- Email: [email protected]
Removal of 2,4-dichlorophenoxyacetic acid from aqueous solutions by partially characterized organophilic sepiolite: thermodynamic and kinetic calculations
Journal of Colloid and Interface Science 281 (2005) 27–32 www.elsevier.com/locate/jcis
Removal of 2,4-dichlorophenoxyacetic acid from aqueous solutions by partially characterized organophilic sepiolite: thermodynamic and kinetic calculations Gülten Akçay a , Mehmet Akçay a,∗ , Kadir Yurdakoç b a Department of Chemistry, Faculty of Science and Art, University of Dicle, TR 21280, Diyarbakır, Turkey b Department of Chemistry, Faculty of Science and Art, University of Dokuz Eylül, Izmir, ˙ Turkey
Received 5 May 2004; accepted 10 August 2004
Abstract The adsorption of 2,4-dichlorophenoxyacetic acid (2,4-D) on organophilic sepiolite (dodecylammonium sepiolite, DAS) was studied as a function of solution concentration and temperature. The observed adsorption rates were found to be equal to the first-order kinetics. The rate constants were calculated for temperatures ranging between 25 and 40 ◦ C at constant concentration. The adsorption energies, E, and adsorption capacity, qm , for 2,4-D adsorption on organophilic sepiolite was estimated using the Dubinin–Radushkevic equation. Thermodynamic parameters (g a , ha , s a ) were determined by a new approximation from the isotherm of 2,4-D adsorption on DAS. Also, S 0 and H 0 values were calculated from the van’t Hoff equation. These isotherms were modeled according to the Freundlich and Dubinin–Radushkevic adsorption equations. The amount of adsorption of this herbicide on organophilic sepiolite was found to be dependent on the relative energies of adsorbent–adsorbate, adsorbate–solvent, and adsorbate–adsorbate interaction. 2004 Elsevier Inc. All rights reserved. Keywords: Herbicides; Removal; Adsorption; Thermodynamics; Kinetics; Organophilic sepiolite
1. Introduction Sepiolite is a fibrous hydrated magnesium silicate with a formula of [Si12 Mg8 O30 (OH)4 ](H2 O)4 ·8H2 O. The general structure of sepiolite is formed by an alternation of blocks and tunnels that grow up in the fiber direction (c-axis). Each block is constructed by two tetrahedral silica sheets enclosing a central magnesia sheet. In some respects sepiolite is similar to other 2:1 trioctahedral silicates, such as talc, but it has discontinuities and inversions of the silica sheets that give rise to structural tunnels. In the inner blocks, all corners of silica tetrahedra are connected to adjacent blocks, but in outer blocks, some of the corners are Si atoms bound to hydroxyl groups (Si–OH). These silanol groups at the external surface of the silicate are usually accessible to organic * Corresponding author. Fax: +90-412-2488039.
E-mail address: [email protected] (M. Akçay). 0021-9797/$ – see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2004.08.080
species, acting as neutral adsorption sites. In addition to this, some isomorphic substitutions in the tetrahedral sheet of the mineral lattice, such as Al3+ instead of Si4+ , form negatively charged adsorption sites. Such sites are occupied by exchangeable cations that compensate for the electrical charge. These characteristics of sepiolite make it a powerful sorbent for neutral organic molecules and organic cations [1]. Immobilization or separation of contaminants contained in polluted water is an objective of increasing importance in a variety of environmental settings. The extent of pesticide contamination of the water environment has recently raised much concern because of the potential health hazards associated with the entry of these compounds into the food chains of humans and animals [2]. The pollution of the water environment with pesticides is effected through their use in the control of aquatic weeds and insects, leaching runoff from agricultural and forest lands, deposition from aerial applications, and discharge of industrial waste water.
28
G. Akçay et al. / Journal of Colloid and Interface Science 281 (2005) 27–32
The wide range of pesticides used makes it extremely difficult to produce a single method for pesticide disposal that applies universally. Therefore, several methods for removal and disposal of these chemicals may be required to solve this problem. Adsorption on solid sorbents, such as clays or activated carbons, is one of the methods which have been used for removing pesticides from water [3]. In an attempt to design new adsorbents to remove nonionic organic contaminants from waste water, several research workers have replaced the inorganic cations of clays with large alkylammonium cations in order to change the hydrophilic surface of clay into a hydrophobic surface [4,5]. A current environmental concern is the contamination of aquatic systems due to pesticide discharges from manufacturing plants, surface runoff, leaching accidental spills, and other sources. Among the numerous agrochemicals in use today, the herbicide 2,4-D has been widely applied to control broadleaved weeds in gardens and farming. 2,4-D is commonly preferred because of its low cost and good selectivity. It is considered moderately toxic and the maximum allowable concentration in drinking water is 100 ppb. On the other hand, 2,4-D is a poorly biodegradable pollutant. Consequently, it has been frequently detected in water bodies in various regions of the world. The wide range of pesticides used makes it extremely difficult to produce a single method for pesticide disposal that applies universally [6–10]. The removal of malathion and butacolor from aqueous solution has been achieved by organoclays [11]. E.G. Pradas et al. have studied the adsorption of atrazine from aqueous solution on heat-treated kerolites [12]. However, no study on adsorption from aqueous solutions of 2,4-D onto dodecylammonium sepiolite (DAS) has appeared to date. It is therefore the objective of this study to investigate the removal of herbicide contaminants from aqueous solutions by organophilic sepiolite. The present work was also undertaken to investigate the sorption capacity of sepiolite modified by dodecylammonium cations.
2. Material and methods 2.1. The preparation and characterization of organophilic sepiolite The natural sepiolite sample used in this study was obtained from the Eski¸sehir region of Anatolia (Turkey). The sample was characterized by X-ray diffraction, thermogravimetric, and FTIR spectroscopic analyses. The chemical composition of the sepiolite was found to be as follows: SiO2 60.0%, MgO 25.4%, Al2 O3 0.3%, K2 O 0.5%, CaO 0.6%, TiO2 0.05%, Na2 O 0.05%, P2 O5 0.01%, MnO 0.01%, (Fe2 O3 + FeO) 0.04%, 13.5% loss on ignition. The 2,4dichlorophenoxyacetic acid was reagents grade obtained from sigma.
The cation-exchange capacity (CEC) of the sepiolite sample was determined according to the ammonium acetate saturation method and was found to be 82 mmol per 100 g dry clay. The sample have a BET specific surface area of 160 m2 /g. The hydrochloride salt solution of dodecylamine was prepared by mixing the appropriate amount of amine with a 0.05 mol/L HCl solution. The hydrogen ion concentration was approximately 20% in excess of the stoichiometric amount to ensure complete conversion of the amine to the salt form. The clay was previously dried at 110 ◦ C for 24 h and the desiccated samples were mixed with the amine salt, the concentrations of which were greater than the CEC of the clay. The mixture was subjected to mechanical shaking for 42 h at a constant temperature of 25 ◦ C. The treated sample was separated from the mixture by centrifugation and washed several times with ethanol and ethanol–water (1:1) until chloride-free, dried at 40 ◦ C for 24 h, and mechanically ground to 200 mesh. The surface area of the DAS sample was determined after the adsorption of N2 at 77 K using the BET method. The clay sample was analyzed by X-ray powder diffraction using a Siemens D-500 diffractometer and CuKα radiation, thermogravimetric analysis with a Shimadzu TGA 50 analyzer, and infrared spectroscopic analysis with a Maidac Model 1700M FTIR analyses before and after the modification process. The carbon content of the DAS adsorbents was analyzed on a Carlo Erba CHN analyzer and found to have 11.4% carbon. 2.2. Sorption studies The stock solutions were prepared in 10 mmol/L CaCl2 in order to promote the flocculation and to have a constant background electrolyte concentration. The adsorption isotherms were determined by using the batch equilibration technique. The kinetics and equilibrium adsorption of 2,4-D were carried out using the method of aqueous solution. The 20 mL of 2,4-D solutions of constant volume and various concentrations were equilibrated with 0.1 g of adsorbent in 50-mL flasks with Teflon caps. The suspension was shaken for 20 h at 25 ◦ C and centrifuged and the supernatant was analyzed. The equilibrium concentrations of the 2,4-D were calculated from the calibration curves after their absorbance values were measured using a Perkin–Elmer Model Lambda-2S UV–vis spectrophotometer at a wavelength of 230 nm. The solubility was determined to be 0.73 g/dm3, and also the pKa value was found to be 2.80.
3. Results and discussion 3.1. Characterization of the DAS The XRD patterns of natural sepiolite and DAS were recorded and the basal spacings of 12.0 and 12.4 Å were
G. Akçay et al. / Journal of Colloid and Interface Science 281 (2005) 27–32
Fig. 1. XRD patterns: (a) raw sepiolite, (b) DAS samples.
Fig. 2. TG and DTG thermogram of samples: (a) sepiolite, (b) DAS.
observed. The expansion in the basal spacing of the sepiolite due to the intercalation of DAS was calculated as d = d − 12 Å, where d is the basal spacing of the DAtreated sepiolite and 12 Å is the thickness of a clay layer. d is found to be 0.4 Å. This observation suggest that DA ions intercalate into the interlayers of sepiolite with a monolayer arrangement. The XRD patterns of sepiolite and DAsepiolite samples are shown in Fig. 1. The surface areas of the S and DAS were determined to be 160 and 81 m2 /g by the BET method. The TG and DTG thermograms of raw sepiolite and DAS are shown in Fig. 2a and 2b. Fig. 2a exhibits desorption maxima near to 84, 305, and 672 ◦ C which can be attributed to the desorption of physically adsorbed water and the dehydroxylation of the magnesium silicate layer, respectively.
29
Fig. 3. FT-IR spectra of the raw sepiolite and DAS.
Fig. 2b exhibits desorption maxima near 190 ◦ C, which can be attributed to the deterioration of dodecylammonium. Sepiolite is a clay mineral which is a fibrous magnesium silicate of ideal formula [Si12 Mg8 O30 (OH)4 ](H2 O)4 ·8H2 O. The mesopore size (an average of 37 Å2 ) of sepiolite is extremely interesting in this regard. In addition to its microfibrous morphology, sepiolite is formed by the alternation of blocks and tunnels that grow up in the fiber direction. Each structural block is made up of two tetrahedral silica sheets sandwiching a central sheet of magnesium oxide– hydroxide. As the silica sheets are discontinuous, silanol (Si–OH) groups are present on the external surface of the silica particles [13]. These groups, located at the edges of the channels, are directly accessible to various organic and inorganic reagents [14]. The cross section of sepiolite tunnels is about 11 × 4 Å. The tunnels shelter two types of water molecules: (a) coordinated water molecules bonded to Mg+2 ions located at the edges of the octahedral sheet, and (b) zeolitic water bonded to the coordinated water molecules through hydrogen bonding [15,16]. Ruiz-Hitzky [15] proposes a preferential molecular arrangement of pyridine molecules in the tunnels with the ring lying parallel to the (d100 ) plane of the sepiolite crystal. Due to the crosssectional area of the pyridine molecule being nearly 2.65 Å, the pyridine molecules enter these tunnels [17,18]. The partial characterization of the sepiolites used has been reported in detail in [19]. IR spectra have been recorded in situ at room temperature. The FT-IR spectra of DA adsorbed onto the sepiolite are shown in Fig. 3. The strong absorption bands at 2930 and 2856 cm−1 indicate CH3 and CH2 groups. The weak absorption band at 1468 cm−1 is attributed to ammonium ions. 3.2. The Freundlich and Dubinin–Radushkevic equations The data obtained from the adsorption experiments carried out with 2,4-D with concentrations between 1 and 0.12 mmol/L were fitted with the Freundlich and Dubinin–
30
G. Akçay et al. / Journal of Colloid and Interface Science 281 (2005) 27–32
Fig. 4. The linear Freundlich isotherm for 2,4-D sorption onto DAS at 25 ◦ C.
Fig. 5. The linear Dubinin–Radushkevic isotherm for 2,4-D sorption onto DAS at 25 ◦ C.
Table 1 Sorption constants obtained from the Freundlich and Dubinin–Radushkevic adsorption isotherms of 2,4-D at 25 ◦ C Sample Freundlich n 2,4-D
Dubinin–Radushkevic
R2 KF (mol/g)
0.971 2.44
β E R2 qm (mmol/g) (mol2 /kJ2 ) (kJ/mol)
0.997 4.56
8.4 × 10−3
−7.72
0.994
Radushkevic equations (D–R). The following Freundlich equation was used to describe the equilibrium data [20], ln qe = ln KF + 1/n ln Ce ,
(1)
where KF and n are characteristic constants of adsorbent and adsorbate. The single-solute sorption isotherm for organophilic sepiolite at 25 ◦ C is shown in Fig. 4. The regression equations at 25 ◦ C, along with the constants KF and n and the correlation coefficient R 2 , are listed in Table 1. The D–R isotherm is more general than the Langmuir isotherm, because it does not assume a homogeneous surface or a constant sorption potential [21,22]. The D–R equation is qe = qm exp −βε2 , (2) where qe is the amount of 2,4-D adsorbed at equilibrium, β is a constant related to the adsorption energy, qm is the theoretical saturation capacity, and ε is the Polanyi potential, which is equal to −RT exp(1/Ce ). The linear form of Eq. (2) is ln qe = ln qm − βε2 .
(3)
The slope of the plot of ln qe versus ε2 gives β (mol2 /kJ2 ) and the intercept yields the sorption capacity, qm (mol/g). The D–R parameters are listed in Table 1. According to the following equation [22], the sorption free energy was obtained: E = (2β)−1/2.
(4)
The magnitude of E is useful for estimating the type of adsorption process. The magnitude of sorption energy may give an idea about the type of sorption. Two main types of adsorption may occur: physical and chemical adsorption. In physical adsorption equilibrium is usually rapidly attained
Fig. 6. The adsorption curves of 2,4-D obtained at 0.4 mmol/dm3 for initial 2,4-D concentration for 25 and 40 ◦ C.
Table 2 Kinetic parameters for 2,4-dichlorophenoxyacetic acid adsorption onto DAS at 25 and 40 ◦ Ca Temperature (◦ C)
Kads (mol/g min)
qe (µmol/g)
R2
t1/2 (min)
Ce (mol/L)
25 40
0.0154 0.0247
20.20 12.75
0.921 0.962
45.42 28.16
0.154 0.116
a Adsorption conditions: initial concentration = 0.4 mmol/L, adsorbent dosage = 1 g/L, and agitation speed = 160 rpm.
and easily reversible, because the energy requirements are small (usually no more than 4.2 kJ/mol). Chemical adsorption is specific adsorption due to the forces involved in the physical adsorption. Therefore, the sorption energy for chemical adsorption is of the same magnitude as the head of chemical reactions. Fig. 5 shows the D–R isotherm for 2,4-D sorption onto DAS. As shown in Table 1, data correlating with Freundlich and D–R equations are used to interpret the results. 3.3. Sorption kinetics of 2,4-D Plots of qt vs t at 25 and 40 ◦ C are shown in Fig. 6. The sorption process still confirms the pseudo-first-order model with high correlation coefficient. The values of model parameters are given in Table 2. The rate of 2,4-D adsorption by DAS at 25 and 40 ◦ C tends to follow Lagergren’s first-order rate equation [23], ln(qe − qt ) = ln qe − Kads t,
(5)
G. Akçay et al. / Journal of Colloid and Interface Science 281 (2005) 27–32
31
Table 3 Thermodynamic parameters for 2,4-dichlorophenoxyacetic acid adsorption onto DAS at 25 ◦ C g a ha s a H 0 S 0 Temperature Kd (L/g) (kJ/mol) (kJ/mol) (kJ/mol K) (kJ/mol) (kJ/mol K) (◦ C) 25 40
0.319 −2.831 0.489 −1.863
−24.577 −0.0633 −25.440 −0.0812
−23.084 −0.0742
0 These values were calculated by the van’t Hoff equation.
Fig. 7. The determination of the rate constant for sorption of 2,4-D by DAS in a single-solute system at 25 and 40 ◦ C.
where qt is the amount of adsorbate adsorbed at time t, qe is the amount of herbicides adsorbed at equilibrium, and Kads is the adsorption rate constant. Kads values were obtained in the linear plots (Fig. 7). The adsorption rate constants were determined from the slopes of the plots. The sorption is rapid during the first 30 min and equilibrium is then attained within 120 min. The Kads constant for 2,4-D sorption on organophilic sepiolite was determined at 25 and 40 ◦ C, respectively, in Table 2.
The amounts of sorption of 2,4-D by DAS clay are measured at temperatures of 25 and 40 ◦ C. The integrals of free energy, enthalpy, and entropy (Ga , H a , and S a ) are measured in joules per kilogram on adsorbent. The integral and immersional functions derived from solution thermodynamics are related to the molar and differential functions generated by surface thermodynamics. Integral functions are converted to molar variables by dividing each function by the total amount adsorbed nai , g a = Ga /nai , ha = H a /nai , s a = S a /nai , where the molar integral functions g a and ha have units of J/mol. The differential functions for component i in the mixture are obtained from the integral functions. Differentiation of the functions Ga , H a , and S a yields differential functions identical with the molar functions. The differential mixture enthalpies (whose absolute values are called isosteric heats in the literature on adsorption) are functions of loading and can be measured experimentally or predicted from single-gas adsorption data [24]. For component adsorption, the integral is na H =
a
a
h dn
(constant T ).
ha = RT ln Cads .
(6)
0
The differential enthalpy ha is a function of loading na , but its variation with temperature is weak and the assumption of its constancy over a moderate range of temperatures is a useful approximation. The enthalpy depends on the volume of the adsorbed phase because initially the volume of the adsorbed phase is zero [25]. These approximations were
(7)
The entropic portion of the free energy term will be approximated as shown in the equation s a = R ln(Cads /Cfree ),
(8)
where the change in entropy is from concentration of the adsorbed component, Cads , relative to concentration of the free component, Cfree . The units of Cads and Cfree are mol/dm3. The equilibrium partition constant Ke can be calculated as Kd = Cads /Cfree .
3.4. Thermodynamic parameters
a
applied to the adsorption of organic substances [26–29] from aqueous solutions. Thus, we will approximate the differential or molar enthalpy in the solution (ha ) as shown in the equation
(9)
The differential Gibbs free energy term of the adsorption can be given as shown in the equations g a = RT ln Kd .
(10)
The relationship between Kd and temperature is given by the van’t Hoff equation, ln Kd = (S 0 /R) − (H 0 /RT ).
(11)
Molar enthalpies, entropies, and Gibbs free energies of 2,4-D on DAS were presented in Table 3. At low Ce , the interactions between adsorbate and adsorbent are displayed; direct interaction between solute and adsorbent and formation of donor–acceptor adducts can explain the high adsorption enthalpy change of 2,4-D. In addition and the increase of the interaction between adsorbate and adsorbent, the interaction between solute in solution and solute adsorbed on sorbent gradually is also seen.
4. Conclusion The Freundlich and Dubinin–Radushkevic adsorption models were used for the mathematical description of the adsorption equilibrium of 2,4-D onto DAS depending on temperature, and isotherm constants evaluated from the isotherms were used to compare the adsorptive capacity of the DAS. The adsorption of 2,4-D onto DAS was found to exhibit nonlinear favorable adsorption behavior that could be characterized well by the Freundlich and Dubinin– Radushkevic isotherm models in the studied concentration range at all the temperatures studied.
32
G. Akçay et al. / Journal of Colloid and Interface Science 281 (2005) 27–32
The magnitude of E is useful for estimating the type of adsorption process. The magnitude of sorption energy may give an idea about the type of sorption. Two main types of adsorption may occur: physical and chemical adsorption. In physical adsorption equilibrium is usually rapidly attained and easily reversible, because the energy requirements are small (usually no more than 4.2 kJ/mol), since the forces involved in the physical adsorption are weak. The chemical adsorption is specific and involves forces much stronger than those in physical adsorption. So the sorption energy for chemical adsorption is of the same magnitude as the head of chemical reactions (8.4 and 83.7 kJ/mol) [22]. Thermodynamic constants were also evaluated using equilibrium constants changing with temperature. The negative values of g a , ha , and s a showed the exothermic nature. In addition, the values are indicative of the adsorption, both physical and chemical. The following conclusions can be drawn from this study: (a) By analyzing the qm values it can be seen that affinity of adsorbent to this sample is high. (b) By analyzing differential adsorption enthalpies it is seen that organophilic sepiolite is a good adsorbent for this sample. (c) By analyzing KF and n values it can be said that organophilic sepiolite is a good adsorbent for removing organic pollutant. (d) By analyzing sorption energy (E) values it can be seen that DAS is a good adsorbent for this compound. (e) It is seen that DAS as a adsorbent is very convenient for 2,4-D at low and high temperatures. By using these kinetic and thermodynamic parameters, we believe that sorption by DAS can be suitable for the purification of waste water and the removal of 2,4-D from industrial waste waters.
Acknowledgment The authors express their thanks to TU Chemnitz, Germany, for XRD and BET surface area results.
References [1] A. Alvarez, A. Singer, E. Galan, Development in Sedimentology, vol. 37, Elsevier, Amsterdam, 1984, p. 253. [2] E. Gonzalez-Pradas, M. Villafranca-Sanchez, A. Gallego Campo, D. Urena-Amate, M. Fernandez-Perez, J. Chem. Technol. Biotechnol. 74 (1999) 49. [3] E. Gonzalez-Pradas, M. Villafranca-Sanchez, A. Gallego Campo, D. Urena-Amate, M. Fernandez-Perez, J. Chem. Technol. Biotechnol. 56 (1993) 67. [4] G. Sheng, S. Xu, S.A. Boyd, Water Res. 30 (1996) 1483. [5] L. Zhu, Y. Li, J. Zhang, Environ. Sci. Technol. 31 (1997) 1407. [6] A. Farba, R. Dufford, A. Evangelista de Dufford, Bull. Environ. Contam. Toxicol. 59 (1997) 645. [7] R. Celis, M.C. Hermosin, L. Cox, J. Cornejo, Environ. Sci. Technol. 33 (1999) 1200. [8] M. Mahramanlıo˘glu, I. Kızılcıklı, I.O. Biçer, M. Tunçay, J. Environ. Sci. Health B 35 (2000) 187. [9] J.L. Sotelo, G. Ovejero, A. Delgado, I. Martinez, Water. Res. 36 (2001) 599. [10] P. Benoit, E. Barriuso, R. Calvet, Chemosphere 37 (1998) 1271. [11] O.R. Pal, A.K. Vanjara, Sep. Purif. Technol. 24 (2001) 167. [12] E. Gonzalez-Pradas, M. Socias-Viciara, M. Sufi, M.D. Urena-Amate, F. Flores-Cespedes, M. Fernandez-Perez, M. Villafranca-Sanches, Chemosphere 5 (2003) 85. [13] J.L. Ahlrichs, C. Serna, J.M. Serratosa, Clays Clay Miner. 23 (1975) 119. [14] M.N. Fernandez-Hernandez, E. Ruiz-Hitzky, Clay Miner. 14 (1979) 295. [15] E. Ruiz-Hitzky, J. Mater. Chem. 11 (2001) 86. [16] B. Casal, J. Merino, J.M. Serratosa, E. Ruiz-Hitzky, Appl. Clay Sci. 18 (2001) 245. [17] T.J. Dines, L.D. MacGregor, C.H. Rochester, J. Colloid Interface Sci. 245 (2002) 221. [18] E. Sabah, M.S. Çelik, J. Colloid Interface Sci. 251 (2002) 33. [19] M. Akçay, J. Mol. Struct. 694 (2004) 21. [20] F.L. Slejko, Adsorption Technology Step by Step, Dekker, New York, 1985, p. 13. [21] S.H. Lin, R.S. Juang, J. Hazard. Mater. 92 (2002) 315. [22] C. Tohsieh, H. Teng, J. Colloid Interface Sci. 230 (2000) 171. [23] E. Haribabu, Y.D. Upadhya, S.N. Upadhyay, J. Environ. Stud. 43 (1993) 169. [24] F.R. Sperstein, A.L. Myers, AIChE J. 47 (2001) 1141. [25] A.L. Myers, AIChE J. 48 (2002) 145. [26] S.A. Morton III, D.J. Kefer, R.M. Counce, D.W. DePaoli, M.Z.C. Hu, J. Colloid Interface Sci. 270 (2004) 229. [27] A.R. Cestari, F.S. Eunice Vieira, A. Simoni Jose de, C. Airoldi, Thermochim. Acta 384 (2000) 25. [28] M.G. Fonseca, A.S. Jose, C. Airoldi, Thermochim. Acta 369 (2001) 17. [29] M.G. Fonseca, C. Airoldi, J. Colloid Interface Sci. 240 (2001) 229.
Removal of 2,4-dichlorophenoxyacetic acid from aqueous solutions by partially characterized organophilic sepiolite: thermodynamic and kinetic calculations Gülten Akçay a , Mehmet Akçay a,∗ , Kadir Yurdakoç b a Department of Chemistry, Faculty of Science and Art, University of Dicle, TR 21280, Diyarbakır, Turkey b Department of Chemistry, Faculty of Science and Art, University of Dokuz Eylül, Izmir, ˙ Turkey
Received 5 May 2004; accepted 10 August 2004
Abstract The adsorption of 2,4-dichlorophenoxyacetic acid (2,4-D) on organophilic sepiolite (dodecylammonium sepiolite, DAS) was studied as a function of solution concentration and temperature. The observed adsorption rates were found to be equal to the first-order kinetics. The rate constants were calculated for temperatures ranging between 25 and 40 ◦ C at constant concentration. The adsorption energies, E, and adsorption capacity, qm , for 2,4-D adsorption on organophilic sepiolite was estimated using the Dubinin–Radushkevic equation. Thermodynamic parameters (g a , ha , s a ) were determined by a new approximation from the isotherm of 2,4-D adsorption on DAS. Also, S 0 and H 0 values were calculated from the van’t Hoff equation. These isotherms were modeled according to the Freundlich and Dubinin–Radushkevic adsorption equations. The amount of adsorption of this herbicide on organophilic sepiolite was found to be dependent on the relative energies of adsorbent–adsorbate, adsorbate–solvent, and adsorbate–adsorbate interaction. 2004 Elsevier Inc. All rights reserved. Keywords: Herbicides; Removal; Adsorption; Thermodynamics; Kinetics; Organophilic sepiolite
1. Introduction Sepiolite is a fibrous hydrated magnesium silicate with a formula of [Si12 Mg8 O30 (OH)4 ](H2 O)4 ·8H2 O. The general structure of sepiolite is formed by an alternation of blocks and tunnels that grow up in the fiber direction (c-axis). Each block is constructed by two tetrahedral silica sheets enclosing a central magnesia sheet. In some respects sepiolite is similar to other 2:1 trioctahedral silicates, such as talc, but it has discontinuities and inversions of the silica sheets that give rise to structural tunnels. In the inner blocks, all corners of silica tetrahedra are connected to adjacent blocks, but in outer blocks, some of the corners are Si atoms bound to hydroxyl groups (Si–OH). These silanol groups at the external surface of the silicate are usually accessible to organic * Corresponding author. Fax: +90-412-2488039.
E-mail address: [email protected] (M. Akçay). 0021-9797/$ – see front matter 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2004.08.080
species, acting as neutral adsorption sites. In addition to this, some isomorphic substitutions in the tetrahedral sheet of the mineral lattice, such as Al3+ instead of Si4+ , form negatively charged adsorption sites. Such sites are occupied by exchangeable cations that compensate for the electrical charge. These characteristics of sepiolite make it a powerful sorbent for neutral organic molecules and organic cations [1]. Immobilization or separation of contaminants contained in polluted water is an objective of increasing importance in a variety of environmental settings. The extent of pesticide contamination of the water environment has recently raised much concern because of the potential health hazards associated with the entry of these compounds into the food chains of humans and animals [2]. The pollution of the water environment with pesticides is effected through their use in the control of aquatic weeds and insects, leaching runoff from agricultural and forest lands, deposition from aerial applications, and discharge of industrial waste water.
28
G. Akçay et al. / Journal of Colloid and Interface Science 281 (2005) 27–32
The wide range of pesticides used makes it extremely difficult to produce a single method for pesticide disposal that applies universally. Therefore, several methods for removal and disposal of these chemicals may be required to solve this problem. Adsorption on solid sorbents, such as clays or activated carbons, is one of the methods which have been used for removing pesticides from water [3]. In an attempt to design new adsorbents to remove nonionic organic contaminants from waste water, several research workers have replaced the inorganic cations of clays with large alkylammonium cations in order to change the hydrophilic surface of clay into a hydrophobic surface [4,5]. A current environmental concern is the contamination of aquatic systems due to pesticide discharges from manufacturing plants, surface runoff, leaching accidental spills, and other sources. Among the numerous agrochemicals in use today, the herbicide 2,4-D has been widely applied to control broadleaved weeds in gardens and farming. 2,4-D is commonly preferred because of its low cost and good selectivity. It is considered moderately toxic and the maximum allowable concentration in drinking water is 100 ppb. On the other hand, 2,4-D is a poorly biodegradable pollutant. Consequently, it has been frequently detected in water bodies in various regions of the world. The wide range of pesticides used makes it extremely difficult to produce a single method for pesticide disposal that applies universally [6–10]. The removal of malathion and butacolor from aqueous solution has been achieved by organoclays [11]. E.G. Pradas et al. have studied the adsorption of atrazine from aqueous solution on heat-treated kerolites [12]. However, no study on adsorption from aqueous solutions of 2,4-D onto dodecylammonium sepiolite (DAS) has appeared to date. It is therefore the objective of this study to investigate the removal of herbicide contaminants from aqueous solutions by organophilic sepiolite. The present work was also undertaken to investigate the sorption capacity of sepiolite modified by dodecylammonium cations.
2. Material and methods 2.1. The preparation and characterization of organophilic sepiolite The natural sepiolite sample used in this study was obtained from the Eski¸sehir region of Anatolia (Turkey). The sample was characterized by X-ray diffraction, thermogravimetric, and FTIR spectroscopic analyses. The chemical composition of the sepiolite was found to be as follows: SiO2 60.0%, MgO 25.4%, Al2 O3 0.3%, K2 O 0.5%, CaO 0.6%, TiO2 0.05%, Na2 O 0.05%, P2 O5 0.01%, MnO 0.01%, (Fe2 O3 + FeO) 0.04%, 13.5% loss on ignition. The 2,4dichlorophenoxyacetic acid was reagents grade obtained from sigma.
The cation-exchange capacity (CEC) of the sepiolite sample was determined according to the ammonium acetate saturation method and was found to be 82 mmol per 100 g dry clay. The sample have a BET specific surface area of 160 m2 /g. The hydrochloride salt solution of dodecylamine was prepared by mixing the appropriate amount of amine with a 0.05 mol/L HCl solution. The hydrogen ion concentration was approximately 20% in excess of the stoichiometric amount to ensure complete conversion of the amine to the salt form. The clay was previously dried at 110 ◦ C for 24 h and the desiccated samples were mixed with the amine salt, the concentrations of which were greater than the CEC of the clay. The mixture was subjected to mechanical shaking for 42 h at a constant temperature of 25 ◦ C. The treated sample was separated from the mixture by centrifugation and washed several times with ethanol and ethanol–water (1:1) until chloride-free, dried at 40 ◦ C for 24 h, and mechanically ground to 200 mesh. The surface area of the DAS sample was determined after the adsorption of N2 at 77 K using the BET method. The clay sample was analyzed by X-ray powder diffraction using a Siemens D-500 diffractometer and CuKα radiation, thermogravimetric analysis with a Shimadzu TGA 50 analyzer, and infrared spectroscopic analysis with a Maidac Model 1700M FTIR analyses before and after the modification process. The carbon content of the DAS adsorbents was analyzed on a Carlo Erba CHN analyzer and found to have 11.4% carbon. 2.2. Sorption studies The stock solutions were prepared in 10 mmol/L CaCl2 in order to promote the flocculation and to have a constant background electrolyte concentration. The adsorption isotherms were determined by using the batch equilibration technique. The kinetics and equilibrium adsorption of 2,4-D were carried out using the method of aqueous solution. The 20 mL of 2,4-D solutions of constant volume and various concentrations were equilibrated with 0.1 g of adsorbent in 50-mL flasks with Teflon caps. The suspension was shaken for 20 h at 25 ◦ C and centrifuged and the supernatant was analyzed. The equilibrium concentrations of the 2,4-D were calculated from the calibration curves after their absorbance values were measured using a Perkin–Elmer Model Lambda-2S UV–vis spectrophotometer at a wavelength of 230 nm. The solubility was determined to be 0.73 g/dm3, and also the pKa value was found to be 2.80.
3. Results and discussion 3.1. Characterization of the DAS The XRD patterns of natural sepiolite and DAS were recorded and the basal spacings of 12.0 and 12.4 Å were
G. Akçay et al. / Journal of Colloid and Interface Science 281 (2005) 27–32
Fig. 1. XRD patterns: (a) raw sepiolite, (b) DAS samples.
Fig. 2. TG and DTG thermogram of samples: (a) sepiolite, (b) DAS.
observed. The expansion in the basal spacing of the sepiolite due to the intercalation of DAS was calculated as d = d − 12 Å, where d is the basal spacing of the DAtreated sepiolite and 12 Å is the thickness of a clay layer. d is found to be 0.4 Å. This observation suggest that DA ions intercalate into the interlayers of sepiolite with a monolayer arrangement. The XRD patterns of sepiolite and DAsepiolite samples are shown in Fig. 1. The surface areas of the S and DAS were determined to be 160 and 81 m2 /g by the BET method. The TG and DTG thermograms of raw sepiolite and DAS are shown in Fig. 2a and 2b. Fig. 2a exhibits desorption maxima near to 84, 305, and 672 ◦ C which can be attributed to the desorption of physically adsorbed water and the dehydroxylation of the magnesium silicate layer, respectively.
29
Fig. 3. FT-IR spectra of the raw sepiolite and DAS.
Fig. 2b exhibits desorption maxima near 190 ◦ C, which can be attributed to the deterioration of dodecylammonium. Sepiolite is a clay mineral which is a fibrous magnesium silicate of ideal formula [Si12 Mg8 O30 (OH)4 ](H2 O)4 ·8H2 O. The mesopore size (an average of 37 Å2 ) of sepiolite is extremely interesting in this regard. In addition to its microfibrous morphology, sepiolite is formed by the alternation of blocks and tunnels that grow up in the fiber direction. Each structural block is made up of two tetrahedral silica sheets sandwiching a central sheet of magnesium oxide– hydroxide. As the silica sheets are discontinuous, silanol (Si–OH) groups are present on the external surface of the silica particles [13]. These groups, located at the edges of the channels, are directly accessible to various organic and inorganic reagents [14]. The cross section of sepiolite tunnels is about 11 × 4 Å. The tunnels shelter two types of water molecules: (a) coordinated water molecules bonded to Mg+2 ions located at the edges of the octahedral sheet, and (b) zeolitic water bonded to the coordinated water molecules through hydrogen bonding [15,16]. Ruiz-Hitzky [15] proposes a preferential molecular arrangement of pyridine molecules in the tunnels with the ring lying parallel to the (d100 ) plane of the sepiolite crystal. Due to the crosssectional area of the pyridine molecule being nearly 2.65 Å, the pyridine molecules enter these tunnels [17,18]. The partial characterization of the sepiolites used has been reported in detail in [19]. IR spectra have been recorded in situ at room temperature. The FT-IR spectra of DA adsorbed onto the sepiolite are shown in Fig. 3. The strong absorption bands at 2930 and 2856 cm−1 indicate CH3 and CH2 groups. The weak absorption band at 1468 cm−1 is attributed to ammonium ions. 3.2. The Freundlich and Dubinin–Radushkevic equations The data obtained from the adsorption experiments carried out with 2,4-D with concentrations between 1 and 0.12 mmol/L were fitted with the Freundlich and Dubinin–
30
G. Akçay et al. / Journal of Colloid and Interface Science 281 (2005) 27–32
Fig. 4. The linear Freundlich isotherm for 2,4-D sorption onto DAS at 25 ◦ C.
Fig. 5. The linear Dubinin–Radushkevic isotherm for 2,4-D sorption onto DAS at 25 ◦ C.
Table 1 Sorption constants obtained from the Freundlich and Dubinin–Radushkevic adsorption isotherms of 2,4-D at 25 ◦ C Sample Freundlich n 2,4-D
Dubinin–Radushkevic
R2 KF (mol/g)
0.971 2.44
β E R2 qm (mmol/g) (mol2 /kJ2 ) (kJ/mol)
0.997 4.56
8.4 × 10−3
−7.72
0.994
Radushkevic equations (D–R). The following Freundlich equation was used to describe the equilibrium data [20], ln qe = ln KF + 1/n ln Ce ,
(1)
where KF and n are characteristic constants of adsorbent and adsorbate. The single-solute sorption isotherm for organophilic sepiolite at 25 ◦ C is shown in Fig. 4. The regression equations at 25 ◦ C, along with the constants KF and n and the correlation coefficient R 2 , are listed in Table 1. The D–R isotherm is more general than the Langmuir isotherm, because it does not assume a homogeneous surface or a constant sorption potential [21,22]. The D–R equation is qe = qm exp −βε2 , (2) where qe is the amount of 2,4-D adsorbed at equilibrium, β is a constant related to the adsorption energy, qm is the theoretical saturation capacity, and ε is the Polanyi potential, which is equal to −RT exp(1/Ce ). The linear form of Eq. (2) is ln qe = ln qm − βε2 .
(3)
The slope of the plot of ln qe versus ε2 gives β (mol2 /kJ2 ) and the intercept yields the sorption capacity, qm (mol/g). The D–R parameters are listed in Table 1. According to the following equation [22], the sorption free energy was obtained: E = (2β)−1/2.
(4)
The magnitude of E is useful for estimating the type of adsorption process. The magnitude of sorption energy may give an idea about the type of sorption. Two main types of adsorption may occur: physical and chemical adsorption. In physical adsorption equilibrium is usually rapidly attained
Fig. 6. The adsorption curves of 2,4-D obtained at 0.4 mmol/dm3 for initial 2,4-D concentration for 25 and 40 ◦ C.
Table 2 Kinetic parameters for 2,4-dichlorophenoxyacetic acid adsorption onto DAS at 25 and 40 ◦ Ca Temperature (◦ C)
Kads (mol/g min)
qe (µmol/g)
R2
t1/2 (min)
Ce (mol/L)
25 40
0.0154 0.0247
20.20 12.75
0.921 0.962
45.42 28.16
0.154 0.116
a Adsorption conditions: initial concentration = 0.4 mmol/L, adsorbent dosage = 1 g/L, and agitation speed = 160 rpm.
and easily reversible, because the energy requirements are small (usually no more than 4.2 kJ/mol). Chemical adsorption is specific adsorption due to the forces involved in the physical adsorption. Therefore, the sorption energy for chemical adsorption is of the same magnitude as the head of chemical reactions. Fig. 5 shows the D–R isotherm for 2,4-D sorption onto DAS. As shown in Table 1, data correlating with Freundlich and D–R equations are used to interpret the results. 3.3. Sorption kinetics of 2,4-D Plots of qt vs t at 25 and 40 ◦ C are shown in Fig. 6. The sorption process still confirms the pseudo-first-order model with high correlation coefficient. The values of model parameters are given in Table 2. The rate of 2,4-D adsorption by DAS at 25 and 40 ◦ C tends to follow Lagergren’s first-order rate equation [23], ln(qe − qt ) = ln qe − Kads t,
(5)
G. Akçay et al. / Journal of Colloid and Interface Science 281 (2005) 27–32
31
Table 3 Thermodynamic parameters for 2,4-dichlorophenoxyacetic acid adsorption onto DAS at 25 ◦ C g a ha s a H 0 S 0 Temperature Kd (L/g) (kJ/mol) (kJ/mol) (kJ/mol K) (kJ/mol) (kJ/mol K) (◦ C) 25 40
0.319 −2.831 0.489 −1.863
−24.577 −0.0633 −25.440 −0.0812
−23.084 −0.0742
0 These values were calculated by the van’t Hoff equation.
Fig. 7. The determination of the rate constant for sorption of 2,4-D by DAS in a single-solute system at 25 and 40 ◦ C.
where qt is the amount of adsorbate adsorbed at time t, qe is the amount of herbicides adsorbed at equilibrium, and Kads is the adsorption rate constant. Kads values were obtained in the linear plots (Fig. 7). The adsorption rate constants were determined from the slopes of the plots. The sorption is rapid during the first 30 min and equilibrium is then attained within 120 min. The Kads constant for 2,4-D sorption on organophilic sepiolite was determined at 25 and 40 ◦ C, respectively, in Table 2.
The amounts of sorption of 2,4-D by DAS clay are measured at temperatures of 25 and 40 ◦ C. The integrals of free energy, enthalpy, and entropy (Ga , H a , and S a ) are measured in joules per kilogram on adsorbent. The integral and immersional functions derived from solution thermodynamics are related to the molar and differential functions generated by surface thermodynamics. Integral functions are converted to molar variables by dividing each function by the total amount adsorbed nai , g a = Ga /nai , ha = H a /nai , s a = S a /nai , where the molar integral functions g a and ha have units of J/mol. The differential functions for component i in the mixture are obtained from the integral functions. Differentiation of the functions Ga , H a , and S a yields differential functions identical with the molar functions. The differential mixture enthalpies (whose absolute values are called isosteric heats in the literature on adsorption) are functions of loading and can be measured experimentally or predicted from single-gas adsorption data [24]. For component adsorption, the integral is na H =
a
a
h dn
(constant T ).
ha = RT ln Cads .
(6)
0
The differential enthalpy ha is a function of loading na , but its variation with temperature is weak and the assumption of its constancy over a moderate range of temperatures is a useful approximation. The enthalpy depends on the volume of the adsorbed phase because initially the volume of the adsorbed phase is zero [25]. These approximations were
(7)
The entropic portion of the free energy term will be approximated as shown in the equation s a = R ln(Cads /Cfree ),
(8)
where the change in entropy is from concentration of the adsorbed component, Cads , relative to concentration of the free component, Cfree . The units of Cads and Cfree are mol/dm3. The equilibrium partition constant Ke can be calculated as Kd = Cads /Cfree .
3.4. Thermodynamic parameters
a
applied to the adsorption of organic substances [26–29] from aqueous solutions. Thus, we will approximate the differential or molar enthalpy in the solution (ha ) as shown in the equation
(9)
The differential Gibbs free energy term of the adsorption can be given as shown in the equations g a = RT ln Kd .
(10)
The relationship between Kd and temperature is given by the van’t Hoff equation, ln Kd = (S 0 /R) − (H 0 /RT ).
(11)
Molar enthalpies, entropies, and Gibbs free energies of 2,4-D on DAS were presented in Table 3. At low Ce , the interactions between adsorbate and adsorbent are displayed; direct interaction between solute and adsorbent and formation of donor–acceptor adducts can explain the high adsorption enthalpy change of 2,4-D. In addition and the increase of the interaction between adsorbate and adsorbent, the interaction between solute in solution and solute adsorbed on sorbent gradually is also seen.
4. Conclusion The Freundlich and Dubinin–Radushkevic adsorption models were used for the mathematical description of the adsorption equilibrium of 2,4-D onto DAS depending on temperature, and isotherm constants evaluated from the isotherms were used to compare the adsorptive capacity of the DAS. The adsorption of 2,4-D onto DAS was found to exhibit nonlinear favorable adsorption behavior that could be characterized well by the Freundlich and Dubinin– Radushkevic isotherm models in the studied concentration range at all the temperatures studied.
32
G. Akçay et al. / Journal of Colloid and Interface Science 281 (2005) 27–32
The magnitude of E is useful for estimating the type of adsorption process. The magnitude of sorption energy may give an idea about the type of sorption. Two main types of adsorption may occur: physical and chemical adsorption. In physical adsorption equilibrium is usually rapidly attained and easily reversible, because the energy requirements are small (usually no more than 4.2 kJ/mol), since the forces involved in the physical adsorption are weak. The chemical adsorption is specific and involves forces much stronger than those in physical adsorption. So the sorption energy for chemical adsorption is of the same magnitude as the head of chemical reactions (8.4 and 83.7 kJ/mol) [22]. Thermodynamic constants were also evaluated using equilibrium constants changing with temperature. The negative values of g a , ha , and s a showed the exothermic nature. In addition, the values are indicative of the adsorption, both physical and chemical. The following conclusions can be drawn from this study: (a) By analyzing the qm values it can be seen that affinity of adsorbent to this sample is high. (b) By analyzing differential adsorption enthalpies it is seen that organophilic sepiolite is a good adsorbent for this sample. (c) By analyzing KF and n values it can be said that organophilic sepiolite is a good adsorbent for removing organic pollutant. (d) By analyzing sorption energy (E) values it can be seen that DAS is a good adsorbent for this compound. (e) It is seen that DAS as a adsorbent is very convenient for 2,4-D at low and high temperatures. By using these kinetic and thermodynamic parameters, we believe that sorption by DAS can be suitable for the purification of waste water and the removal of 2,4-D from industrial waste waters.
Acknowledgment The authors express their thanks to TU Chemnitz, Germany, for XRD and BET surface area results.
References [1] A. Alvarez, A. Singer, E. Galan, Development in Sedimentology, vol. 37, Elsevier, Amsterdam, 1984, p. 253. [2] E. Gonzalez-Pradas, M. Villafranca-Sanchez, A. Gallego Campo, D. Urena-Amate, M. Fernandez-Perez, J. Chem. Technol. Biotechnol. 74 (1999) 49. [3] E. Gonzalez-Pradas, M. Villafranca-Sanchez, A. Gallego Campo, D. Urena-Amate, M. Fernandez-Perez, J. Chem. Technol. Biotechnol. 56 (1993) 67. [4] G. Sheng, S. Xu, S.A. Boyd, Water Res. 30 (1996) 1483. [5] L. Zhu, Y. Li, J. Zhang, Environ. Sci. Technol. 31 (1997) 1407. [6] A. Farba, R. Dufford, A. Evangelista de Dufford, Bull. Environ. Contam. Toxicol. 59 (1997) 645. [7] R. Celis, M.C. Hermosin, L. Cox, J. Cornejo, Environ. Sci. Technol. 33 (1999) 1200. [8] M. Mahramanlıo˘glu, I. Kızılcıklı, I.O. Biçer, M. Tunçay, J. Environ. Sci. Health B 35 (2000) 187. [9] J.L. Sotelo, G. Ovejero, A. Delgado, I. Martinez, Water. Res. 36 (2001) 599. [10] P. Benoit, E. Barriuso, R. Calvet, Chemosphere 37 (1998) 1271. [11] O.R. Pal, A.K. Vanjara, Sep. Purif. Technol. 24 (2001) 167. [12] E. Gonzalez-Pradas, M. Socias-Viciara, M. Sufi, M.D. Urena-Amate, F. Flores-Cespedes, M. Fernandez-Perez, M. Villafranca-Sanches, Chemosphere 5 (2003) 85. [13] J.L. Ahlrichs, C. Serna, J.M. Serratosa, Clays Clay Miner. 23 (1975) 119. [14] M.N. Fernandez-Hernandez, E. Ruiz-Hitzky, Clay Miner. 14 (1979) 295. [15] E. Ruiz-Hitzky, J. Mater. Chem. 11 (2001) 86. [16] B. Casal, J. Merino, J.M. Serratosa, E. Ruiz-Hitzky, Appl. Clay Sci. 18 (2001) 245. [17] T.J. Dines, L.D. MacGregor, C.H. Rochester, J. Colloid Interface Sci. 245 (2002) 221. [18] E. Sabah, M.S. Çelik, J. Colloid Interface Sci. 251 (2002) 33. [19] M. Akçay, J. Mol. Struct. 694 (2004) 21. [20] F.L. Slejko, Adsorption Technology Step by Step, Dekker, New York, 1985, p. 13. [21] S.H. Lin, R.S. Juang, J. Hazard. Mater. 92 (2002) 315. [22] C. Tohsieh, H. Teng, J. Colloid Interface Sci. 230 (2000) 171. [23] E. Haribabu, Y.D. Upadhya, S.N. Upadhyay, J. Environ. Stud. 43 (1993) 169. [24] F.R. Sperstein, A.L. Myers, AIChE J. 47 (2001) 1141. [25] A.L. Myers, AIChE J. 48 (2002) 145. [26] S.A. Morton III, D.J. Kefer, R.M. Counce, D.W. DePaoli, M.Z.C. Hu, J. Colloid Interface Sci. 270 (2004) 229. [27] A.R. Cestari, F.S. Eunice Vieira, A. Simoni Jose de, C. Airoldi, Thermochim. Acta 384 (2000) 25. [28] M.G. Fonseca, A.S. Jose, C. Airoldi, Thermochim. Acta 369 (2001) 17. [29] M.G. Fonseca, C. Airoldi, J. Colloid Interface Sci. 240 (2001) 229.