- Email: [email protected]
Sorption equilibria of chlorinated anilines in aqueous solution on resin-bound cobalt ion
Colloids and Surfaces A: Physicochemical and Engineering Aspects 207 (2002) 41 – 47 www.elsevier.com/locate/colsurfa
Sorption equilibria of chlorinated anilines in aqueous solution on resin-bound cobalt ion Mustafa Uc¸an, Ahmet Ayar * Department of Chemistry, Faculty of Sciences and Arts, Uni6ersity of Nig˘de, Nig˘de, Turkey Received 11 October 2001; accepted 16 January 2002
Abstract Studies have been made of the sorption equilibria of chlorinated anilines in aqueous solution on ligand exchange resin. The chlorinated anilines used included 2-chloroaniline, 3-chloroaniline, 4-chloroaniline and 2,5-dichloroaniline. A mini-column apparatus was used to study sorption of chlorinated anilines onto ligand exchange resin. The experiments were conducted in a constant temperature at 25 9 0.1 °C. The sorption behaviour of these chlorinated anilines on Co(II)-loaded carboxylated diaminoethyl sporopollenin (CDAE-sporopollenin) can be expressed by the Langmuir and Freundlich isotherms. The characteristics of the sorption process were also investigated using Scatchard plot analysis (q/C vs. q). When the Scatchard plot showed a deviation from linearity, greater emphasis was placed on the analysis of the adsorption data in terms of the Freundlich model, in order to construct the adsorption isotherms of the ligand(s) at particular concentration(s) in solutions. Equilibrium binding data for ligands gave rise to a linear plot, indicating that the Langmuir model could be applied. Ligand adsorption constants, dissociation constant (Kd) and correlation coefficients for the ligands were calculated from Langmuir and Freundlich isotherms. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Breakthrough; Ligand exchange; Sorption; Adsorption; Sporopollenin; Aromatic amines
1. Introduction Aromatic amines such as aniline and its substituted analogue are widely used as industrial intermediates in the manufacture of carbamate and urethane pesticides, dyestuffs, cosmetics and medicines. These amines are also employed in the rubber industry as antioxidants and antiozonants and as components in epoxy and polyurethane polymers [1]. Aromatic amines are in general harmful to living things. They are known to be * Corresponding author E-mail address: [email protected] (A. Ayar).
toxic water pollutants and their presence in wastewater even in very low concentrations has been shown to be harmful to aquatic life [2]. Aromatic amines are a source of serious social and hygienic problems as important occupational and environmental pollutants. Chlorinated anilines such as p-chloroaniline and 3,4-dichloroaniline were also found as degradation products and intermediates of various phenylurea and phenylcarbamate pesticides [3]. Thus, the removal and recovery of aromatic amines from wastewater are important problems, due both to the difficulty in treating them by the usual activated sludge method and to the necessity of reusing of rela-
0927-7757/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 7 - 7 7 5 7 ( 0 2 ) 0 0 1 3 4 - 6
42
M. Uc¸ an, A. Ayar / Colloids and Surfaces A: Physicochem. Eng. Aspects 207 (2002) 41–47
tively expensive chemicals. In view of the importance of these compounds, a rapid and sensitive method of analysis is needed to detect them in the environment. We therefore considered the possibility of using a Co(II)-CDAE-sporopollenin resin for removal of aromatic amines from wastewater by ligand exchange sorption. Although ligand exchange [4] is being increasingly employed for the chromatographic separation and purification of organic compounds, the possibility of its application to wastewater treatment and pollution control has not received any significant attention. A factor, which might have contributed to this neglect, is the problem associated with the reactivation of depleted adsorbent and the separation of the sorbet species from the complexing metal ion in the eluate. Another reason may be the unfavourable economics of using costlier transition metals, such as copper, nickel and cobalt, as complexing ions in a ligand exchange process for any large-scale application in wastewater treatment. These facts suggested the possibility of a new process for wastewater treatment based on ligand exchange. Ligand exchange using chelating resinbound cobalt ion was thus found by us to be very effective for the removal and recovery of aromatic amines from wastewater in low concentrations [2]. This process was termed ligand exchange by Helf-
ferich who first demonstrated this phenomenon by using a nickel-loaded carboxylic acid exchanger for replacement of amines by elution with ammonia. Ligand exchange is a highly selective process and even very similar ligands, under a proper set of conditions, may exhibit differences in the degree of formation of their metal-ligand complexes [5]. Sorption methods offer an effective removal and separation of organic weak base from aqueous media, and have been studied extensively on several adsorbents, e.g. porous resins and ligand exchanger resins. However, it is still important to conduct a systematic study of these sorption equilibria of aromatic amines to explore an adsorbent that provides selective separation. This study was conducted to confirm the possibilities of ligand sorption of chlorinated anilines from aqueous media on chelating CDAE-sporopollenin [6–8] resin. Sporopollenin is a natural polymer obtained from Lycopodium cla6atum, which is highly resistant to chemicals, has a high capacity, is stable, has a constant chemical structure, and occurs naturally as a component of spore walls and exhibits very good stability after even prolonged exposure to mineral acids and alkalies [9,10]. Ethylenediamine complexes posses a very stable structure and have a very minor dissociation tendency, and they have suitable functional groups for a Ligand-exchange matrix [12]. Steps of complex formations were as follows:
M. Uc¸ an, A. Ayar / Colloids and Surfaces A: Physicochem. Eng. Aspects 207 (2002) 41–47
The sorption characteristics were studied in terms of sorption isotherms in a column operation. Polymeric chelating resins, which can selectively remove target contaminants and be regenerated efficiently, are highly desirable for large-scale commercial applications [13]. In this regard, sorption of chlorinated anilines has been investigated on the Co(II)-CDAE-sporopollenin. The chlorinated anilines used included 2chloroaniline, 3-chloroaniline, 4-chloroaniline and 2,5-dichloroaniline. Sporopollenin is a natural polymer [9–11], obtained from L. cla6atum, can be a suitable skeleton for the ligand-exchange resin structures and occurs naturally as a component of spore walls. CDAE-sporopollenin was prepared from L. cla6atum, 1,2-diaminoethane, and bromoacetic acid. Ethylenediamine complex possesses a very stable structure and has a very minor dissociation tendency, and they act as suitable functional groups for a ligand exchange matrix.
2. Experimental
2.1. Materials L. cla6atum was purchased from BDH Chemicals. The chlorinated anilines were purchased from Sigma. All other chemicals were purchased from Merck. The ligand exchanger resin was prepared in this laboratory as previously described [6]. The chlorinated anilines were dissolved in deionized water and adjusted to a desired concentration. Laboratory reagent grade chlorinated anilines were used for the sorption studies. The purity of the products was checked by chemical analysis. All other chemicals used were reagent grade. Prior to sorption experiment, the resin was washed by a column operation.
2.2. Sorption experiment Column of 4 mm i.d. by 15 mm lengths was utilized. The adsorbent particles were packed between two layers of glass wool. To avoid air bubbles, the column was carefully packed under
43
water. The solution containing ligands was passed through the ligand-exchange column, which was filled with the Co(II)-CDAE-sporopollenin in the complexing metal-ligand form. The performance of the ligand exchanger in continuous operation was studied by conducting column runs. The flow of the solution was started at time t= 0 and samples of the effluent were recorded by a spectrometer (UV-160 A Shimadzu) at wavelength 260 nm. The amount of sorbed ligand was calculated from the breakthrough curves. The sorption of ligand in the resin was also controlled by the difference between the initial ligand concentrations in the solution and in the resin. The experiments were conducted in a constant temperature at 2590.1 °C.
3. Results and discussion From the experimental data obtained from the sorption experiments with the Co(II)-CDAEsporopollenin, the sorption isotherms reflecting the equilibrium capacity of the adsorbent, at a particular concentration of ligand in solution, were constructed. The maximum capacity (qm) and the dissociation constant (Kd) were extracted from the corresponding semi-reciprocal plots (C/q vs. C). The characteristics of the sorption process were also investigated using Scatchard plot analysis (q/C vs. q). When the Scatchard plot showed a deviation from linearity, greater emphasis was placed on the analysis of the adsorption data in terms of the Freundlich model, in order to construct the adsorption isotherms of the ligand(s) at particular concentration(s) in solutions. Fig. 1 shows the adsorption isotherms of ligands with ligand exchanger, whilst Fig. 2 presents the adsorption characteristics assessed from the Scatchard plot. Equilibrium binding data for ligands gave rise to a linear plot, indicating that the Langmuir model could be applied for sorption process [14]. In the sorptions of ligands, deviation from linearty in the plot of q/C vs. q was observed, indicating the presence of multi-model interaction and non-Langmuirean behaviour (Fig. 2). At least three types of interactions can occur during the adsorption process, namely: (i) specific
44
M. Uc¸ an, A. Ayar / Colloids and Surfaces A: Physicochem. Eng. Aspects 207 (2002) 41–47
Fig. 1. Adsorption isotherms for the equilibrium binding of ligand onto resin.
binding involving inductive effects of Cl atoms on the ligands; (ii) non-specific binding between the ligand and other classes of binding sites on the adsorbent; and (iii) ligand-mobile phase interaction. The corresponding semi-reciprocal transformations (Fig. 3) of the equilibrium binding data for 2-chloroaniline, 4-chloroaniline and 2,5dichloroaniline onto ligand exchanger gave rise to a linear plot, indicating that the Langmuir model could be applied in this case. The resulting Scatchard plot analysis indicated that the adsorp-
tion of ligands onto ligand exchanger might also diverge from the Langmuirean behaviour (Fig. 2). Although the Langmuir isothermal model was initially conceived to describe the adsorption of gases to glass and mica surfaces, where equilibrium is quickly established, this adsorption model has also found relatively wide application as a convenient representation of the simplest cases of the interaction of a ligand with a metal complex, immobilised onto a support material. Various practical examples of the use of the Langmuir isothermal model can be found in the scientific literature, which include the sorption of aromatic amines on resin-bound ferrous ion [2], sorption of pyridine derivatives on porous resins [15] and ion exchange equilibrium of heavy metals on chelating resins [16]. When the ligand is relatively small and the immobilised metal complex is readily accessible, the overall dynamics of the adsorption can be fast and the assumption of local equilibrium may be valid. The Langmuir model represents a simplified case of ligand adsorption, since this model assumes: (i) reversible adsorption; (ii) no change in the properties of the adsorbed molecules; (iii) no lateral interaction between adsorbed molecules; (iv) one adsorption site per molecule; and (v) that all adsorption sites have the same affinity for the
Fig. 2. Scatchard plot of the experimental data for the adsorption isotherms.
M. Uc¸ an, A. Ayar / Colloids and Surfaces A: Physicochem. Eng. Aspects 207 (2002) 41–47
45
where M is metal ion, L is ligand, m is coordinative valence of polymer-bound metal ion for ligand, the forward (k1) and reversed (k2) interaction rates take into account the mass transfer resistances, which include the ligand movement from the mobile phase to the adsorbent surface layer, the ligand transfer across a stagnant film layer surrounding the adsorbent particles, the ligand diffusion into the pores of the particles and the interaction and adsorption of ligand to the solid phase. Eq. (1) can be expressed in the form of a rate equation with a second-order forward and first-order reverse kinetics, namely: dq = k1C(qm − q)− k2q* dt
(2)
where C is the concentration of the ligand in the mobile phase, q is the amount of ligand adsorbed onto the matrix and qm is the maximum capacity of the matrix. At equilibrium, the rate of forward interaction becomes equal to the rate of reverse interaction (denoted by the asterisk), dq/dt becomes zero and Eq. (2) can be converted to: C/q=
Fig. 3. Semi-reciprocal plot of the experimental data for the adsorption of ligands onto resin.
ligand. Under such assumptions, the binding of a ligand onto an immobilised metal complex can be represented [2,14] by the following equilibrium: k1
MLm − 1 + L X MLm m =1, 2, 3, . . .,
(1)
Kd 1 +C qm qm
(3)
where Kd represents the dissociation constant. The equilibrium sorption data on Co2 + forms of the CDAE-sporopollenin are plotted according to Eq. (3) in Fig. 3. The values of qm and Kd derived from these plots are presented in Table 1. To compare the adsorption behaviours of ligands onto support material, the equilibrium adsorption curves for the binding of ligands onto support material are shown in Fig. 1. As evident from these data, the adsorption isotherms of 3chloroaniline and 2,5-dichloroaniline were steeper
k2
Table 1 Parameters of Langmuir and Freundlich isotherms for adsorption of ligands onto resin Ligands
2-Chloroaniline 3-Chloroaniline 4-Chloroaniline 2,5-Dichloroaniline
Langmuir isotherm (Eq. (3))
Freundlich isotherm (Eq. (4))
qm (mmol g−1 resin)
Kd (M)
r2
k (mmol g−1 resin)
n
r2
0.39 2.57 0.29 1.23
0.005 0.019 0.007 0.007
0.750 0.189 0.829 0.800
18.17 116.61 13.01 37.21
0.83 0.99 0.86 0.81
0.935 0.964 0.975 0.993
46
M. Uc¸ an, A. Ayar / Colloids and Surfaces A: Physicochem. Eng. Aspects 207 (2002) 41–47
kind of binding site involved in the adsorption of 3-chloroaniline onto support material, and indicative of more complex binding phenomena than can be explained in terms of the independent and univalent binding site Langmuirean model. The Freundlich equation is the empirical relationship whereby it is assumed that the adsorption energy of a ligand binding to a site on an adsorbent depends on whether or not the adjacent sites are already occupied. To test the fit of data, the Freundlich isotherms equation is written as q= k(C)n
Fig. 4. Log – log plot of the experimental data for the adsorption of ligands onto resin.
than the corresponding isotherm for other ligands, indicating a greater affinity of 3-chloroaniline and 2,5-dichloroaniline onto Co(II) matrix. The corresponding semi-reciprocal transformations of the equilibrium binding data for 2chloroaniline, 4-chloroaniline and 2,5-dichloroaniline onto ligand exchanger gave rise to a linear plot. However, for 3-chloroaniline, divergence from the linear semi-reciprocal fit was again evident, consistent with the participation of secondary equilibrium effects in the adsorption process (Fig. 3). The binding of 3-chloroaniline could not be as adequately represented by the Langmuir model, but rather followed more closely the Freundlich model (Fig. 4(a)). As noted above, possible reasons for this difference may be associated with the presence of more than one
(4)
where k and n are empirical coefficients. Equilibrium binding parameters were derived from Langmuir or Freundlich-type analysis by regression analysis of the experimental data for ligands with the adsorbent. In the case of Langmuir-type fit of the experimental data, the semireciprocal plot of C/q vs. C was employed to generate the intercept of Kd/qm and the slope of 1/qm. In the case of the Freundlich-type fit of the experimental data, the plot of ln C vs. ln q was employed to generate the intercept value of ln k and the slope of n. Ligand adsorption constants, dissociation constant (Kd) and correlation coefficients for the ligands were calculated from Langmuir and Freundlich isotherms and are given in Table 1. The small decrease in qm for the binding of 2-chloroanilines and 4-chloroanilines onto the Co2 + matrix leads to the conclusion that a mixture of electrical forces and steric hindrance are involved during the binding process with these ligands. The combination of forces involved during the adsorption process with 2-chloroanilines and 4-chloroanilines is not surprising, taking into account the inductive effects of ortho-Cl and para-Cl atoms of these ligands and the ability of them to acts as multi-functional ligands that contain both steric hindrance and charged groups. Once more, steric hindrance around the amino nitrogen (ortho-substitution) loosens the attachment to the metal ions and causes the substance to migrate faster [17]. The results of this investigation have provided further insight into the nature of the adsorption process of ligands with sporopollenin adsorbent.
M. Uc¸ an, A. Ayar / Colloids and Surfaces A: Physicochem. Eng. Aspects 207 (2002) 41–47
The validity of two isothermal models to appropriately anticipate the experimental findings was also examined. Since the application of methods of isothermal analysis can be used to yield useful information about the maximum capacity and the dissociation constants of a specified ligand, equilibrium binding approaches provide a relatively straight forward procedure for acquiring essential data, examining variations in the isothermal behaviour of ligands with adsorbent.
References [1] K. Kijima, H. Kataoka, M. Makita, J. Chromatogr. A 738 (1996) 83. [2] M. Chanda, O’Driscoll, G.L. Rempel, Reactive Polymers, 2 (1984) 279. [3] L. Mu¨ ller, E. Fattore, E. Benfenati, J. Chromatogr. A 791 (1997) 221. [4] F. Helfferich, Nature (London) 189 (1961) 1001.
47
[5] F. Helfferich, Ion Exchange, McGraw-Hill, New York, 1962. [6] A. Ayar, S. Yildiz, E. Pehlivan, Sep. Sci. Technol. 30 (15) (1995) 3081. [7] M. Erso¨ z, U.S. Vural, M. Yigitoglu, M. Sezgin, J. Coll. Inter. Sci. 184 (1996) 319. [8] M. Erso¨ z, M. Yig˘ itog˘ lu, A. Ayar, J. Appl. Polym. Sci. 64 (1997) 1225. [9] G. Shaw, in: J.B. Harbone (Ed.), Phyto Chemical Phylogency, Academic Press, London, 1970. [10] J. Brooks, G. Shaw, Nature 220 (1968) 678. [11] G. Mackenzie, G. Shaw, Int. J. Pep. Prot. Res. 15 (1980) 298. [12] S. Yildiz, E. Pehlivan, M. Erso¨ z, M. Pehlivan, J. Chromatogr. Sci. 31 (1993) 150. [13] U.S. Vural, M. Erso¨ z, E. Pehlivan, J. Appl. Polym. Sci. 58 (1995) 2423. [14] G.M.S. Finette, Q. Mao, M.T.W. Hearn, J. Chromatogr. A 763 (1997) 71. [15] S. Akita, H. Takeuchi, J. Chem. Eng. Jpn. 26 (3) (1993) 237. [16] M. Erso¨ z, E. Pehlivan, H.J. Duncan, S. Yildiz, M. Pehlivan, Reactive Polym. 24 (1995) 195. [17] V.A. Davankov, J.D. Navratil, H.F. Walton, Ligand Exchange Chromatography, CRC Press, US, 1988.
Sorption equilibria of chlorinated anilines in aqueous solution on resin-bound cobalt ion Mustafa Uc¸an, Ahmet Ayar * Department of Chemistry, Faculty of Sciences and Arts, Uni6ersity of Nig˘de, Nig˘de, Turkey Received 11 October 2001; accepted 16 January 2002
Abstract Studies have been made of the sorption equilibria of chlorinated anilines in aqueous solution on ligand exchange resin. The chlorinated anilines used included 2-chloroaniline, 3-chloroaniline, 4-chloroaniline and 2,5-dichloroaniline. A mini-column apparatus was used to study sorption of chlorinated anilines onto ligand exchange resin. The experiments were conducted in a constant temperature at 25 9 0.1 °C. The sorption behaviour of these chlorinated anilines on Co(II)-loaded carboxylated diaminoethyl sporopollenin (CDAE-sporopollenin) can be expressed by the Langmuir and Freundlich isotherms. The characteristics of the sorption process were also investigated using Scatchard plot analysis (q/C vs. q). When the Scatchard plot showed a deviation from linearity, greater emphasis was placed on the analysis of the adsorption data in terms of the Freundlich model, in order to construct the adsorption isotherms of the ligand(s) at particular concentration(s) in solutions. Equilibrium binding data for ligands gave rise to a linear plot, indicating that the Langmuir model could be applied. Ligand adsorption constants, dissociation constant (Kd) and correlation coefficients for the ligands were calculated from Langmuir and Freundlich isotherms. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Breakthrough; Ligand exchange; Sorption; Adsorption; Sporopollenin; Aromatic amines
1. Introduction Aromatic amines such as aniline and its substituted analogue are widely used as industrial intermediates in the manufacture of carbamate and urethane pesticides, dyestuffs, cosmetics and medicines. These amines are also employed in the rubber industry as antioxidants and antiozonants and as components in epoxy and polyurethane polymers [1]. Aromatic amines are in general harmful to living things. They are known to be * Corresponding author E-mail address: [email protected] (A. Ayar).
toxic water pollutants and their presence in wastewater even in very low concentrations has been shown to be harmful to aquatic life [2]. Aromatic amines are a source of serious social and hygienic problems as important occupational and environmental pollutants. Chlorinated anilines such as p-chloroaniline and 3,4-dichloroaniline were also found as degradation products and intermediates of various phenylurea and phenylcarbamate pesticides [3]. Thus, the removal and recovery of aromatic amines from wastewater are important problems, due both to the difficulty in treating them by the usual activated sludge method and to the necessity of reusing of rela-
0927-7757/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 7 - 7 7 5 7 ( 0 2 ) 0 0 1 3 4 - 6
42
M. Uc¸ an, A. Ayar / Colloids and Surfaces A: Physicochem. Eng. Aspects 207 (2002) 41–47
tively expensive chemicals. In view of the importance of these compounds, a rapid and sensitive method of analysis is needed to detect them in the environment. We therefore considered the possibility of using a Co(II)-CDAE-sporopollenin resin for removal of aromatic amines from wastewater by ligand exchange sorption. Although ligand exchange [4] is being increasingly employed for the chromatographic separation and purification of organic compounds, the possibility of its application to wastewater treatment and pollution control has not received any significant attention. A factor, which might have contributed to this neglect, is the problem associated with the reactivation of depleted adsorbent and the separation of the sorbet species from the complexing metal ion in the eluate. Another reason may be the unfavourable economics of using costlier transition metals, such as copper, nickel and cobalt, as complexing ions in a ligand exchange process for any large-scale application in wastewater treatment. These facts suggested the possibility of a new process for wastewater treatment based on ligand exchange. Ligand exchange using chelating resinbound cobalt ion was thus found by us to be very effective for the removal and recovery of aromatic amines from wastewater in low concentrations [2]. This process was termed ligand exchange by Helf-
ferich who first demonstrated this phenomenon by using a nickel-loaded carboxylic acid exchanger for replacement of amines by elution with ammonia. Ligand exchange is a highly selective process and even very similar ligands, under a proper set of conditions, may exhibit differences in the degree of formation of their metal-ligand complexes [5]. Sorption methods offer an effective removal and separation of organic weak base from aqueous media, and have been studied extensively on several adsorbents, e.g. porous resins and ligand exchanger resins. However, it is still important to conduct a systematic study of these sorption equilibria of aromatic amines to explore an adsorbent that provides selective separation. This study was conducted to confirm the possibilities of ligand sorption of chlorinated anilines from aqueous media on chelating CDAE-sporopollenin [6–8] resin. Sporopollenin is a natural polymer obtained from Lycopodium cla6atum, which is highly resistant to chemicals, has a high capacity, is stable, has a constant chemical structure, and occurs naturally as a component of spore walls and exhibits very good stability after even prolonged exposure to mineral acids and alkalies [9,10]. Ethylenediamine complexes posses a very stable structure and have a very minor dissociation tendency, and they have suitable functional groups for a Ligand-exchange matrix [12]. Steps of complex formations were as follows:
M. Uc¸ an, A. Ayar / Colloids and Surfaces A: Physicochem. Eng. Aspects 207 (2002) 41–47
The sorption characteristics were studied in terms of sorption isotherms in a column operation. Polymeric chelating resins, which can selectively remove target contaminants and be regenerated efficiently, are highly desirable for large-scale commercial applications [13]. In this regard, sorption of chlorinated anilines has been investigated on the Co(II)-CDAE-sporopollenin. The chlorinated anilines used included 2chloroaniline, 3-chloroaniline, 4-chloroaniline and 2,5-dichloroaniline. Sporopollenin is a natural polymer [9–11], obtained from L. cla6atum, can be a suitable skeleton for the ligand-exchange resin structures and occurs naturally as a component of spore walls. CDAE-sporopollenin was prepared from L. cla6atum, 1,2-diaminoethane, and bromoacetic acid. Ethylenediamine complex possesses a very stable structure and has a very minor dissociation tendency, and they act as suitable functional groups for a ligand exchange matrix.
2. Experimental
2.1. Materials L. cla6atum was purchased from BDH Chemicals. The chlorinated anilines were purchased from Sigma. All other chemicals were purchased from Merck. The ligand exchanger resin was prepared in this laboratory as previously described [6]. The chlorinated anilines were dissolved in deionized water and adjusted to a desired concentration. Laboratory reagent grade chlorinated anilines were used for the sorption studies. The purity of the products was checked by chemical analysis. All other chemicals used were reagent grade. Prior to sorption experiment, the resin was washed by a column operation.
2.2. Sorption experiment Column of 4 mm i.d. by 15 mm lengths was utilized. The adsorbent particles were packed between two layers of glass wool. To avoid air bubbles, the column was carefully packed under
43
water. The solution containing ligands was passed through the ligand-exchange column, which was filled with the Co(II)-CDAE-sporopollenin in the complexing metal-ligand form. The performance of the ligand exchanger in continuous operation was studied by conducting column runs. The flow of the solution was started at time t= 0 and samples of the effluent were recorded by a spectrometer (UV-160 A Shimadzu) at wavelength 260 nm. The amount of sorbed ligand was calculated from the breakthrough curves. The sorption of ligand in the resin was also controlled by the difference between the initial ligand concentrations in the solution and in the resin. The experiments were conducted in a constant temperature at 2590.1 °C.
3. Results and discussion From the experimental data obtained from the sorption experiments with the Co(II)-CDAEsporopollenin, the sorption isotherms reflecting the equilibrium capacity of the adsorbent, at a particular concentration of ligand in solution, were constructed. The maximum capacity (qm) and the dissociation constant (Kd) were extracted from the corresponding semi-reciprocal plots (C/q vs. C). The characteristics of the sorption process were also investigated using Scatchard plot analysis (q/C vs. q). When the Scatchard plot showed a deviation from linearity, greater emphasis was placed on the analysis of the adsorption data in terms of the Freundlich model, in order to construct the adsorption isotherms of the ligand(s) at particular concentration(s) in solutions. Fig. 1 shows the adsorption isotherms of ligands with ligand exchanger, whilst Fig. 2 presents the adsorption characteristics assessed from the Scatchard plot. Equilibrium binding data for ligands gave rise to a linear plot, indicating that the Langmuir model could be applied for sorption process [14]. In the sorptions of ligands, deviation from linearty in the plot of q/C vs. q was observed, indicating the presence of multi-model interaction and non-Langmuirean behaviour (Fig. 2). At least three types of interactions can occur during the adsorption process, namely: (i) specific
44
M. Uc¸ an, A. Ayar / Colloids and Surfaces A: Physicochem. Eng. Aspects 207 (2002) 41–47
Fig. 1. Adsorption isotherms for the equilibrium binding of ligand onto resin.
binding involving inductive effects of Cl atoms on the ligands; (ii) non-specific binding between the ligand and other classes of binding sites on the adsorbent; and (iii) ligand-mobile phase interaction. The corresponding semi-reciprocal transformations (Fig. 3) of the equilibrium binding data for 2-chloroaniline, 4-chloroaniline and 2,5dichloroaniline onto ligand exchanger gave rise to a linear plot, indicating that the Langmuir model could be applied in this case. The resulting Scatchard plot analysis indicated that the adsorp-
tion of ligands onto ligand exchanger might also diverge from the Langmuirean behaviour (Fig. 2). Although the Langmuir isothermal model was initially conceived to describe the adsorption of gases to glass and mica surfaces, where equilibrium is quickly established, this adsorption model has also found relatively wide application as a convenient representation of the simplest cases of the interaction of a ligand with a metal complex, immobilised onto a support material. Various practical examples of the use of the Langmuir isothermal model can be found in the scientific literature, which include the sorption of aromatic amines on resin-bound ferrous ion [2], sorption of pyridine derivatives on porous resins [15] and ion exchange equilibrium of heavy metals on chelating resins [16]. When the ligand is relatively small and the immobilised metal complex is readily accessible, the overall dynamics of the adsorption can be fast and the assumption of local equilibrium may be valid. The Langmuir model represents a simplified case of ligand adsorption, since this model assumes: (i) reversible adsorption; (ii) no change in the properties of the adsorbed molecules; (iii) no lateral interaction between adsorbed molecules; (iv) one adsorption site per molecule; and (v) that all adsorption sites have the same affinity for the
Fig. 2. Scatchard plot of the experimental data for the adsorption isotherms.
M. Uc¸ an, A. Ayar / Colloids and Surfaces A: Physicochem. Eng. Aspects 207 (2002) 41–47
45
where M is metal ion, L is ligand, m is coordinative valence of polymer-bound metal ion for ligand, the forward (k1) and reversed (k2) interaction rates take into account the mass transfer resistances, which include the ligand movement from the mobile phase to the adsorbent surface layer, the ligand transfer across a stagnant film layer surrounding the adsorbent particles, the ligand diffusion into the pores of the particles and the interaction and adsorption of ligand to the solid phase. Eq. (1) can be expressed in the form of a rate equation with a second-order forward and first-order reverse kinetics, namely: dq = k1C(qm − q)− k2q* dt
(2)
where C is the concentration of the ligand in the mobile phase, q is the amount of ligand adsorbed onto the matrix and qm is the maximum capacity of the matrix. At equilibrium, the rate of forward interaction becomes equal to the rate of reverse interaction (denoted by the asterisk), dq/dt becomes zero and Eq. (2) can be converted to: C/q=
Fig. 3. Semi-reciprocal plot of the experimental data for the adsorption of ligands onto resin.
ligand. Under such assumptions, the binding of a ligand onto an immobilised metal complex can be represented [2,14] by the following equilibrium: k1
MLm − 1 + L X MLm m =1, 2, 3, . . .,
(1)
Kd 1 +C qm qm
(3)
where Kd represents the dissociation constant. The equilibrium sorption data on Co2 + forms of the CDAE-sporopollenin are plotted according to Eq. (3) in Fig. 3. The values of qm and Kd derived from these plots are presented in Table 1. To compare the adsorption behaviours of ligands onto support material, the equilibrium adsorption curves for the binding of ligands onto support material are shown in Fig. 1. As evident from these data, the adsorption isotherms of 3chloroaniline and 2,5-dichloroaniline were steeper
k2
Table 1 Parameters of Langmuir and Freundlich isotherms for adsorption of ligands onto resin Ligands
2-Chloroaniline 3-Chloroaniline 4-Chloroaniline 2,5-Dichloroaniline
Langmuir isotherm (Eq. (3))
Freundlich isotherm (Eq. (4))
qm (mmol g−1 resin)
Kd (M)
r2
k (mmol g−1 resin)
n
r2
0.39 2.57 0.29 1.23
0.005 0.019 0.007 0.007
0.750 0.189 0.829 0.800
18.17 116.61 13.01 37.21
0.83 0.99 0.86 0.81
0.935 0.964 0.975 0.993
46
M. Uc¸ an, A. Ayar / Colloids and Surfaces A: Physicochem. Eng. Aspects 207 (2002) 41–47
kind of binding site involved in the adsorption of 3-chloroaniline onto support material, and indicative of more complex binding phenomena than can be explained in terms of the independent and univalent binding site Langmuirean model. The Freundlich equation is the empirical relationship whereby it is assumed that the adsorption energy of a ligand binding to a site on an adsorbent depends on whether or not the adjacent sites are already occupied. To test the fit of data, the Freundlich isotherms equation is written as q= k(C)n
Fig. 4. Log – log plot of the experimental data for the adsorption of ligands onto resin.
than the corresponding isotherm for other ligands, indicating a greater affinity of 3-chloroaniline and 2,5-dichloroaniline onto Co(II) matrix. The corresponding semi-reciprocal transformations of the equilibrium binding data for 2chloroaniline, 4-chloroaniline and 2,5-dichloroaniline onto ligand exchanger gave rise to a linear plot. However, for 3-chloroaniline, divergence from the linear semi-reciprocal fit was again evident, consistent with the participation of secondary equilibrium effects in the adsorption process (Fig. 3). The binding of 3-chloroaniline could not be as adequately represented by the Langmuir model, but rather followed more closely the Freundlich model (Fig. 4(a)). As noted above, possible reasons for this difference may be associated with the presence of more than one
(4)
where k and n are empirical coefficients. Equilibrium binding parameters were derived from Langmuir or Freundlich-type analysis by regression analysis of the experimental data for ligands with the adsorbent. In the case of Langmuir-type fit of the experimental data, the semireciprocal plot of C/q vs. C was employed to generate the intercept of Kd/qm and the slope of 1/qm. In the case of the Freundlich-type fit of the experimental data, the plot of ln C vs. ln q was employed to generate the intercept value of ln k and the slope of n. Ligand adsorption constants, dissociation constant (Kd) and correlation coefficients for the ligands were calculated from Langmuir and Freundlich isotherms and are given in Table 1. The small decrease in qm for the binding of 2-chloroanilines and 4-chloroanilines onto the Co2 + matrix leads to the conclusion that a mixture of electrical forces and steric hindrance are involved during the binding process with these ligands. The combination of forces involved during the adsorption process with 2-chloroanilines and 4-chloroanilines is not surprising, taking into account the inductive effects of ortho-Cl and para-Cl atoms of these ligands and the ability of them to acts as multi-functional ligands that contain both steric hindrance and charged groups. Once more, steric hindrance around the amino nitrogen (ortho-substitution) loosens the attachment to the metal ions and causes the substance to migrate faster [17]. The results of this investigation have provided further insight into the nature of the adsorption process of ligands with sporopollenin adsorbent.
M. Uc¸ an, A. Ayar / Colloids and Surfaces A: Physicochem. Eng. Aspects 207 (2002) 41–47
The validity of two isothermal models to appropriately anticipate the experimental findings was also examined. Since the application of methods of isothermal analysis can be used to yield useful information about the maximum capacity and the dissociation constants of a specified ligand, equilibrium binding approaches provide a relatively straight forward procedure for acquiring essential data, examining variations in the isothermal behaviour of ligands with adsorbent.
References [1] K. Kijima, H. Kataoka, M. Makita, J. Chromatogr. A 738 (1996) 83. [2] M. Chanda, O’Driscoll, G.L. Rempel, Reactive Polymers, 2 (1984) 279. [3] L. Mu¨ ller, E. Fattore, E. Benfenati, J. Chromatogr. A 791 (1997) 221. [4] F. Helfferich, Nature (London) 189 (1961) 1001.
47
[5] F. Helfferich, Ion Exchange, McGraw-Hill, New York, 1962. [6] A. Ayar, S. Yildiz, E. Pehlivan, Sep. Sci. Technol. 30 (15) (1995) 3081. [7] M. Erso¨ z, U.S. Vural, M. Yigitoglu, M. Sezgin, J. Coll. Inter. Sci. 184 (1996) 319. [8] M. Erso¨ z, M. Yig˘ itog˘ lu, A. Ayar, J. Appl. Polym. Sci. 64 (1997) 1225. [9] G. Shaw, in: J.B. Harbone (Ed.), Phyto Chemical Phylogency, Academic Press, London, 1970. [10] J. Brooks, G. Shaw, Nature 220 (1968) 678. [11] G. Mackenzie, G. Shaw, Int. J. Pep. Prot. Res. 15 (1980) 298. [12] S. Yildiz, E. Pehlivan, M. Erso¨ z, M. Pehlivan, J. Chromatogr. Sci. 31 (1993) 150. [13] U.S. Vural, M. Erso¨ z, E. Pehlivan, J. Appl. Polym. Sci. 58 (1995) 2423. [14] G.M.S. Finette, Q. Mao, M.T.W. Hearn, J. Chromatogr. A 763 (1997) 71. [15] S. Akita, H. Takeuchi, J. Chem. Eng. Jpn. 26 (3) (1993) 237. [16] M. Erso¨ z, E. Pehlivan, H.J. Duncan, S. Yildiz, M. Pehlivan, Reactive Polym. 24 (1995) 195. [17] V.A. Davankov, J.D. Navratil, H.F. Walton, Ligand Exchange Chromatography, CRC Press, US, 1988.