24,25]. The strong 13.69, 5.26 and 3.27 Å peaks and relatively weak 9.50 Å peak in ethylene glycol-solvated case are related to mixed-layered illite/smectite. Using 9.50 (001/002 Ill/S) and 5.26 Å (002/003 Ill/S) peaks, the estimated percent of illite in the mixed-layered. The 7.30 and 3.61 Å peaks indicate that kaolinite is also present .
XRF and EDS study
The chemical composition of this clay was determined with XRF and EDS (Fig. 1c). Results of XRF analysis: SiO2: 54.47; Al2O3: 20.92; MgO: 3.65; SO3: 0.32; K2O: 1.82; CaO: 1.14; Na2O: 0.16; TiO2: 0.37; Fe2O3: 3.13; PbO: 0.16; SrO: 0.10; ZrO2: 0.05; As2O3: 0.02; Loss-on-ignition corrections (L.O.I.): 13.7.
To prepare the clay sample for FT-IR spectroscopy, an electric vacuum oven was used to dry (at 50°C for 6 h) and cool the clay. A FT-IR spectrum was indicated in the range of 400-4000 cm-1 with the KBr pellet method. Mixed-layered illite/smectite is distinguished according to the absorption bands near 3627 and 1029 cm-1, bands near 914 and 836 cm-1. Sharp bands at 3694 and 3927 cm-1 belong to kaolinite . The FT-IR analysis, also confirmed that the Ghezeljeh montmorillonite clay is mainly composed of montmorillonite minerals (Fig. 1d) .
Cation exchange capacity (CEC)
The cation exchange capacity (CEC) is the amount of equals of interchangeable charge per quantity of clay, which is equal with the layer charge . The CEC of the Ghezeljeh montmorillonite nanoclay was calculated with 0.01 M Cu-triethylentetramine [28,29]. The CEC value of 160.0 meq/100 g for Ghezeljeh montmorillonite nanoclay was found. The significant value of CEC is in a good agreement with what has been reported for Montmorillonite in the literature .
The specific surface area (SBET), pore radius and pore volume of the Ghezeljeh montmorillonite nanoclay were obtained from N2 adsorption isotherms attained at liquid nitrogen temperature (at 77 K) by means of a Belsorp mini II instrument (BelJapan, Japan). Prior to the surface area calculations, humidity and vapor on the solid surface or entered in the open holes were cleaned off by heating under vacuum at 100⁰C for 12 h. The Ghezeljeh montmorillonite nanoclay possesses a specific surface area of 90.916 m2/g, pore radius of 4.8 nm and pore volume of 0.147 cm3/g [19,31].
Zeta potential measurement
The zeta potential of the Ghezeljeh nanoclay was obtained from electrophoretic mobility measurements at 21.31⁰C, with Zetameter ZetaCAD instruments). The zeta potential obtained at a natural pH of 5.64 is -25.970 mV equivalents to zeta potentials of montmorillonite (−21.2 mV) .
RESULTS AND DISCUSSION
Prior to extraction-preconcentration step of mercury ions from real samples, standard solutions were subjected to optimize a number of operating parameters involved in the extraction-preconcentration of mercury ions. The parameters were quantity of adsorbent, eluent characteristics (type, volume, and concentration), pH of the buffer solutions, buffer type, shaking time, volume of the standard solutions, and initial mercury ions concentration (adsorption capacity). The role of desorption time and centrifugation time was also studied. A summary of the main findings is as follows.
Effect of adsorbent amount
Ten quantity levels of the Ghezeljeh montmorillonite clay were studied: 0.1, 0.2, 0.3, 0.4, 0.5, 1, 1.5, 2, 2.5 and 3 g. The standard solution was 60 ml composing of 50 ml of doubly distilled water containing 14.63 mg/lof mercury ions and 10 ml of buffer solution added at pH 7. The adsorption of the Hg(II) ions onto the clay improved as the amount of the Ghezeljeh nanoclay was increased. However, the adsorption declined at adsorbent amounts higher than 2 g. Reduction in the adsorption could be explained by the point that when the adsorbent amount is less than 2 g, the mercury ions can simply come into contact with the adsorption positions, while when the adsorbent content exceeds 2 g, the amount of such positions per unit quantity decreases, due to accumulation and flocculation of adsorbent fragments [33-35].
Effect of eluent kind, volume and concentration
To obtain suitable eluent, HCl and HNO3 solutions were used at various concentrations (1-5 M) with varying volumes (5-15 ml) for the elution of mercury ions adsorbed onto the Ghezeljeh nanoclay. The adsorbed ions were readily eluted (desorbed) from the nanoclay only when 10 ml of 3 M HCl was used.
Effect of pH of buffer solutions
To research the effect of pH of the buffer solutions in adsorption of mercury ions onto the Ghezeljeh nanoclay, pH was adjusted in the cover of 2 to 8 at room temperature, using buffer solutions given in section reagents and solutions. At pH higher than 8, Hg(OH)2 solid phase was formed, therefore the Mercury maintenance ability reduces. Mercury ions were optimally adsorbed on the Ghezeljeh nanoclay at pH 7. For subsequent runs of the experiment, pH 7 was applied as the optimum pH level for phosphate buffer solution. The results are shown in Fig. 2a. Clays are identified to have a negative surface charge in solution, as pH changes, surface charge also changes, and the adsorption of charged species is affected. At low pH values, there are extra H3O+ ions in solution, a competitiveness occurs between the positively charged hydrogen ions and metal ions for the accessible adsorption positions on the negatively charged clay surface.
Effect of the type of buffer solutions
Three types of buffer solutions were compared at a concentration of 0.1 M at pH 7 in terms of their effect on mercury ions adsorption: phosphate, carbonate and citrate buffer solutions. The phosphate buffer solution led to a higher level of mercury ions adsorption. The percentage of the recovery of mercury ions was approximately consistent with 47.1%, using carbonate and citrate buffer solutions.
Effect of the concentration and pH of phosphate buffer solution
To understand the effect of concentration and pH of phosphate buffer solution on adsorption mercury ions on the nanoclay, concentrations of phosphoric acid in the range of 0.001 M to 3 M and pH from 5 to 8 at room temperature are changed. The maximum percentage of recovery is obtained at 0.1 M and pH 7.
Effect of agitation time
To study the effect of agitation (shaking) time (or contact time), the adsorption of mercury ions onto the Ghezeljeh nanoclay was measured after 10, 20, 30, 40, 50 min of shaking the standard solutions (Fig. 2b). It was completed after 20 min adsorption. Consequently Hg(II)–clay interactions arrived at balance state in less than 20 min. It showed that the adsorption locations on the clay minerals were quickly covered using the mercury ion. On the basis of the consequences, a 20 min of agitation time was attained appropriate for the extreme adsorption. This optimum value was used in the rest of experiments.
Effect of volume of standard solutions
To study the influence of the total volume of the standard solution (sample + buffer), on the adsorption of mercury ions onto the Ghezeljeh nanoclay, seven quantities of 30, 60, 90, 120, 300, 420, and 600 ml were investigated. This was aimed at attaining a high preconcentration factor. It was found that recovery is over 95% at quantities of 30–90 ml, but it declined to below 95% when the volume of the solution exceeded 90 ml. Now, given that ending solution volume to be determined by FAAS is 10 ml, the preconcentration factor is 9. The results are recorded in Fig. 2c.
Effect of primary mercury ion concentration
The adsorption capacity of an adsorbent is defined as the largest amount of metal adsorbed onto 1 g of the adsorbent . In order to evaluate the adsorption capacity of the Ghezeljeh nanoclay, 2 g of the clay was added to different standard solutions containing 0.59, 0.88, 1.17, 2.34, 3.51, 4.68, and 5.85 mg of mercury ions (Fig. 2d). The adsorption capacity was evaluated to be 1.17 mg/g (relative error smaller than ±5%). At lesser concentrations, a great number of adsorption positions on the nanoclay are accessible to the metal ions and this condition is improved with elevation of metal ion concentration and the competition for adsorption locations becomes difficult.
Effect of desorption time
Desorption time is defined as the length of time that an eluent is in contact with the adsorbent containing metal ions. The desorption time in this study was studied by measuring the recovery of mercury ions from the clay after 5, 10, 15, 20, 25, and 30 min of contact between HCl and nanoclay (Fig. 2e). Desorption time of 25 min was found to lead to the highest degree of desorption. This value was applied in the remaining experiments.
Effect of centrifugation time
To explore the effect of centrifugation time on the desorption of mercury ions from the clay, aliquots taken from the standard solutions after 25 min of desorption time were centrifuged for 10, 20, 30, 35, and 40 min at a rotation speed of 3500 rpm (Fig. 2f). The highest recovery of mercury ions was obtained at 30 min of centrifugation.
Effect of Interference from other ions
To evaluate the feasible analytical applications of the preconcentration technique current, the consequence of several foreign ions interfering with the trace measurement of mercury ions on nanoclay was examined under the optimized conditions. Ions were considered to be interfering when they produced an error larger than ±5% in the recovery of mercury ions. None of the added ions caused interference.
Figures of merit
The figures of merit for mercury ions in the present study were measured under optimal experimental situations. The limit of detection (LOD) established on three times replication of the standard deviations of the blank solution (k = 3, n = 10) turned out to be 0.033 ng/ml. The value for the limit of quantification (LOQ) was 0.11 ng/ml, preconcentration factor 9, dynamic linear range (DLR) from 0.11 ng/ml to 22.2 μg/ml, adsorption capacity was calculated to be 1.17 mg/g. On the whole, full recovery (100%) was obtained under the optimized conditions with standard solutions. The interaction between mercury ions and the Ghezeljeh nanoclay was rapid, with the equilibrium batch process being attained in less than 10 min, meaning that the interaction was thermodynamically favorable.
The experimental process was exercised to a variety of real water and fish samples.
After the parameters involved in the adsorption and desorption steps of mercury ions were optimized, the method used in this research was separately applied to a variety of natural water samples: Caspian Sea, Iran, Karun River (inside and outside the city of Ahvaz), Iran, Persian gulf (20 km away from the coast of Bandar Abbas), Iran, Well water (10 km outside the city of Herat), Afghanistan, Haryrood river (residential area), Afghanistan, Tap water (from Herat city), Afghanistan. Beforehand the examination, the samples were strained through a Whatman blue band filter paper. Prior to the standard addition, pH of the samples was balanced to the best pH level. Spiking tests using multiple standard addition procedure checked trustworthiness. Each natural water sample was spiked with three standard solutions. Mercury ion level was determined by Varian 220Z atomic absorption spectrometer with VGA system. The recovery was explained as the ratio of the concentration of analytes found to the concentration of analytes spiked. The consequences are recorded in Table 2. The recoveries of the added standard solutions were in the cover of 71–101% with low relative standard deviations (less than 2%), which showed that good recoveries can be attained using this way.
The process used in this research was separately employed to five different fish samples: Trout, canned Tuna (Poolak), Persian Gulf Tiger tooth Croaker; Caspian Pike and Caspian Whitefish were purchased from the local markets in Qazvin, Iran. Fish samples were cleaned with distilled water, and made uniform using an electrical mixer and then dried in an electrical oven at temperature of 100°C for 24 h. The dried sample was homogenized again and was stored in polyethylene bottles for subsequent analysis. Samples were digested using wet digestion . One gram was taken from each fish sample. Then, 16 milliliters of a mixture of HNO3 and H2O2 (6:2) was added to each sample. The digestion vessel was heated on the hot plate up to 130°C for 4 h. The sample was permitted to cool. Then, 25 ml of doubly distilled water was increased while stirring. The resulting solution was filtered through Whatman blue band filter paper. Afterwards, appropriate amounts of 4 M sodium hydroxide were added to adjust the pH level. The sample was then diluted using doubly distilled water until it was 50 ml. The digested food samples were poured into an Erlenmeyer flask. Then, Ten mL of the buffer solution was added. After 0.5 min of agitation, 2 g of the Ghezeljeh nanoclay was added. Subsequently, the extraction-preconcentration process was performed. The blank digestions were performed in the same way. Mercury ion level was determined by Varian 220Z atomic absorption spectrometer with VGA system. The results are reported in Table 2. A decline in the recovery of mercury ions was observed. However, the recovery percentage was still significant with low relative standard deviations (less than 4%).
Comparison between this research and similar studies
The detection limit (LOD) and preconcentration factor (P.F) of the offered process for extraction of mercury ions have been matched with those of other extraction procedures reported in the literatures, and the results are summarized in Table 3.
Sorption isotherm is the equation or curve that links the metal concentration that has been adsorbed on the solid part by metal concentration in the solution at balance state for definite temperature. Balance state models can be categorized into empirical and mechanistic models. Empirical models cannot show the mechanisms of the sorbate uptake but can be applied to predict the experimental consequences; however the mechanistic models can describe the system mechanisms. Consequently, mechanistic models explain the fundamental interactions that happen between the metal ions in the solution and the charged surface. Empirical models are commonly established on simple mathematical relations between the liquid part equilibrium concentration and the solid part equilibrium concentration . In terms of the balance state investigation, for many cases, the Langmuir equation is in a good correspondence with the investigational data, whereas the Freundlich equation has also been applied to fit the investigational data in numerous cases. Several investigators have also effectively used other isotherm equations, such as the Temkin and Dubinin–Radushkevich (DR) models to expect the adsorption balance [38,39]. Accordingly, in this study, the isotherm data was examined using the Langmuir, Freundlich, Temkin, and Dubinin–Radushkevich equations.
Adsorption isotherm experiments
A sequences of 50 ml glass bottles were filled up with 30 ml of standard solutions composing of 25 ml of doubly distilled water containing primary concentration of mercury ions from 10 to 50 mg/l, and 5 ml of buffer solution added at pH 7, then an equivalent quantity of Ghezeljeh nanoclay (1 g) was added into each bottle at the required temperature, immediately, adsorption of mercury ions on Ghezeljeh nanoclay was investigated after 5, 10, 15, 20, 30, 40, 50, 100, 150, 200 and 300 min of shaking in incubator shaker. The resulting solutions were centrifuged and the supernatant liquids were subjected for the calculation of mercury ions. The concentration of metal ions residue in the solution was determined by taking the difference of primary and final metal ion concentrations (Fig. 3). Adsorption procedure was calculated by computing the sorption percentage (Adsorption %) as determined by the following:
where C0 is the primary concentration and Ce is the equilibrium concentration, mg/l. The quantity of ions adsorbed per unit mass of adsorbed, qe (mg/g) is estimated by the subsequent expression:
at which V(ml) is the volume of metal ions solution, and w (mg) is the weight of adsorbent. Adsorption isotherm displays the correlation between the adsorption capacity (qe) and the equilibrium concentration (Ce) of ions in the liquid part.
As shown in (Fig. 3), at lesser concentrations, a great amount of adsorption positions on the nanoclay are obtainable for the metal ions and this condition is altered with the addition of metal ion concentration and the competition for adsorption positions gets hard. Adsorption isotherm, at consistent temperature, displays the correlation between the adsorption capacity (qe) and the equilibrium concentration (Ce) of ions in the liquid part. Adsorption isotherm models are commonly applied for fitting the data, and give essential information about the mechanism of adsorption and support us in the plan of new adsorbing structures . In this research, the isotherm data was analyzed by the Temkin, Dubinin–Radushkevich, Langmuir, and Freundlich equations.
The Langmuir equation is frequently applied to explain adsorption of solute from liquid solutions, and its corresponding model accepts the monolayer presentation of the adsorption surface with a finite number of adsorption sites, by monolayer adsorption without any interaction between adsorbed molecules and is expressed by the subsequent equation .
Values of qm and KL are evaluated from the plot of (Ce/qe) verses (1/Ce). Ce is the equilibrium concentration of metal ions onto the adsorbent (mg/L), qe is the quantity of metal ions adsorbed per unit quantity of nanoclay at equilibrium concentration (mg/g), qm (mg/g) is the extreme adsorption capacity, and KL is Langmuir constant correlated for sorption energy; specifically, KL displays adsorption enthalpy which commonly differs with temperature . The Langmuir isotherm was used to our investigational data and the results are revealed in Fig. 4a and Table 4. One of the critical factors of Langmuir equation is the equilibrium or separation factor (RL) (Fig. 4b). RL can be determined by the subsequent equation:
where C0 (mg/L) is the maximum primary solute concentration. The RL indicates the kind of isotherm being acceptable (0 < RL < 1) or unacceptable (RL > 1) or irreversible (RL = 0) . This parameter indicated that nanoclay is an appropriate adsorbent for the adsorption of mercury ions from aqueous solutions.
The Freundlich isotherm model confirms that the surface is dissimilar and the energy of sorption is not consistent. This model moreover accepts the multilayer adsorption. The linear formula of the Freundlich equation is specified in the subsequent equation [20,41],
where qe explains the quantity of metal species adsorbed at equilibrium in mg/g, Ce is the solute equilibrium concentration in mg/l, KF and n are Freundlich consistents correlated to the adsorption capacity and intensity of adsorption, respectively. KF and n were defined from plot of log qe verses log Ce(Table 4). A satisfactory adsorption has n values in the range of 1–10; for this research, we established a favorable value for n.
The Dubinin–Radushkevich isotherm is extra general since it does not accept a uniform surface or consistent sorption potential. It is applied to discriminate between the physical and chemical adsorption of metal ions on surfaces . The Dubinin–Radushkevich calculation is known by Eq. (6):
where qe and qm have the equal significance as before, k (mol2/J2) is a consistent correlated to the adsorption energy, and ε is the extreme adsorption capability known by Eq. (7):
R (J/molK) is the gas constant, and T (K) is the absolute temperature. The constant k gives the mean free energy E (kJ/mol) of sorption per molecule of the sorbate when it is carried to the surface of the solid from an infinite distance in the solution and can be calculated using Eq. (8) and is applied to evaluate the kind of adsorption process. If E < 8 kJ/mol, adsorption procedure is of a physical nature while, if value 8 < E < 16 kJ/mol, the adsorption procedure can be described by ion interchange mechanism.
In this research, small values of E show that the adsorption procedure is a physical nature (Table 4).
Temkin and Pyzhev  evaluated the properties of some indirect adsorbate/adsorbate interactions onto adsorption isotherms. They proposed, regardless the concentration of these adsorbates, their interactions cause that the heat of adsorption of all molecules in the layer would reduce linearly with coverage. The derivation of the Temkin isotherm accepts that the reduction in the heat of sorption is more linear rather than logarithmic, as indicated in the Freundlich equation . The Temkin isotherm has been frequently used in the subsequent formula:
where B = RT/b, R is gas constant (8.314 J/molK), T is the temperature (K), KT is equilibrium binding constant (L/g); b is correlated to heat of adsorption (J/mol). The sorption data can be investigated using Eq. (10). Consequently, a plot of qe versus ln Ce permits one to define the consistent as shown in Table 4.
Data from Fig. 4 and Table 4 show that Langmuir and Freundlich model acceptably fit the experimental data for mercury ions.
To estimate the kinetic mechanism for the adsorption of mercury ions on Ghezeljeh nanoclay, and its potential rate-controlling stages that comprise mass transport and chemical reaction procedures, kinetic models similar to pseudo-first-order, pseudo-second-order, and intra-particle diffusion were examined .
Pseudo- first-order model was commonly explained as follows :
where qe is the quantity of metal ions adsorbed per unit mass of adsorbent at equilibrium i.e., adsorption capacity (mg/g), qt is the quantity of adsorbent adsorbed (mg/g) at any time t and k1is the rate constant. The value of k1was computed from the slope of the linear plot of log (qe−qt) versus t (Table 5).
Pseudo-second-order rate model
Pseudo-second-order rate model is known as follows:
where k2 is the rate constant. The values of k2can be computed from the plot of t / qt versus t; see Fig. 5a, and Table 5.
Intra-particle diffusion model (Waber–Morris model)
To define the rate-controlling stage, intraparticle diffusion model was used to adsorption kinetic data as known via the subsequent equation [38,39].
where qt is the quantity of metal ions adsorbed onto nanoclay at time t, and kint is the rate constant for intraparticle diffusion. Fig. 5b displays a plot of qt verses t1/2. It may show multi-linearity which displays two or more stages occurring in the adsorption procedure. The first sharper part (t ≤ 20 min) is the outside surface adsorption or instant adsorption step. The second part is the slow adsorption step where the intraparticle diffusion rate is measured. The third is the ending equilibrium step where intraparticle diffusion begins to slow down because of enormously small solute concentration in the solution. The factors computed are brought in Table 5. The quantity of kint was superior at the greater concentrations. The multi-stepped adsorption detected for all the metal ions and finest fitting found for the investigational data in height regression coefficient quantities shows that pseudo second order kinetic model might play an important role in the adsorption of metal ions onto nanoclay.
Determination of thermodynamic parameters
Three thermodynamic factors free energy change (ΔG0) enthalpy change (ΔH0) and entropy change (ΔS0) were computed by the subsequent equations:
where R is the universal gas constant, 8.314 J/molK, T is the absolute temperature (K), and kL is the Langmuir constant (mol/l). ΔS0and ΔH0could be found from the slope and intercept of ln kL versus 1/T according to the equation (17). Quantities of ΔS0, ΔH0and ΔG0are displayed in Table 5. The positive values of ΔH° confirm the endothermic nature of the sorption procedure. As clearly shown in Table 5, the positive quantities of ΔG0 is decreased with the rise of sorption temperature, indicating the superior sorption at upper temperature. Also, the positive quantities of entropy change (ΔS0) show that the randomness at the solid–liquid boundary throughout the adsorption procedure increases. The small enthalpy quantities of ΔH° < 20 kJ/mol show that the physisorption is involved in the procedure of adsorption. The estimated quantities of ΔH° for the current system were greater than 20 kJ/mol and therefore, the procedure may include a spontaneous sorption mechanism as ion exchange where chemical links are not of powerful energies .
This research attempted to extract and preconcentrate mercury ions from aqueous solutions with the Ghezeljeh montmorillonite clay as a native adsorbent. To this end, the adsorbent was prepared using the Galehouse way and made distinctive via Fourier transform infrared spectroscopy (FT-IR), scanning electron microscopy-energy dispersive spectrometer operating (SEM-EDS), X-ray diffractometry (XRD), X-ray fluorescence (XRF), cation exchange capacity (CEC) measurements and specific surface area (SBET). The results of XRD, FT-IR, zeta potential and CEC of the Ghezeljeh clay confirm that montmorillonite is the dominant mineral phase. On the basis of SEM images of clay, the distance between the plates is in nm level. A number of effective parameters on extraction-preconcentration were optimized using standard solutions. It was also shown that additional metal ions in the aqueous solution containing mercury ions generally do not have a negative effect on mercury ion recovery. The figures of merit were also calculated: LOD, 0.033 ng/ml; LOQ, 0.11 ng/ml; preconcentration factor, 9; DLR from 0.11 ng/ml to 22.2 μg/ml, and adsorption capacity of the clay was 1.17 mg/g. In the optimized standard solution, full recovery (100%) was obtained. At a later stage, the experimental process was used to a variety of natural water and fish samples under the optimized condition with the recovery being still significant. Good recoveries of spiked samples demonstrate the accuracy of the methods used. In addition, data show that Langmuir and Freundlich models acceptably fit the investigational data for mercury ions. Adsorption of mercury ions onto Ghezeljeh montmorillonite nanoclay obeyed the pseudo-second-order kinetic model. Calculation of ΔG0, ΔH0 and ΔS0 displayed that the nature of mercury ions sorption is endothermic and favorable at upper temperature.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interests regarding the publication of this paper.
The authors are grateful for the financial support of this work by the Imam Khomeini International University (IKIU), and Mines and Mining Industries Development and Renovation Organization of Iran (IMIDRO).