Bentonite from the Birjand area of Iran was characterized by X-ray diffraction, X-ray fluorescence, and Fourier transform infrared spectroscopy. The removal of cadmium from aqueous solution by this bentonite was investigated as a function of conditions such as contact time, metal concentrations, pH, stirring speed, temperature, particle size, and amount of bentonite. The adsorption isotherm was studied with different models: the Freundlich and Dubinin–Radushkevich models had the highest correlation coefficients, 0.9922 and 0.9988, respectively. The corresponding Langmuir model indicates a maximum adsorption capacity of 13.50 mg/g. Firstorder, pseudo-second-order, and intra-particle diffusion equations were used to study the mechanism of adsorption; the experimental data fit well with pseudo-second-order kinetics. Thermodynamic parameters of adsorption were calculated at temperatures of 293, 303, 313, and 323 K, and indicated that the Cd adsorption was exothermic and spontaneous.
The presence of cadmium ions in industrial wastes affects human and aquatic lives (Godt et al. 2006; Lavelle 1995; Sharma 1995). Hence, many processes for Cd removal have been studied (Bedoui et al. 2008; Jianru et al. 2007; Lodeiro et al. 2006; Namasivayam and Ranganathan 1995; Wu and Xiong 2003), and various types of bentonite have been used for the removal of metals and other pollutants (Babel and Kurniawan 2003; Koswojo et al. 2010; Li et al. 2011). The ability to absorb metals such as Cd, Zn, Fe, Hg, Cr, Cu, Mn, and Pb onto bentonite is well recognized (Chen et al. 2011; Inglezakis et al. 2007; Karapinar and Donat 2009; Vieira et al. 2010; Wang et al. 2011) and many studies have focused on optimum conditions, thermodynamics, isotherm models, and kinetics for removal of Cd using bentonite (Hamidpour et al. 2010; Huang et al. 2011; Purna et al. 2006; Zhao et al. 2011). The maximum removal of Cd onto raw and modified bentonite is summarized in Table 1, which shows that the best Cd adsorption belongs to a modified bentonite with an intercalation of 8-hydroxyquinolinium (Bentouami and Ouali 2006). However, the suitability of Iranian bentonite for Cd adsorption has not been characterized; this study was conducted to investigate its effectiveness and key factors involved in the adsorption process.
Materials and Methods A representative sample of bentonite from the Birjand area in southeastern Iran was used without any chemical pretreatment. The sample was ground and sieved by ASTM standard sieves to obtain the nominal particle size fractions of -600 to ?425, -425 to ?300, -300 to ?150, and -150 lm in diameter. X-ray diffraction (XRD) and X-ray fluorescence (XRF) were used to determine the mineralogy of the sample and its elemental analysis, respectively. XRD spectra and XRF were obtained using a Philips X-ray diffractometer 1140 (a = 1.54 A, 40 kV, 30 mA, calibrated with Si-standard) and a Philips X-ray diffractometer Xunique II (80 kV, 40 mA, calibrated with a Si-standard), respectively. The XRD of the sample is shown in Fig. 1, and the elemental analysis of the sample is shown in Table 1. This result indicated that the main mineral of the sample is montmorillonite. All chemical compounds were purchased from Merck and used without further purification.
Infrared spectra from 4,000 to 400 cm-1 were recorded on a Shimadzu 470 FT-IR instrument, using KBr pellets. Fourier transform infrared (FTIR) spectroscopy has been used for chemical functional groups. The Cd concentration after adsorption was determined using a Unicom 939 atomic absorption (AA) spectrometer. The Cd adsorption was calculated by the amount of Cd in solution.
The Cd adsorption experiments were carried out using batch equilibrium. All of the adsorption experiments were conducted in a 250.0 mL glass reactor using a magnetic stirrer for mixing at ambient temperature. In this study, the influence of the parameters, such as mass of bentonite from 5.00 to 60.00 g/L, particle size from -150 to -600 lm, initial Cd concentration from 50.0 to 2,000.0 ppm, contact time from 15 to 180 min, and stirring speed from 400 to 900 rpm were investigated, and the optimized conditions for maximizing Cd adsorption were determined. All other parameters were held constant for the investigation of each parameter in each test. The adsorption isotherms models of Langmuir, Freundlich, Temkin, and Dubinin–Radushkevich (D–R) were studied using 5 g of bentonite added to 100 mL of solution containing different concentrations of Cd, ranging from 50 to 2,000 ppm. All of the solutions were immediately filtered after each test.
For the kinetic investigation, 5 g of bentonite was mixed with 100 mL of Cd at various concentrations of 100.00, 250.00, 500.00, 750.00, 1,000.00, and 1,500.0 mg/L. Each batch test was conducted at various time intervals to determine the period required to reach the adsorption equilibrium and maximum removal of Cd. For the thermodynamic investigation, 5.00 g of bentonite was added to 100 mL of Cd solution at concentrations of 100.0 mg/L and the amount of removed Cd was determined at various temperatures (293, 303, 313, and 323 K). The amount of Cd adsorbed onto the bentonite was determined from the difference between the initial and remaining concentrations of Cd in solution after adsorption.
Result and Discussion
Characterization of the Bentonite
The mineralogical study indicated that quartz, oligoclase, gypsum, montmorillonite, and illite were the main constituents of this sample (Supplementary Fig. 1; supplementary files appear with the on-line version of this paper). The diffraction peaks at 8.9329, 19.8985, and 26.6978 (2O–) correspond to planes (001), (002), and (003), although plane (001) overlapped with the illite diffraction peak. The bentonite contains little illite; the main impurity in this sample was quartz, which is observed in the XRD pattern with the main diffraction peak at 27.7311 (2O–) (Caglar et al. 2009). The main mineral is montmorillonite; the presence of iron likely indicates that some adsorption has already taken place (Supplementary Table 1).
Infrared data are usually used to identify solid-state structures and functional groups of clays. Presence of two stretching bands at 3,627.54 and 3,432.54 cm-1 indicate O–H bond linkage (Supplementary Fig. 2). The stretching band at 3,627 cm-1 is due to the hydroxyl linkage of the bentonite structure and the board band at 3,432 cm-1 indicates the presence of water molecule in the structure, due to hydrogen bonding between hydrogen and oxygen of different water molecules (Wang et al. 2009). The bending vibration band of H–O–H in the water molecule of bentonite is observed at 1,635 cm-1 . The strong and broad band at 1,040 cm-1 can be attributed to Si–O of stretching vibration in the Si–O–Si functional group in tetrahedral sheets, for the montmorillonite structure (Wang et al. 2011). Two bending vibration bands at 467 and 519 cm-1 represent Si–O–Al and Si–O–Si bonds, respectively. The vibration band at 693 cm-1 and its small shoulder band (with a lower intensity, which is not assigned in Supplementary Fig. 2) can be related to the deformation and bending modes of the Si–O–Si bond. The presence of quartz in this sample is indicated by the FTIR shoulder bonding at 796–777 cm-1 (Klinkenberg et al. 2006; Yang et al. 2010).
In order to study the effect of bentonite quantity on the removal of Cd from solution, experiments were conducted with weights of 5.00–60.00 g/L, with the same concentration (1,000.00 ppm Cd2?) at 25 C, a stirring speed of 500 rpm, and a particle size of -150 lm. Supplementary Fig. 3 presents the results of Cd adsorption with different bentonite values. It is obvious that the adsorption percentage is a function of bentonite mass. The increase in adsorbent to liquid ratio caused the percentage of adsorption to increase, to a maximum value at 50.0 g/L of bentonite, as the amount of available sites for ion exchange increased with the increasing mass of bentonite (Abollino et al. 2003). The maximum value of Cd removed from solution for 1.00 g of bentonite was 160.00 mg, which is considerable compared with other reported values (Bentouami and Ouali 2006; Hamidpour et al.