Journal of Environmental Management 156 (2015) 200e208
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Modeling, kinetic, and equilibrium characterization of paraquat adsorption onto polyurethane foam using the ion-pairing technique Walkimar M. Carneiro a, Claudio F. Lima b, Jonas O. Vinhal a, b, Mateus R. Lage a, Jose Ricardo J. Cassella a, * ~o em Química, Universidade Federal Fluminense, Outeiro de Sa ~o Joa ~o Batista s/n, Centro, Nitero s Graduaça i, RJ Instituto de Química, Programa de Po 24020-141, Brazil b Departamento de Química, Universidade Federal de Viçosa, Viçosa, MG 36570-000, Brazil a
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 June 2014 Received in revised form 2 March 2015 Accepted 13 March 2015 Available online
We studied the adsorption of paraquat onto polyurethane foam (PUF) when it was in a medium containing sodium dodecylsulfate (SDS). The adsorption efficiency was dependent on the concentration of SDS in solution, because the formation of an ion-associate between the cationic paraquat and the dodecylsulfate anion was found to be a fundamental step in the process. A computational study was carried out to identify the possible structure of the ion-associate in aqueous medium. The obtained data demonstrated that the structure is probably formed from four units of dodecylsulfate bonded to one paraquat moiety. The results showed that 94% of the paraquat present in 45 mL of a solution containing 3.90 105 mol L1 could be retained by 300 mg of PUF, resulting in the removal of 2.20 mg of paraquat. The experimental data were reasonably adjusted to the Freundlich isotherm and to the pseudo-secondorder kinetic model. Also, the application of MorriseWeber and Reichenberg models indicated that both film-diffusion and intraparticle-diffusion processes were active during the control of the adsorption kinetics. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Paraquat Polyurethane foam Adsorption Ion-pairing
1. Introduction Nowadays, contamination of aquatic systems can be considered as one of the main problems in modern society, because the shortage of water is a phenomenon already observed in several parts of the world (El-Shahawi, 1997). The discharge of pesticides into water bodies is one process that leads to the contamination of aquatic systems (Brigante and Schulz, 2011). Paraquat (1,10 -dimethyl-4,40 -dipyridyn, Fig. 1) is a quaternary ammonium salt that is largely employed as a herbicide in various types of cultures, owing to its chemical and physical properties such as high solubility in water (Matolcsy et al., 1988; Bromilow, 2004). Paraquat has been proven to be toxic to mammals, directly affecting the lungs, liver, and kidneys, among other organs (Clark et al., 1966; Melchiorri et al., 1996). Despite the fact that some studies have associated Parkinson's disease with paraquat contamination, it is not possible to affirm that this herbicide is
* Corresponding author. E-mail address: [email protected] (R.J. Cassella). http://dx.doi.org/10.1016/j.jenvman.2015.03.022 0301-4797/© 2015 Elsevier Ltd. All rights reserved.
responsible for the development of the disease (Dinis-Oliveira et al., 2006; Mandel et al., 2012). Different strategies can be used to eliminate and/or remove pollutants from water. Among these strategies, adsorption processes represent one of the most important strategies, especially because of the advantages observed when they are applied. In general, adsorption processes are simple, efficient, and versatile, because the use of different adsorbents allows the removal of solutes with varying chemical characteristics. Several studies have been conducted to evaluate the removal of paraquat from water by adsorption processes. Various materials have already been used for this purpose, such as clay minerals (Seki and Yurdakoç, 2005; Tsai and Lai, 2006; Brigante et al., 2010; Iglesias et al., 2010), diatomaceous earth (Tsai et al., 2005), rice husk (Hsu et al., 2009), mesoporous silica (Brigante and Schulz, 2011), sawdust (Nanseu-Njiki et al., 2010), activated carbon (Hamadi et al., 2004), and chitosan (Silva et al., 2011). In order to choose an adsorbent for the removal of chemical species from aqueous media, several aspects must be considered. Undoubtedly, the most important aspect is the solute affinity for the adsorbent, which will determine the efficiency of the process.
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
Fig. 1. Structure of paraquat in its ionic form.
Other aspects, such as cost, availability, and the need for pretreatment, are also important. Polyurethane foams (PUFs) satisfy all of these criteria, as they are low-cost, readily available, and do not require pretreatment, and PUFs can easily be obtained in the market in different forms and compositions (Braun et al., 1985). They are able to retain polar and non-polar solutes, owing to the different functional groups present in their structure (Bowen, 1970). PUFs have already been employed, with success, in the adsorption of metallic complexes (Sant’Ana et al., 2003; Sant’Ana et al., 2004a; Sant’Ana et al., 2004b; Almeida et al., 2007), other pesticides (Farag et al., 1986; El-Shahawi, 1993; El-Shahawi and Aldhaheri, 1996; El-Shahawi, 1997; Cassella et al., 2000), dyes (Baldez et al., 2008; Robaina et al., 2009; Baldez et al., 2009; Mori and Cassella, 2009; Silveira Neta et al., 2011; Leite et al., 2012), and other organic pollutants (El-Shahawi, 1994). There is, however, no report regarding the use of a PUF for the adsorption of paraquat or any other cationic herbicide. The main goal of this work was to optimize and characterize, in terms of kinetics and the equilibrium, the adsorption of paraquat from an aqueous medium by PUF, using dodecylsulfate as the counter ion. Additionally, a computational study was carried out in order to reveal the possible species that is actually adsorbed and to provide information about the adsorption mechanism. 2. Experimental 2.1. Apparatus Determination of the paraquat concentration in the solutions was performed by spectrophotometry using an Agilent UVevisible spectrophotometer (model Cary 60, Palo Alto, CA, USA). The spectra were registered in the range of 200e400 nm, and the absorbance at 257 nm was employed for quantitative purposes. All measurements were performed with a standard quartz cuvette with an optical path length of 10 mm. Agitation of the solutions was carried out on a horizontal roller mixer by Biomixer (Curitiba, Brazil). Capped polyethylene flasks with a 50 mL capacity were employed in the experiments. The total volume of solution used in each experiment was always 45 mL. The Spartan 6.0 software (Wavefunction Inc., CA, USA) was used to design the chemical structures for the input files employed in the computational calculations (Shao et al., 2006). This software was also used to perform conformational analysis of the molecules present in each system. Finally, Gaussian 09 (Gaussian Inc., Wallingford, CT, USA) software was used to run all density functional calculations (Frisch et al., 2009) and Chemcraft was used to view and edit the Gaussian output files. 2.2. Reagents and solutions All reagents used in this work were of analytical grade or higher and were employed without further purification. The deionized water employed in the preparation of solutions was purified in a Direct-Q System (Millipore, Bedford, USA). A 1000 mg L1 (3.90 103 mol L1) paraquat stock solution was prepared by dissolving 50 mg of the reagent (SigmaeAldrich, Steinheim, Germany) in approximately 20 mL of deionized water.
201
Afterwards, the obtained solution was quantitatively transferred to a 50 mL volumetric flask and the volume was made up to the mark with deionized water. The stock solution was stored in a refrigerator. The paraquat solutions employed in the experiments were prepared through the convenient dilution of the stock solution with deionized water. A 1000 mg L1 (3.90 103 mol L1) stock solution of sodium dodecylsulfate (SDS) was prepared by dissolving 0.250 g of the reagent (Vetec, Rio de Janeiro, Brazil) in approximately 100 mL of deionized water. Then, the solution was transferred to a 250 mL volumetric flask and the volume was made up to the mark with deionized water. This solution was stable for 1 week when stored at an ambient laboratory temperature (22 ± 1 C) in a light-free location. Diluted solutions of SDS, employed in the adsorption experiments, were prepared through the convenient dilution of the stock solution with deionized water. Open-cell, polyether-type PUF (Guararapes Ltda, Brazil) with a density of 22.5 mg cm3 was employed in all experiments. The foam was comminuted in a blender with purified water. Then, the powdered PUF was dried at ambient temperature and sieved through a 2 mm plastic sieve. The treated PUF was stored in a plastic flask, which was maintained in a light- and dust-free environment. 2.3. General procedure The adsorption experiments were performed by shaking 45 mL solutions, containing known concentrations of paraquat and SDS, with adequate mass of powdered PUF. The shaking was performed with a horizontal mixer operated at 90 rpm. The concentration of paraquat that remained in the solution after shaking was determined by spectrophotometry (at 257 nm), at certain time intervals, as stipulated in each experiment according to the information desired. All experiments were carried out at ambient temperature, which was always 22 ± 1 C. The removal percentage (R) was calculated using Equation (1):
Rð%Þ ¼ 100
Ct 100 Co
(1)
in which Co is the initial concentration of paraquat in the solution and Ct is the concentration of paraquat in the solution at a given time. 3. Results and discussion 3.1. Influence of SDS concentration The most important process variable, related to the adsorption procedure, investigated in this work, is the formation of an ionassociate between the cationic paraquat and the dodecylsulfate anion, because the highly soluble paraquat cation cannot be retained by the hydrophobic PUF, as previously observed in a study on the adsorption of cationic dyes by PUFs (Baldez et al., 2008, 2009; Mori and Cassella, 2009; Leite et al., 2012). Therefore, the addition of a correct concentration of SDS to the medium was fundamental in order to enhance paraquat adsorption by PUF. This parameter was evaluated in the range of 0 (no addition of SDS) to 400 mg L1 (0e1.40 103 mol L1). The mass of PUF employed in the experiment was 200 mg, the shaking speed was 90 rpm, and the initial concentration of paraquat in solution was 5.5 mg L1 (or 2.15 105 mol L1). The adsorption efficiency increased with increasing SDS concentration up to 200 mg L1. Above this SDS concentration, the removal efficiency remained virtually constant, once the variation
202
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
of the amount of paraquat adsorbed was very low (Fig. 2). In order to ensure quantitative (higher than 90%) adsorption of the herbicide, the SDS concentration was set at 300 mg L1 (1.05 103 mol L1) in all further experiments. For each of the SDS concentrations tested, the time required to attain equilibrium was 30 min, which indicated that the SDS concentration had no significant effect on the adsorption kinetics. The addition of SDS at higher concentrations ensured the conversion of all paraquat that was present in the solution into its respective ion-associate, which is a very important step of the process, as the ion-associate is the substance that can actually be adsorbed onto the PUF. The concentration of SDS needed to achieve the highest adsorption efficiency was 49 times higher than the concentration of paraquat, which demonstrated the low value of the formation constant of the ion-associate in aqueous medium. As demonstrated, the adsorption of paraquat depended on its conversion into the ion-associate and other several factors which will be investigated below. However, the structure of the ionassociate could not be determined from the experimental data. In order to elucidate the substance that was really adsorbed onto PUF, a theoretical study was performed using molecular modeling. 3.2. Influence of PUF mass
3.3. Influence of pH and ionic strength In most liquidesolid extraction processes, the pH of the medium can play a major role because it can affect the characteristic of the adsorbent surface and/or the acid-base distribution of the solute species in aqueous medium. As acid-base chemistry of paraquat in aqueous solutions is restricted, since there are no acid-base groups in its structure, the influence of the pH on the present work could be related only to the interaction of Hþ with nitrogen atoms present in the PUF structure. The effect of pH was tested in the range of 2.0e9.0. Solutions with pH higher than 9.0 were not tested because, in this situation, the molecule of paraquat was degraded due to the cleavage of pyridine rings in basic medium (Matolcsy et al., 1988). The control of the pH of the solutions was achieved with the use of the BrittoneRobinson buffer (phosphoric acid/acetate/borate) at 0.10 mol L1 total concentration. Solutions containing paraquat and SDS at concentrations of 2.15 105 and 1.05 103 mol L1, respectively, were employed in the experiments. Maximum extraction of paraquat was observed in the range of 7.0e9.0 (Fig. 3a). In the acid range, the extraction was impaired, probably due to the protonation of nitrogen atoms present in urethane groups of the solid-phase, which resulted in the formation of
The influence of the mass of adsorbent was investigated in the range of 50e400 mg of PUF. The evaluation of the effect of this parameter is very important because it controls the number of active sites exposed for adsorption and, in general, presents noticeable effect on both extraction efficiency and kinetics. The concentrations of paraquat and SDS in the solutions employed in this experiment were 2.15 105 and 1.05 103 mol L1, respectively. The extraction efficiency increased with the increasing of the mass of adsorbent up to 300 mg, achieving a removal percentage of approximately 95%. When the mass of adsorbent was 400 mg, no improvement of the removal percentage was verified, probably due to the excess of active sites already present in the resin. Therefore, an adsorbent dose of 300 mg was used in all further experiments. No noticeable effect of the mass of adsorbent on the adsorption kinetics was observed probably because the transportation of the solute from the solution to the liquidesolid interface was efficient at the shaking conditions established in the experiment.
Fig. 2. Effect of SDS concentration on the removal of paraquat by PUF in aqueous solution. The initial concentration of paraquat was 2.15 105 mol L1 (or 5.5 mg L1).
Fig. 3. Effect of (a) pH and (b) ionic strength on the removal of paraquat by PUF in aqueous solution. The initial concentration of paraquat was 2.15 105 mol L1 (or 5.5 mg L1).
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
a positively charged PUF surface and decreased the interaction of the hydrophobic ion-associate with the resin. The influence of the ionic strength has been verified in the liquideliquid extraction of the ion-associate paraquat-SDS. According to Jarvie and Stewart (1979), the extraction was affected by the ionic strength and the influence depended on the type and concentration of the electrolyte added to the solution. In this context, the effect of the ionic strength was evaluated in the range of 0.005e0.10 mol L1 of NaCl. This experiment was performed using paraquat and SDS in concentrations of 2.15 105 and 1.05 103 mol L1, respectively. The results obtained in this experiment (Fig. 3b) revealed that the ionic strength influenced the adsorption of paraquat by PUF. It was observed a significant decrease of the herbicide adsorption onto PUF with the increase of the NaCl concentration in the medium. According to Tsai and Lai (2006) and Brigante et al. (2010), there could be a competition between the cations of the electrolyte and paraquat by the negative sites of the adsorbent. However, the adsorption mechanism proposed in this work does not take into account the direct interaction between the cationic solute and negative sites present on the adsorbent surface. Therefore, it is believed that this behavior could be probably related to the formation of a positive electrostatic “shield” around dodecylsulfate as a consequence of the increase of the concentration of cations in the solution. This phenomenon impairs the formation of the ionassociate, a fundamental step of the process. 3.4. Influence of shaking speed The shaking speed can affect directly the adsorption process. The increase of shaking speed of the system increases the average kinetic energy of the species and hence the number of effective collisions between the solutes and the active sites of the adsorbent material. So, the shaking speed was evaluated in the range of 15e90 rpm. This experiment was performed using a horizontal shaker with 300 mg of PUF. The concentrations of paraquat and SDS were 2.15 105 and 1.05 103 mol L1, respectively. The results obtained in this experiment revealed that the shaking speed did not influence the extraction efficiency of paraquat. Extraction efficiencies of approximately 95% were observed in the whole range studied. 3.5. Computational studies on ion-associate formation in water Although the extraction of paraquat in the presence of SDS has already been reported (Jarvie and Stewart, 1979; Leite et al., 2013), no work demonstrating the occurrence and/or proposing the structure of the possible ion-associate has been reported. In order to fill this gap, a computational study was carried out to model the formation of the paraquateSDS ion-associate in aqueous solution and to identify the molecular structure of the solute that is actually adsorbed. As previously mentioned, computational calculations were performed in order to improve the understanding of the interaction between paraquat and dodecylsulfate ions and to evaluate the most likely stoichiometry and structure of the ion-associate. For this purpose, calculations were performed in the gas phase and in the presence of the solvent (water). In the first step, conformational analysis was carried out using the semi-empirical method Austin Model 1 (AM1) in Spartan 6.0 code (Dewar et al., 1985; Shao et al., 2006). This conformational study was performed to show that the most likely geometry for the ions could be obtained and to make the geometry optimization easier to converge. The optimization of the geometries of paraquat, the dodecylsulfate anion, and the ionassociate, in the gas phase, was performed at the B3LYP/6-31G(d) level using Gaussian 09 software. It provided geometries that
203
were confirmed as energy minima by calculating the second-order Hessian matrix (no negative eigenvalues) (Lee et al., 1988; Miehlich et al., 1989; Ditchfield et al., 1971; Hehre et al., 1972; Frish et al., 2009). Fig. 4 shows the calculated geometries of the isolated ions (Fig. 4a and b) and for the ion-associate that resulted from the interaction of one paraquat ion with one dodecylsulfate anion (Fig. 4c). According to B3LYP/6-31G(d) calculations (Kohn and Sham, 1965; Parr and Yang, 1989; Lee et al., 1988), the formation of the ion-associate between dodecylsulfate and paraquat ions (shown in Fig. 4c) must occur through two types of interactions: (i) the interaction between one oxygen atom of the dodecylsulfate anion, with one nitrogen atom of paraquat, or (ii) the interaction between other oxygen atom of dodecylsulfate with one hydrogen atom of paraquat (hydrogen bonding). These interactions were favored by the partial charges of the atoms involved. The dodecylsulfate oxygen atoms have a negative partial charge, whereas the nitrogen atoms, and consequently the pyridine hydrogen atoms, have a positive partial charge. When more dodecylsulfate ions were added to the system, the interactions followed the same trend as those observed in the first interaction. Density functional calculations were carried out by considering isolated ions in the gas phase and by including solvation effects (water) using the polarizable continuum solvation model (Tomasi et al., 2005). The approach consisted of the addition of between one and five dodecylsulfate ions, sequentially, to a single paraquat unity. The addition of the dodecylsulfate ions was continued until five anions were included, because the stability of the system was strongly affected when a sixth dodecylsulfate ion was added. In the situation with six dodecylsulfate ions, the ion-associate decomposes, probably due to the strong repulsion caused by the excess negative charges. The geometry of each ion-associate was fully optimized, which was followed by the calculation of vibrational frequencies to confirm the geometries as true minima and to obtain the thermodynamic parameters needed to calculate the interaction energies. The results are shown in Fig. 5. As seen in Fig. 5a, in the absence of solvent, the minimum for the free energy of interaction (more stable arrangement) was achieved when one paraquat unity interacted with three dodecylsulfate ions, yielding an ion-associate with a total charge of 1. In the presence of the solvent (Fig. 5b), the most stable arrangement was obtained when four dodecylsulfate ions were attached to one paraquat ion, resulting in an ion-associate with a total charge of 2. This change occurs because the solvent stabilizes the excess of negative charges present in the system. These results indicate that the most likely form of the ion-associate occurs when one paraquat ion is surrounded by four dodecylsulfate anions, in which the anions are attached to the paraquat molecule through interactions between the sulfate groups and the nitrogen or hydrogen atoms of paraquat. In this way, the hydrophobic part of each dodecylsulfate ion must be directed outside, as shown in Fig. 4c, and is responsible for ionassociate adsorption onto the hydrophobic PUF. The obtained results highlight the importance of the large excess of surfactant that must be added to the solution, which was verified in previous experiments, as several dodecylsulfate ions are needed to interact with each paraquat unit. 3.6. Influence of initial paraquat concentration The influence of the initial concentration of paraquat on the adsorption process was investigated in the 3.90 106e3.90 105 mol L1 concentration range. As optimized previously, the concentration of surfactant added to the solutions was always 50 times higher than the paraquat concentration. The mass of PUF employed was 300 mg.
204
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
Fig. 4. Optimized geometries of the substances of interest in the gas phase. (a) The dodecylsulfate anion, (b) the paraquat ion, and (c) the interaction between paraquat and the dodecylsulfate ions (first approach for ion-associate).
The removal percentage did not vary significantly with the variation of the initial solute concentration, thus indicating that the number of active sites present in 300 mg of the PUF was sufficient for the adsorption of 2.2 mg of paraquat. The mean removal percentage was 90.9 ± 1.5%, which proved that the material had a high capacity for paraquat removal from aqueous solutions under the optimized conditions. This high capacity can be explained by the elevated number of active sites on PUF and/or by the occurrence of multilayer adsorption. Another important aspect is the time required to achieve the equilibrium. In all cases, equilibrium was achieved in approximately 10e15 min, evidencing that the initial concentration of the solute did not influence the adsorption kinetics. The amount (mass) of paraquat adsorbed increased linearly with the increase of its initial concentration, indicating that a process similar to a liquideliquid extraction regulated the adsorption. Some authors support that, in this case, PUF acts as a solid extractor, in a similar way to an organic solvent in liquideliquid extraction (Gough and Gesser, 1975; El-Shahawi, 1994; Mori and Cassella, 2009; Leite et al., 2012). The process could be described by the Nernst partition law, and the slope of the straight line, which represents the partition constant (Kd), was 1.50 L g1. This indicated that the solvent-like mechanism was predominant in the regulation of the adsorption process. It is important to note that the predominance of a solvent-like mechanism is compatible with the structure of the ion-associate predicted in the computational study, because the partition of the “hydrophobic” solute should be strongly dependent on the hydrophobic chain of the dodecylsulfate anion that surrounds the paraquat moiety. 3.7. Evaluation of the equilibrium conditions (isotherms)
Fig. 5. Interaction free energies between paraquat and the dodecylsulfate anions, according to the total charge of the ion-associate that is formed: (a) in the gas phase and (b) in aqueous medium.
The characterization of a given adsorption system can be performed through the application of adsorption isotherms, which provide a quantitative description of the adsorption capacity under equilibrium conditions. Several isotherms have been reported in previous studies, but the most employed isotherms are, undoubtedly, Langmuir and Freundlich isotherms, which were used to evaluate the adsorption process studied in this work. Firstly, the Langmuir isotherm was applied to the system, because this isotherm has already been successfully employed to
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
describe paraquat adsorption from aqueous media when using other materials (Cheah et al., 1997; Hsu et al., 2009; Nanseu-Njiki et al., 2010). The Langmuir isotherm is very restrictive and good adjustment of the data can only be verified when some conditions are satisfied: (i) the adsorption must occur on a monolayer; (ii) each molecule of solute must not interact with other adsorbed molecules, and (iii) each active site of the adsorbent must have an identical capacity for adsorption. The Langmuir isotherm can be written in its linear form as shown in Equation (2):
Ce 1 Ce þ ¼ qe Q $b Qe
(2)
where Ce (mol L1) is the concentration of solute that remained in the solution at equilibrium, qe (mol g1) is the amount of solute retained on the adsorbent at equilibrium, Q is the maximum adsorption capacity of the monolayer, and b is the apparent adsorption equilibrium constant. The Langmuir isotherm was not able to describe the adsorption of paraquat (in the form of an ion-associate with SDS) by PUF. A poor linear correlation was observed between Ce and Ce/qe (r2 ¼ 0.3465), indicating that the experimental data could not be adjusted to the Langmuir equation. This behavior is probably associated with the fact that one or more of the conditions of the Langmuir isotherm are not satisfied. As adsorption can occur on multilayers, as stated previously, it was expected that the Langmuir isotherm could not be used to characterize this system. The Freundlich isotherm was subsequently tested; this model is represented by an empirical equation that is largely employed to describe the behavior of heterogeneous systems (Proctor and ToroVazquez, 1996), such as the system studied in the present work. The Freundlich equation can be written in its linear form as shown in Equation (3):
1 log qe ¼ log KF þ log Ce n
(3)
where KF and 1/n are parameters related to the adsorption extension and degree of nonlinearity between solution concentration and adsorption, respectively. Satisfactory adjustment of the data to the Freundlich isotherm was observed for the present system, with an equation of log qe ¼ 1.9935 þ 1.3265 log Ce and a coefficient of determination (r2) of 0.9574. The adjustment of the data to the Freundlich isotherm reinforced the idea that the adsorption process was carried out through a solvent-like mechanism, considering PUF as a solid polymeric extractor. The adsorption parameters 1/n and KF presented values of 1.33 and 98.5 mol g1, respectively. According to Al-Duri (1995), values of 1/n higher than unity can indicate the occurrence of physical adsorption, which is in accordance with previous data verified in this work. Also, the elevated value of KF evidenced the high capacity of PUF to retain the ion-associate, which can be associated with multilayer adsorption (Magdy and Daifullah, 1998).
205
system under study was reached after only 10e15 min of shaking, which was independent of the initial solute concentration in the medium. Two kinetic models were used to evaluate the adsorption of paraquat. Firstly, the Lagergren pseudo-first-order equation (Eq. (4)) was applied to evaluate adsorption kinetics (Tseng et al., 2010):
logðqe qt Þ ¼ log qe
k t 2:303
(4)
where qt is the concentration of solute adsorbed at time t (mol g1) and k is the overall rate constant. Although a satisfactory correlation between log (qe e qt) and t could be observed in the application of the model (with r2 in the range of 0.975e0.998, depending on the initial concentration of paraquat), its predictive capacity was very poor, showing that the adsorption of paraquat (as an ion-associate with SDS) must not follow pseudo-first-order kinetics. Elevated errors, up to 80%, were verified when the pseudo-first-order model equation was employed to predict the amount of paraquat adsorbed at equilibrium for the different initial concentrations of solute. However, the pseudo-second-order model presented an excellent predictive capacity for the adsorption of paraquat by PUF, similarly to other adsorption systems (Hamadi et al., 2004; Brigante and Schulz, 2011; Silva et al., 2011). The pseudo-second-order model can be expressed as shown in Equation (5) (Ho and McKay, 1999):
t 1 1 ¼ þ t qt k2 $q2e qe
(5)
where k2 (g min1 mol1) is the pseudo-second-order adsorption constant. In the case of the present system, an excellent correlation (r2 higher than 0.999) could be observed between t/qt and t when the pseudo-second-order model was applied (Fig. 6), which was independent of the initial paraquat concentration in solution. Additionally, the use of this model allowed the satisfactory prediction of the paraquat concentration adsorbed by PUF under equilibrium conditions, with the observed differences always lower than 3%. The results obtained during the application of the pseudo-secondorder model for the different initial concentrations of paraquat are shown in Table 1. Kinetic evaluation of the adsorption systems was also
3.8. Kinetic characterization of the system Kinetic studies of the adsorption systems can provide important information about their behavior, such as the time required to attain equilibrium, the adsorption rate, and the prediction of solute concentration in each phase after equilibrium is reached. Also, such data are important in the understanding of the influence of some practical variables on the adsorption process (Lazaridis and Asouhidou, 2003). As mentioned previously, the adsorption equilibrium of the
Fig. 6. Results obtained during the application of the pseudo-second-order kinetic model for the adsorption of paraquat onto PUF. [SDS]/[paraquat] ¼ 49.
206
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
performed by applying diffusion models, which were used to evaluate whether the solute transference from the liquid to the solid phase was regulated by film-diffusion and/or intraparticlediffusion processes (Qiu et al., 2009). Diffusion models are generally based on the observation that three processes occur before the solute molecules are truly adsorbed: (i) diffusion of solute molecules through the interfacial film that encompasses the adsorbent particles, which is usually referred to as film diffusion; (ii) diffusion of solute molecules within the adsorbent pores (intraparticle diffusion); and (iii) actual adsorption/absorption of the solute molecules on/in active sites of the adsorbent. The occurrence of such processes can be verified by testing the MorriseWeber and Reichenberg models. The MorriseWeber model is useful for verifying whether a diffusion process (intraparticle diffusion, film diffusion, or both) is controlling the adsorption rate. This model considers that a linear relationship exists between the adsorption capacity and the square root of time. If a straight line is obtained when t1/2 is plotted against qt, a diffusion process is active in controlling the adsorption kinetics. The MorriseWeber model was firstly applied in this scenario, which can be described by Equation (6):
qt ¼ kd $t 1=2 þ C
Fig. 7. Application of the intraparticle-diffusion model (MorriseWeber) for the adsorption of paraquat onto PUF. [SDS]/[paraquat] ¼ 49.
(Reichenberg, 1953). The Reichenberg model has been used to evaluate diffusion processes, providing information about their nature (film or intraparticle diffusion). The Reichenberg model can be expressed by Equation (7):
(6)
where kd is the intraparticle diffusion constant. It is important to remember that C is a parameter of the function that indicates the resistance of the interfacial film (Crini et al., 2007) The application of the MorriseWeber model to paraquat adsorption by PUF resulted in the appearance of a multilinear relationship between t1/2 and qt (Fig. 7). This behavior confirmed that a diffusion process controls the adsorption rate of paraquat by PUF. Within the first 4 min, a straight line passing through the origin was observed. According to Crini et al. (2007), the resistance of the interfacial film is practically null (C z 0) under these conditions, and the intraparticle diffusion process must regulate the transference of the solute from the liquid to the solid phase. Between 4 and 10 min, the slope decreases, leading to the occurrence of a value of C (if the line is extrapolated) higher than 0. In this situation, it can be concluded that the resistance of the interfacial film increases, probably due to its occupation with the solute molecules. The smaller inclination of the straight line indicates a decrease in the influence of the intraparticle diffusion process on the control of solute adsorption, as the film-diffusion process becomes active. Finally, after 10 min extraction, the function becomes parallel to the t1/2 axis, indicating that the equilibrium condition has been reached and that the film resistance (C) has achieved its maximum value. Under these conditions, diffusion processes no longer influence adsorption (slope ¼ 0) and the exchange of solute molecules between the liquid and solid phases should be neglected, owing to a very low concentration of paraquat remaining in the solution. In order to confirm the occurrence of intraparticle diffusion as the rate-limiting step of adsorption, especially in the beginning of the extraction process, the Reichenberg model was tested
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Bt ¼ 6:28318 3:2899F 6:28318 ð1 1:047FÞ
(7)
where Bt is the Reichenberg parameter and F is the qt/qe ratio. The results obtained during the application of the Reichenberg model showed, as expected, that intraparticle diffusion was very active in the beginning of the adsorption process. Straight lines (r2 higher than 0.90), passing through the origin, were obtained by the application of the Reichenberg model to the data obtained in the first 10 min of extraction, proving that intraparticle diffusion controlled the adsorption rate within this period of time. 4. Conclusions The results obtained in this work showed that PUF can be used to remove paraquat from aqueous media in the presence of SDS. The adsorption process was dependent upon the concentration of SDS in solution, which controlled the formation of an ion-associate between paraquat and the dodecylsulfate anion. The highest removal percentage was observed when the concentration of SDS was approximately 50 times higher than the paraquat concentration. Under optimized conditions, and using 300 mg of the adsorbent, PUF was able to retain more than 95% of the paraquat present in 45 mL of a 2.05 105 mol L1 solution, which represents approximately 2.2 mg of paraquat. Computational studies indicated that the ion-associate that formed in aqueous solution was composed of one unit of paraquat surrounded by four dodecylsulfate anion, which were bonded to the pesticide through interactions between the oxygen atoms of the
Table 1 Evaluation of the pseudo-second-order kinetic model for the adsorption of paraquat by PUF at different initial concentrations. Parameter
r2 k2 (g min1 mol1) qpredicted (mol g1) e qobserved (mol g1) e Difference (%)
Initial concentration of paraquat ( 106 mol L1) 3.90
7.80
15.5
21.5
29.5
39.0
1.000 4.96 106 5.00 107 4.95 107 1.1
1.000 3.18 106 1.03 106 1.03 106 0.6
0.9999 1.76 106 2.13 106 2.12 106 0.8
0.9999 1.54 106 3.03 106 2.94 106 3.0
0.9999 2.80 105 4,05 106 4.03 106 0.6
1.000 2.75 105 5.31 106 5.27 106 0.7
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
sulfate moiety and the nitrogen and hydrogen atoms of the paraquat. The system could be modeled by using the Freundlich isotherm and followed pseudo-second-order kinetics. The application of diffusion models (MorriseWeber and Reichenberg) demonstrated that solute transference from the liquid to the solid phase was controlled by both film- and intraparticle-diffusion processes. Acknowledgments ~o de Aperfeiçoamento de The authors are grateful to Coordenaça ~o Carlos Chagas Filho de Pessoal de Nível Superior (CAPES), Fundaça Pesquisa do Estado do Rio de Janeiro (FAPERJ), Fundaça ~o Amparo a Pesquisa do Estado de Minas Gerais (FAPEMIG) and to de Amparo a gico Conselho Nacional de Desenvolvimento Científico e Tecnolo (CNPq) for providing grants, scholarships and financial support. References Al-Duri, B., 1995. Adsorption modeling and mass transfer. In: McKay, G. (Ed.), Use of Adsorbents for the Removal of Pollutants from Wastewaters. CRC Press Inc., Boca Raton. Almeida, G.N., Souza, L.M., Pereira Netto, A.D., Cassella, R.J., 2007. Characterization of solid-phase extraction of Fe(III) by unloaded polyurethane foam as thiocyanate complex. J. Colloid Interface Sci. 315, 63e69. Baldez, E.E., Robaina, N.F., Cassella, R.J., 2008. Employment of polyurethane foam for the adsorption of methylene blue in aqueous medium. J. Hazard. Mater. 159, 580e586. Baldez, E.E., Robaina, N.F., Cassella, R.J., 2009. Study of rhodamine B retention by polyurethane foam from aqueous medium in presence of sodium dodecylsulfate. Sep. Sci. Technol. 44, 3128e3149. Bowen, H.J.M., 1970. Adsorption by polyurethane foams: new method of separation. J. Chem. Soc. A 7, 1082e1085. Braun, T.D., Navratil, J.D., Farag, A.B., 1985. Polyurethane Foam Sorbents in Separation Science. CRC Press Inc., Boca Raton. Brigante, M., Zanini, G., Avena, M., 2010. Effect of humic acids on the adsorption of paraquat by goethite. J. Hazard. Mater. 184, 241e247. Brigante, M., Schulz, P.C., 2011. Adsorption of paraquat on mesoporous sílica modified with titania: effects of pH, ionic strength and temperature. J. Colloid Interface Sci. 363, 355e361. Bromilow, R.H., 2004. Paraquat and sustainable agriculture. Pest Manage. Sci. 60, 340e349. Cassella, R.J., Garrigues, S., Santelli, R.E., de la Guardia, M., 2000. Retention of carbaryl by polyether type polyurethane foam: a critical study. Analyst 125, 257e261. Cheah, U.-B., Kirkwood, R.C., Lum, K.-Y., 1997. Adsorption, desorption and mobility of four commonly used pesticides in Malaysian agricultural soils. Pestic. Sci. 50, 53e63. Clark, D.G., McElligott, T.F., Hurst, E.W., 1966. The toxicity of paraquat. Brit. J. Ind. Med. 23, 126e132. Crini, G., Peindy, H.N., Gimbert, F., Robert, C., 2007. Removal of C.I. Basic Green 4 (Malachite Green) from aqueous solutions by adsorption using cyclodextrinbased adsorbent: kinetic and equilibrium studies. Sep. Purif. Technol. 53, 97e110. Dewar, M.J.S., Zoebisch, E.G., Healy, E.F., 1985. AM1: a new general purpose quantum mechanical molecular model. J. Am. Chem. Soc. 107, 3902e3909. ~o, F., Carmo, H., Duarte, J.A., Navarro, A.S., Bastos, M.L., Dinis-Oliveira, R.J., Remia Carvalho, F., 2006. Paraquat exposure as an etiological factor of Parkinson's disease. Neurotoxicol 27, 1110e1122. Ditchfield, R., Hehre, W.J., Pople, J.A., 1971. Self-consistent molecular orbital methods. 9. Extended Gaussian-type basis for molecular-orbital studies of organic molecules. J. Chem. Phys. 54, 724e728. El-Shahawi, M.S., 1993. Preconcentration and separation of some organic water pollutants with polyurethane foam and activated carbon. Chromatographia 36, 318e322. El-Shahawi, M.S., 1994. Retention and separation of some organic water pollutants with unloaded and tri-n-octylamine loaded polyester-based polyurethane foams. Talanta 41, 1481e1488. El-Shahawi, M.S., Aldhaheri, S.M., 1996. Preconcentration and separation of acaricides by polyether based polyurethane foam. Anal. Chim. Acta 320, 277e287. El-Shahawi, M.S., 1997. Retention profiles of some commercial pesticides, pyrethroid and acaricide residues and their application to tomato and parsley plants. J. Chromatogr. A 760, 179e192. Farag, A.B., El-Wakil, A.M., El-Shahawi, M.S., 1986. Collection and separation of some organic insecticides on polyurethane foam columns. Fresenius J. Anal. Chem. 324, 59e60. Frisch, M.J., Trucks, G.W., Schlegel, H.B., Scuseria, G.E., Robb, M.A., Cheeseman, J.R., Scalmani, G., Barone, V., Mennucci, B., Petersson, G.A., Nakatsuji, H., Caricato, M., Li, X., Hratchian, H.P., Izmaylov, A.F., Bloino, J., Zheng, G., Sonnenberg, J.L., Hada,
207
M., Ehara, M., Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M., Nakajima, T., Honda, Y., Kitao, O., Nakai, H., Vreven, T., Montgomery, J.A.Jr., Peralta, J.E., Ogliaro, F., Bearpark, M., Heyd, J. J., Brothers, E., Kudin, K.N., Staroverov, V.N., Kobayashi, R., Normand, J., Raghavachari, K., Rendell, A., Burant, J.C., Iyengar, S.S., Tomasi, J., Cossi, M., Rega, N., Millam, N.J., Klene, M., Knox, J.E., Cross, J.B., Bakken, V., Adamo, C., Jaramillo, J., Gomperts, R., Stratmann, R.E., Yazyev, O., Austin, A.J., Cammi, R., Pomelli, C., Ochterski, J.W., Martin, R.L., Morokuma, K., Zakrzewski, V.G., Voth, G.A., Salvador, P., Dannenberg, J.J., Dapprich, S., Daniels, € Foresman, J.B., Ortiz, J.V., Cioslowski, J., Fox, D.J., 2009. Gaussian A.D., Farkas, O., Inc., Wallingford, CT, USA. Gough, K.M., Gesser, H.D., 1975. The extraction and recovery of phthalate esters from water using porous polyurethane foam. J. Chromatogr. 115, 383e390. Hamadi, N.K., Swaminathan, S., Chen, X.D., 2004. Adsorption of paraquat dichloride from aqueous solution by activated carbon derived from used tires. J. Hazard. Mater. 112, 133e141. Hehre, W.J., Ditchfield, R., Pople, J.A., 1972. Self-consistent molecular orbital methods. 12. Further extensions of Gaussian-type basis sets for use in molecular-orbital studies of organic-molecules. J. Chem. Phys. 56, 2257e2261. Ho, Y.S., McKay, G., 1999. Pseudo-second order model for sorption processes. Process Biochem. 34, 451e465. Hsu, S.-T., Chen, L.-C., Lee, C.-C., Pan, T.-C., You, B.-X., Yan, Q.-F., 2009. Preparation of methacrylic acid-modified rice husk improved by Na experimental design and application for paraquat adsorption. J. Hazard. Mater. 171, 465e470. pez, R., Gondar, D., Antelo, J., Fiol, S., Arce, F., 2010. Adsorption of Iglesias, A., Lo paraquat on goethite and humic acid-coated goethite. J. Hazard. Mater. 183, 664e668. Jarvie, D.R., Stewart, M.J., 1979. The rapid extraction of paraquat from plasma using an ion-pairing technique. Clin. Chim. Acta 94, 241e251. Kohn, W., Sham, L.J., 1965. Self-consistent equations including exchange and correlation effects. Phys. Rev. 140, 1133e1138. Lazaridis, N.K., Asouhidou, D.D., 2003. Kinetics of sorptive removal of chromium(VI) from aqueous solutions by calcined Mg-Al-CO3 hydrotalcite. Water Res. 37, 2875e2882. Lee, C., Yang, W., Parr, R.G., 1988. Development of the ColleeSalvetti correlationenergy formula into a functional of the electron density. Phys. Rev. B 37, 785e789. Leite, B.T., Robaina, N.F., dos Reis, L.G.T., Netto, A.D.P., Cassella, R.J., 2012. Removal of malachite Green from aqueous medium employing polyurethane foam as adsorbent and sodium dodecylsulfate as carrier. Water, Air, Soil Pollut. 223, 1303e1313. Leite, M.P., Reis, L.G.T., Robaina, N.F., Pacheco, W.F., Cassella, R.J., 2013. Adsorption of paraquat from aqueous medium by Amberlite XAD-2 and XAD-4 resins using dodecylsulfate as counter ion. Chem. Eng. J. 215e216, 691e698. Magdy, Y.H., Daifullah, A.A.M., 1998. Adsorption of a basic dye from aqueous solutions onto sugar-industry-mud in two modes of operations. Waste Manage 18, 219e226. Mandel, J.S., Adami, H.-O., Cole, P., 2012. Paraquat and Parkinson's disease: an overview of the epidemiology and a review of two recent studies. Regul. Toxicol. Pharmacol. 62, 385e392. dasy, M., Andriska, V., 1988. Studies in Environmental Science 32: Matolcsy, G., Na Pesticide Chemistry. Elsevier, New York, NY, USA. , G., 1996. ParaMelchiorri, D., Reiter, R.J., Sewerynek, E., Hara, M., Chen, L., Nistico quat toxicity and oxidative damage: reduction by melatonin. Biochem.. Pharmacol. 51, 1095e1099. Miehlich, B., Savin, A., Stoll, H., Preuss, H., 1989. Results obtained with the correlation-energy density functionals of Becke and Lee, Yang and Parr. Chem. Phys. Lett. 157, 200e206. Mori, M., Cassella, R.J., 2009. Sorption of crystal violet by polyurethane foam from aqueous medium containing sodium dodecylsulfate. Quím. Nova 32, 2039e2045. Nanseu-Njiki, C.P., Dedzo, G.K., Ngameni, E., 2010. Study of the removal of paraquat from aqueous solution by biosorption onto Ayous (Triplochiton schleroxylon) sawdust. J. Hazard. Mater. 179, 63e71. Parr, R.G., Yang, W., 1989. Density-functional Theory of Atoms and Molecules. Oxford University Press, New York, NY, USA. Proctor, A., Toro-Vazquez, J.F., 1996. The Freundlich isotherm in studying adsorption in oil processing. J. Am. Oil Chem. Soc. 73, 1627e1633. Qiu, H., Lv, L., Pan, B.-C., Zhang, Q.-J., Zhang, W.-M., Zhang, Q.-X., 2009. Critical review in adsorption kinetic models. J. Zhejiang Uni. Sci. A 10, 716e724. Reichenberg, D., 1953. Properties of ion-exchange resins in relation to their structure. III. Kinetics of exchange. J. Am. Chem. Soc. 75, 589e597. Robaina, N.F., Soriano, S., Cassella, R.J., 2009. Polyurethane foam loaded with SDS for the adsorption of cationic dyes from aqueous medium: multivariate optimization of the loading process. J. Hazard. Mater. 167, 653e659. Sant’Ana, O.D., Jesuino, L.S., Cassella, R.J., Carvalho, M.S., Santelli, R.E., 2003. Solidphase extraction of Cu (II) as diethyldithiocarbamate complex by polyurethane foam. J. Braz. Chem. Soc. 14, 728e733. Sant’Ana, O.D., Jesuino, L.S., Cassella, R.J., Carvalho, M.S., Santelli, R.E., 2004a. Determination of lead by electrothermal atomic absorption spectrometry employing a novel sampling strategy of polyurethane foam impregnated with thiazolylazo-p-cresol (TAC). J. Braz. Chem. Soc. 15, 96e102. Sant’Ana, O.D., Jesuino, L.S., Cassella, R.J., Carvalho, M.S., Santelli, R.E., 2004b. Solidphase extraction of Cd (II) as diethyldithiocarbamate complex by polyurethane foam: kinetic and thermodynamic characterization. Sep. Sci. Technol. 39, 2695e2709.
208
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
Seki, Y., Yurdakoç, K., 2005. Paraquat adsorption onto clays and organoclays from aqueous solution. J. Colloid Interface Sci. 287, 1e5. Shao, Y., Molnar, L.F., Jung, Y., Kussmann, J., Ochsenfeld, C., Brown, S.T., Gilbert, A.T.B., Slipchenko, L.V., Levchenko, S.V., O'Neill, D.P., DiStasio Jr., R.A., Lochan, R.C., Wang, R., Beran, G.J.O., Besley, N.A., Herbert, J.M., Lin, C.Y., Van Voorhis, T., Chien, S.H., Sodt, A., Steele, R.P., Rassolov, V.A., Maslen, P.E., Korambath, P.P., Adamson, R.D., Austin, B., Baker, J., Byrd, E.F.C., Dachsel, H., Doerksen, R.J., Dreuw, A., Dunietz, B.D., Dutoi, A.D., Furlani, T.R., Gwaltney, S.R., Heyden, A., Hirata, S., Hsu, C.-P., Kedziora, G., Khalliulin, R.Z., Klunzinger, P., Lee, A.M., Lee, M.S., Liang, W.Z., Lotan, I., Nair, N., Peters, B., Proynov, E.I., Pieniazek, P.A., Rhee, Y.M., Ritchie, J., Rosta, E., Sherrill, C.D., Simmonett, A.C., Subotnik, J.E., Woodcock III, H.L., Zhang, W., Bell, A.T., Chakraborty, A.K., Chipman, D.M., Keil, F.J., Warshel, A., Hehre, W.J., Schaefer, H.F., Kong, J., Krylov, A.I., Gill, P.M.W., Head-Gordon, M., 2006. Advances in methods and algorithms in a modern quantum chemistry program package. Phys. Chem. Chem. Phys. 8, 3172e3191. Silva, M.S., Cocenza, D.S., Grillo, R., Melo, N.F.S., Tonello, P.S., Oliveira, L.C.,
Cassimiro, D.L., Rosa, A.H., Fraceto, L.F., 2011. Paraquat-loaded alginate/chitosan nanoparticles: preparation, characterization and soil sorption studies. J. Hazard. Mater. 190, 366e374. Silveira Neta, J.J., Moreira, G.C., Silva, C.J., Reis, C., Reis, E.L., 2011. Use of polyurethane foams for the removal of the Direct Red 80 and Reactive Blue 21 dyes in aqueous medium. Desalination 281, 55e60. Tomasi, J., Mennucci, B., Cammi, R., 2005. Quantum mechanical continuum solvation models. Chem. Rev. 105, 2999e3093. Tsai, W.T., Hsien, K.J., Chang, Y.M., Lo, C.C., 2005. Removal of herbicide paraquat from an aqueous solution by adsorption onto spent and treated diatomaceous earth. Bioresour. Technol. 96, 657e663. Tsai, W.-T., Lai, C.-W., 2006. Adsorption of herbicide paraquat by clay mineral regenerated from spent bleaching earth. J. Hazard. Mater. 134, 144e148. Tseng, R.-L., Wu, F.-C., Juang, R.-S., 2010. Characteristics and applications of the Lagergren's first-order equation adsorption kinetics. J. Taiwan Inst. Chem. Eng 41, 661e669.
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Modeling, kinetic, and equilibrium characterization of paraquat adsorption onto polyurethane foam using the ion-pairing technique Walkimar M. Carneiro a, Claudio F. Lima b, Jonas O. Vinhal a, b, Mateus R. Lage a, Jose Ricardo J. Cassella a, * ~o em Química, Universidade Federal Fluminense, Outeiro de Sa ~o Joa ~o Batista s/n, Centro, Nitero s Graduaça i, RJ Instituto de Química, Programa de Po 24020-141, Brazil b Departamento de Química, Universidade Federal de Viçosa, Viçosa, MG 36570-000, Brazil a
a r t i c l e i n f o
a b s t r a c t
Article history: Received 9 June 2014 Received in revised form 2 March 2015 Accepted 13 March 2015 Available online
We studied the adsorption of paraquat onto polyurethane foam (PUF) when it was in a medium containing sodium dodecylsulfate (SDS). The adsorption efficiency was dependent on the concentration of SDS in solution, because the formation of an ion-associate between the cationic paraquat and the dodecylsulfate anion was found to be a fundamental step in the process. A computational study was carried out to identify the possible structure of the ion-associate in aqueous medium. The obtained data demonstrated that the structure is probably formed from four units of dodecylsulfate bonded to one paraquat moiety. The results showed that 94% of the paraquat present in 45 mL of a solution containing 3.90 105 mol L1 could be retained by 300 mg of PUF, resulting in the removal of 2.20 mg of paraquat. The experimental data were reasonably adjusted to the Freundlich isotherm and to the pseudo-secondorder kinetic model. Also, the application of MorriseWeber and Reichenberg models indicated that both film-diffusion and intraparticle-diffusion processes were active during the control of the adsorption kinetics. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Paraquat Polyurethane foam Adsorption Ion-pairing
1. Introduction Nowadays, contamination of aquatic systems can be considered as one of the main problems in modern society, because the shortage of water is a phenomenon already observed in several parts of the world (El-Shahawi, 1997). The discharge of pesticides into water bodies is one process that leads to the contamination of aquatic systems (Brigante and Schulz, 2011). Paraquat (1,10 -dimethyl-4,40 -dipyridyn, Fig. 1) is a quaternary ammonium salt that is largely employed as a herbicide in various types of cultures, owing to its chemical and physical properties such as high solubility in water (Matolcsy et al., 1988; Bromilow, 2004). Paraquat has been proven to be toxic to mammals, directly affecting the lungs, liver, and kidneys, among other organs (Clark et al., 1966; Melchiorri et al., 1996). Despite the fact that some studies have associated Parkinson's disease with paraquat contamination, it is not possible to affirm that this herbicide is
* Corresponding author. E-mail address: [email protected] (R.J. Cassella). http://dx.doi.org/10.1016/j.jenvman.2015.03.022 0301-4797/© 2015 Elsevier Ltd. All rights reserved.
responsible for the development of the disease (Dinis-Oliveira et al., 2006; Mandel et al., 2012). Different strategies can be used to eliminate and/or remove pollutants from water. Among these strategies, adsorption processes represent one of the most important strategies, especially because of the advantages observed when they are applied. In general, adsorption processes are simple, efficient, and versatile, because the use of different adsorbents allows the removal of solutes with varying chemical characteristics. Several studies have been conducted to evaluate the removal of paraquat from water by adsorption processes. Various materials have already been used for this purpose, such as clay minerals (Seki and Yurdakoç, 2005; Tsai and Lai, 2006; Brigante et al., 2010; Iglesias et al., 2010), diatomaceous earth (Tsai et al., 2005), rice husk (Hsu et al., 2009), mesoporous silica (Brigante and Schulz, 2011), sawdust (Nanseu-Njiki et al., 2010), activated carbon (Hamadi et al., 2004), and chitosan (Silva et al., 2011). In order to choose an adsorbent for the removal of chemical species from aqueous media, several aspects must be considered. Undoubtedly, the most important aspect is the solute affinity for the adsorbent, which will determine the efficiency of the process.
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
Fig. 1. Structure of paraquat in its ionic form.
Other aspects, such as cost, availability, and the need for pretreatment, are also important. Polyurethane foams (PUFs) satisfy all of these criteria, as they are low-cost, readily available, and do not require pretreatment, and PUFs can easily be obtained in the market in different forms and compositions (Braun et al., 1985). They are able to retain polar and non-polar solutes, owing to the different functional groups present in their structure (Bowen, 1970). PUFs have already been employed, with success, in the adsorption of metallic complexes (Sant’Ana et al., 2003; Sant’Ana et al., 2004a; Sant’Ana et al., 2004b; Almeida et al., 2007), other pesticides (Farag et al., 1986; El-Shahawi, 1993; El-Shahawi and Aldhaheri, 1996; El-Shahawi, 1997; Cassella et al., 2000), dyes (Baldez et al., 2008; Robaina et al., 2009; Baldez et al., 2009; Mori and Cassella, 2009; Silveira Neta et al., 2011; Leite et al., 2012), and other organic pollutants (El-Shahawi, 1994). There is, however, no report regarding the use of a PUF for the adsorption of paraquat or any other cationic herbicide. The main goal of this work was to optimize and characterize, in terms of kinetics and the equilibrium, the adsorption of paraquat from an aqueous medium by PUF, using dodecylsulfate as the counter ion. Additionally, a computational study was carried out in order to reveal the possible species that is actually adsorbed and to provide information about the adsorption mechanism. 2. Experimental 2.1. Apparatus Determination of the paraquat concentration in the solutions was performed by spectrophotometry using an Agilent UVevisible spectrophotometer (model Cary 60, Palo Alto, CA, USA). The spectra were registered in the range of 200e400 nm, and the absorbance at 257 nm was employed for quantitative purposes. All measurements were performed with a standard quartz cuvette with an optical path length of 10 mm. Agitation of the solutions was carried out on a horizontal roller mixer by Biomixer (Curitiba, Brazil). Capped polyethylene flasks with a 50 mL capacity were employed in the experiments. The total volume of solution used in each experiment was always 45 mL. The Spartan 6.0 software (Wavefunction Inc., CA, USA) was used to design the chemical structures for the input files employed in the computational calculations (Shao et al., 2006). This software was also used to perform conformational analysis of the molecules present in each system. Finally, Gaussian 09 (Gaussian Inc., Wallingford, CT, USA) software was used to run all density functional calculations (Frisch et al., 2009) and Chemcraft was used to view and edit the Gaussian output files. 2.2. Reagents and solutions All reagents used in this work were of analytical grade or higher and were employed without further purification. The deionized water employed in the preparation of solutions was purified in a Direct-Q System (Millipore, Bedford, USA). A 1000 mg L1 (3.90 103 mol L1) paraquat stock solution was prepared by dissolving 50 mg of the reagent (SigmaeAldrich, Steinheim, Germany) in approximately 20 mL of deionized water.
201
Afterwards, the obtained solution was quantitatively transferred to a 50 mL volumetric flask and the volume was made up to the mark with deionized water. The stock solution was stored in a refrigerator. The paraquat solutions employed in the experiments were prepared through the convenient dilution of the stock solution with deionized water. A 1000 mg L1 (3.90 103 mol L1) stock solution of sodium dodecylsulfate (SDS) was prepared by dissolving 0.250 g of the reagent (Vetec, Rio de Janeiro, Brazil) in approximately 100 mL of deionized water. Then, the solution was transferred to a 250 mL volumetric flask and the volume was made up to the mark with deionized water. This solution was stable for 1 week when stored at an ambient laboratory temperature (22 ± 1 C) in a light-free location. Diluted solutions of SDS, employed in the adsorption experiments, were prepared through the convenient dilution of the stock solution with deionized water. Open-cell, polyether-type PUF (Guararapes Ltda, Brazil) with a density of 22.5 mg cm3 was employed in all experiments. The foam was comminuted in a blender with purified water. Then, the powdered PUF was dried at ambient temperature and sieved through a 2 mm plastic sieve. The treated PUF was stored in a plastic flask, which was maintained in a light- and dust-free environment. 2.3. General procedure The adsorption experiments were performed by shaking 45 mL solutions, containing known concentrations of paraquat and SDS, with adequate mass of powdered PUF. The shaking was performed with a horizontal mixer operated at 90 rpm. The concentration of paraquat that remained in the solution after shaking was determined by spectrophotometry (at 257 nm), at certain time intervals, as stipulated in each experiment according to the information desired. All experiments were carried out at ambient temperature, which was always 22 ± 1 C. The removal percentage (R) was calculated using Equation (1):
Rð%Þ ¼ 100
Ct 100 Co
(1)
in which Co is the initial concentration of paraquat in the solution and Ct is the concentration of paraquat in the solution at a given time. 3. Results and discussion 3.1. Influence of SDS concentration The most important process variable, related to the adsorption procedure, investigated in this work, is the formation of an ionassociate between the cationic paraquat and the dodecylsulfate anion, because the highly soluble paraquat cation cannot be retained by the hydrophobic PUF, as previously observed in a study on the adsorption of cationic dyes by PUFs (Baldez et al., 2008, 2009; Mori and Cassella, 2009; Leite et al., 2012). Therefore, the addition of a correct concentration of SDS to the medium was fundamental in order to enhance paraquat adsorption by PUF. This parameter was evaluated in the range of 0 (no addition of SDS) to 400 mg L1 (0e1.40 103 mol L1). The mass of PUF employed in the experiment was 200 mg, the shaking speed was 90 rpm, and the initial concentration of paraquat in solution was 5.5 mg L1 (or 2.15 105 mol L1). The adsorption efficiency increased with increasing SDS concentration up to 200 mg L1. Above this SDS concentration, the removal efficiency remained virtually constant, once the variation
202
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
of the amount of paraquat adsorbed was very low (Fig. 2). In order to ensure quantitative (higher than 90%) adsorption of the herbicide, the SDS concentration was set at 300 mg L1 (1.05 103 mol L1) in all further experiments. For each of the SDS concentrations tested, the time required to attain equilibrium was 30 min, which indicated that the SDS concentration had no significant effect on the adsorption kinetics. The addition of SDS at higher concentrations ensured the conversion of all paraquat that was present in the solution into its respective ion-associate, which is a very important step of the process, as the ion-associate is the substance that can actually be adsorbed onto the PUF. The concentration of SDS needed to achieve the highest adsorption efficiency was 49 times higher than the concentration of paraquat, which demonstrated the low value of the formation constant of the ion-associate in aqueous medium. As demonstrated, the adsorption of paraquat depended on its conversion into the ion-associate and other several factors which will be investigated below. However, the structure of the ionassociate could not be determined from the experimental data. In order to elucidate the substance that was really adsorbed onto PUF, a theoretical study was performed using molecular modeling. 3.2. Influence of PUF mass
3.3. Influence of pH and ionic strength In most liquidesolid extraction processes, the pH of the medium can play a major role because it can affect the characteristic of the adsorbent surface and/or the acid-base distribution of the solute species in aqueous medium. As acid-base chemistry of paraquat in aqueous solutions is restricted, since there are no acid-base groups in its structure, the influence of the pH on the present work could be related only to the interaction of Hþ with nitrogen atoms present in the PUF structure. The effect of pH was tested in the range of 2.0e9.0. Solutions with pH higher than 9.0 were not tested because, in this situation, the molecule of paraquat was degraded due to the cleavage of pyridine rings in basic medium (Matolcsy et al., 1988). The control of the pH of the solutions was achieved with the use of the BrittoneRobinson buffer (phosphoric acid/acetate/borate) at 0.10 mol L1 total concentration. Solutions containing paraquat and SDS at concentrations of 2.15 105 and 1.05 103 mol L1, respectively, were employed in the experiments. Maximum extraction of paraquat was observed in the range of 7.0e9.0 (Fig. 3a). In the acid range, the extraction was impaired, probably due to the protonation of nitrogen atoms present in urethane groups of the solid-phase, which resulted in the formation of
The influence of the mass of adsorbent was investigated in the range of 50e400 mg of PUF. The evaluation of the effect of this parameter is very important because it controls the number of active sites exposed for adsorption and, in general, presents noticeable effect on both extraction efficiency and kinetics. The concentrations of paraquat and SDS in the solutions employed in this experiment were 2.15 105 and 1.05 103 mol L1, respectively. The extraction efficiency increased with the increasing of the mass of adsorbent up to 300 mg, achieving a removal percentage of approximately 95%. When the mass of adsorbent was 400 mg, no improvement of the removal percentage was verified, probably due to the excess of active sites already present in the resin. Therefore, an adsorbent dose of 300 mg was used in all further experiments. No noticeable effect of the mass of adsorbent on the adsorption kinetics was observed probably because the transportation of the solute from the solution to the liquidesolid interface was efficient at the shaking conditions established in the experiment.
Fig. 2. Effect of SDS concentration on the removal of paraquat by PUF in aqueous solution. The initial concentration of paraquat was 2.15 105 mol L1 (or 5.5 mg L1).
Fig. 3. Effect of (a) pH and (b) ionic strength on the removal of paraquat by PUF in aqueous solution. The initial concentration of paraquat was 2.15 105 mol L1 (or 5.5 mg L1).
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
a positively charged PUF surface and decreased the interaction of the hydrophobic ion-associate with the resin. The influence of the ionic strength has been verified in the liquideliquid extraction of the ion-associate paraquat-SDS. According to Jarvie and Stewart (1979), the extraction was affected by the ionic strength and the influence depended on the type and concentration of the electrolyte added to the solution. In this context, the effect of the ionic strength was evaluated in the range of 0.005e0.10 mol L1 of NaCl. This experiment was performed using paraquat and SDS in concentrations of 2.15 105 and 1.05 103 mol L1, respectively. The results obtained in this experiment (Fig. 3b) revealed that the ionic strength influenced the adsorption of paraquat by PUF. It was observed a significant decrease of the herbicide adsorption onto PUF with the increase of the NaCl concentration in the medium. According to Tsai and Lai (2006) and Brigante et al. (2010), there could be a competition between the cations of the electrolyte and paraquat by the negative sites of the adsorbent. However, the adsorption mechanism proposed in this work does not take into account the direct interaction between the cationic solute and negative sites present on the adsorbent surface. Therefore, it is believed that this behavior could be probably related to the formation of a positive electrostatic “shield” around dodecylsulfate as a consequence of the increase of the concentration of cations in the solution. This phenomenon impairs the formation of the ionassociate, a fundamental step of the process. 3.4. Influence of shaking speed The shaking speed can affect directly the adsorption process. The increase of shaking speed of the system increases the average kinetic energy of the species and hence the number of effective collisions between the solutes and the active sites of the adsorbent material. So, the shaking speed was evaluated in the range of 15e90 rpm. This experiment was performed using a horizontal shaker with 300 mg of PUF. The concentrations of paraquat and SDS were 2.15 105 and 1.05 103 mol L1, respectively. The results obtained in this experiment revealed that the shaking speed did not influence the extraction efficiency of paraquat. Extraction efficiencies of approximately 95% were observed in the whole range studied. 3.5. Computational studies on ion-associate formation in water Although the extraction of paraquat in the presence of SDS has already been reported (Jarvie and Stewart, 1979; Leite et al., 2013), no work demonstrating the occurrence and/or proposing the structure of the possible ion-associate has been reported. In order to fill this gap, a computational study was carried out to model the formation of the paraquateSDS ion-associate in aqueous solution and to identify the molecular structure of the solute that is actually adsorbed. As previously mentioned, computational calculations were performed in order to improve the understanding of the interaction between paraquat and dodecylsulfate ions and to evaluate the most likely stoichiometry and structure of the ion-associate. For this purpose, calculations were performed in the gas phase and in the presence of the solvent (water). In the first step, conformational analysis was carried out using the semi-empirical method Austin Model 1 (AM1) in Spartan 6.0 code (Dewar et al., 1985; Shao et al., 2006). This conformational study was performed to show that the most likely geometry for the ions could be obtained and to make the geometry optimization easier to converge. The optimization of the geometries of paraquat, the dodecylsulfate anion, and the ionassociate, in the gas phase, was performed at the B3LYP/6-31G(d) level using Gaussian 09 software. It provided geometries that
203
were confirmed as energy minima by calculating the second-order Hessian matrix (no negative eigenvalues) (Lee et al., 1988; Miehlich et al., 1989; Ditchfield et al., 1971; Hehre et al., 1972; Frish et al., 2009). Fig. 4 shows the calculated geometries of the isolated ions (Fig. 4a and b) and for the ion-associate that resulted from the interaction of one paraquat ion with one dodecylsulfate anion (Fig. 4c). According to B3LYP/6-31G(d) calculations (Kohn and Sham, 1965; Parr and Yang, 1989; Lee et al., 1988), the formation of the ion-associate between dodecylsulfate and paraquat ions (shown in Fig. 4c) must occur through two types of interactions: (i) the interaction between one oxygen atom of the dodecylsulfate anion, with one nitrogen atom of paraquat, or (ii) the interaction between other oxygen atom of dodecylsulfate with one hydrogen atom of paraquat (hydrogen bonding). These interactions were favored by the partial charges of the atoms involved. The dodecylsulfate oxygen atoms have a negative partial charge, whereas the nitrogen atoms, and consequently the pyridine hydrogen atoms, have a positive partial charge. When more dodecylsulfate ions were added to the system, the interactions followed the same trend as those observed in the first interaction. Density functional calculations were carried out by considering isolated ions in the gas phase and by including solvation effects (water) using the polarizable continuum solvation model (Tomasi et al., 2005). The approach consisted of the addition of between one and five dodecylsulfate ions, sequentially, to a single paraquat unity. The addition of the dodecylsulfate ions was continued until five anions were included, because the stability of the system was strongly affected when a sixth dodecylsulfate ion was added. In the situation with six dodecylsulfate ions, the ion-associate decomposes, probably due to the strong repulsion caused by the excess negative charges. The geometry of each ion-associate was fully optimized, which was followed by the calculation of vibrational frequencies to confirm the geometries as true minima and to obtain the thermodynamic parameters needed to calculate the interaction energies. The results are shown in Fig. 5. As seen in Fig. 5a, in the absence of solvent, the minimum for the free energy of interaction (more stable arrangement) was achieved when one paraquat unity interacted with three dodecylsulfate ions, yielding an ion-associate with a total charge of 1. In the presence of the solvent (Fig. 5b), the most stable arrangement was obtained when four dodecylsulfate ions were attached to one paraquat ion, resulting in an ion-associate with a total charge of 2. This change occurs because the solvent stabilizes the excess of negative charges present in the system. These results indicate that the most likely form of the ion-associate occurs when one paraquat ion is surrounded by four dodecylsulfate anions, in which the anions are attached to the paraquat molecule through interactions between the sulfate groups and the nitrogen or hydrogen atoms of paraquat. In this way, the hydrophobic part of each dodecylsulfate ion must be directed outside, as shown in Fig. 4c, and is responsible for ionassociate adsorption onto the hydrophobic PUF. The obtained results highlight the importance of the large excess of surfactant that must be added to the solution, which was verified in previous experiments, as several dodecylsulfate ions are needed to interact with each paraquat unit. 3.6. Influence of initial paraquat concentration The influence of the initial concentration of paraquat on the adsorption process was investigated in the 3.90 106e3.90 105 mol L1 concentration range. As optimized previously, the concentration of surfactant added to the solutions was always 50 times higher than the paraquat concentration. The mass of PUF employed was 300 mg.
204
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
Fig. 4. Optimized geometries of the substances of interest in the gas phase. (a) The dodecylsulfate anion, (b) the paraquat ion, and (c) the interaction between paraquat and the dodecylsulfate ions (first approach for ion-associate).
The removal percentage did not vary significantly with the variation of the initial solute concentration, thus indicating that the number of active sites present in 300 mg of the PUF was sufficient for the adsorption of 2.2 mg of paraquat. The mean removal percentage was 90.9 ± 1.5%, which proved that the material had a high capacity for paraquat removal from aqueous solutions under the optimized conditions. This high capacity can be explained by the elevated number of active sites on PUF and/or by the occurrence of multilayer adsorption. Another important aspect is the time required to achieve the equilibrium. In all cases, equilibrium was achieved in approximately 10e15 min, evidencing that the initial concentration of the solute did not influence the adsorption kinetics. The amount (mass) of paraquat adsorbed increased linearly with the increase of its initial concentration, indicating that a process similar to a liquideliquid extraction regulated the adsorption. Some authors support that, in this case, PUF acts as a solid extractor, in a similar way to an organic solvent in liquideliquid extraction (Gough and Gesser, 1975; El-Shahawi, 1994; Mori and Cassella, 2009; Leite et al., 2012). The process could be described by the Nernst partition law, and the slope of the straight line, which represents the partition constant (Kd), was 1.50 L g1. This indicated that the solvent-like mechanism was predominant in the regulation of the adsorption process. It is important to note that the predominance of a solvent-like mechanism is compatible with the structure of the ion-associate predicted in the computational study, because the partition of the “hydrophobic” solute should be strongly dependent on the hydrophobic chain of the dodecylsulfate anion that surrounds the paraquat moiety. 3.7. Evaluation of the equilibrium conditions (isotherms)
Fig. 5. Interaction free energies between paraquat and the dodecylsulfate anions, according to the total charge of the ion-associate that is formed: (a) in the gas phase and (b) in aqueous medium.
The characterization of a given adsorption system can be performed through the application of adsorption isotherms, which provide a quantitative description of the adsorption capacity under equilibrium conditions. Several isotherms have been reported in previous studies, but the most employed isotherms are, undoubtedly, Langmuir and Freundlich isotherms, which were used to evaluate the adsorption process studied in this work. Firstly, the Langmuir isotherm was applied to the system, because this isotherm has already been successfully employed to
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
describe paraquat adsorption from aqueous media when using other materials (Cheah et al., 1997; Hsu et al., 2009; Nanseu-Njiki et al., 2010). The Langmuir isotherm is very restrictive and good adjustment of the data can only be verified when some conditions are satisfied: (i) the adsorption must occur on a monolayer; (ii) each molecule of solute must not interact with other adsorbed molecules, and (iii) each active site of the adsorbent must have an identical capacity for adsorption. The Langmuir isotherm can be written in its linear form as shown in Equation (2):
Ce 1 Ce þ ¼ qe Q $b Qe
(2)
where Ce (mol L1) is the concentration of solute that remained in the solution at equilibrium, qe (mol g1) is the amount of solute retained on the adsorbent at equilibrium, Q is the maximum adsorption capacity of the monolayer, and b is the apparent adsorption equilibrium constant. The Langmuir isotherm was not able to describe the adsorption of paraquat (in the form of an ion-associate with SDS) by PUF. A poor linear correlation was observed between Ce and Ce/qe (r2 ¼ 0.3465), indicating that the experimental data could not be adjusted to the Langmuir equation. This behavior is probably associated with the fact that one or more of the conditions of the Langmuir isotherm are not satisfied. As adsorption can occur on multilayers, as stated previously, it was expected that the Langmuir isotherm could not be used to characterize this system. The Freundlich isotherm was subsequently tested; this model is represented by an empirical equation that is largely employed to describe the behavior of heterogeneous systems (Proctor and ToroVazquez, 1996), such as the system studied in the present work. The Freundlich equation can be written in its linear form as shown in Equation (3):
1 log qe ¼ log KF þ log Ce n
(3)
where KF and 1/n are parameters related to the adsorption extension and degree of nonlinearity between solution concentration and adsorption, respectively. Satisfactory adjustment of the data to the Freundlich isotherm was observed for the present system, with an equation of log qe ¼ 1.9935 þ 1.3265 log Ce and a coefficient of determination (r2) of 0.9574. The adjustment of the data to the Freundlich isotherm reinforced the idea that the adsorption process was carried out through a solvent-like mechanism, considering PUF as a solid polymeric extractor. The adsorption parameters 1/n and KF presented values of 1.33 and 98.5 mol g1, respectively. According to Al-Duri (1995), values of 1/n higher than unity can indicate the occurrence of physical adsorption, which is in accordance with previous data verified in this work. Also, the elevated value of KF evidenced the high capacity of PUF to retain the ion-associate, which can be associated with multilayer adsorption (Magdy and Daifullah, 1998).
205
system under study was reached after only 10e15 min of shaking, which was independent of the initial solute concentration in the medium. Two kinetic models were used to evaluate the adsorption of paraquat. Firstly, the Lagergren pseudo-first-order equation (Eq. (4)) was applied to evaluate adsorption kinetics (Tseng et al., 2010):
logðqe qt Þ ¼ log qe
k t 2:303
(4)
where qt is the concentration of solute adsorbed at time t (mol g1) and k is the overall rate constant. Although a satisfactory correlation between log (qe e qt) and t could be observed in the application of the model (with r2 in the range of 0.975e0.998, depending on the initial concentration of paraquat), its predictive capacity was very poor, showing that the adsorption of paraquat (as an ion-associate with SDS) must not follow pseudo-first-order kinetics. Elevated errors, up to 80%, were verified when the pseudo-first-order model equation was employed to predict the amount of paraquat adsorbed at equilibrium for the different initial concentrations of solute. However, the pseudo-second-order model presented an excellent predictive capacity for the adsorption of paraquat by PUF, similarly to other adsorption systems (Hamadi et al., 2004; Brigante and Schulz, 2011; Silva et al., 2011). The pseudo-second-order model can be expressed as shown in Equation (5) (Ho and McKay, 1999):
t 1 1 ¼ þ t qt k2 $q2e qe
(5)
where k2 (g min1 mol1) is the pseudo-second-order adsorption constant. In the case of the present system, an excellent correlation (r2 higher than 0.999) could be observed between t/qt and t when the pseudo-second-order model was applied (Fig. 6), which was independent of the initial paraquat concentration in solution. Additionally, the use of this model allowed the satisfactory prediction of the paraquat concentration adsorbed by PUF under equilibrium conditions, with the observed differences always lower than 3%. The results obtained during the application of the pseudo-secondorder model for the different initial concentrations of paraquat are shown in Table 1. Kinetic evaluation of the adsorption systems was also
3.8. Kinetic characterization of the system Kinetic studies of the adsorption systems can provide important information about their behavior, such as the time required to attain equilibrium, the adsorption rate, and the prediction of solute concentration in each phase after equilibrium is reached. Also, such data are important in the understanding of the influence of some practical variables on the adsorption process (Lazaridis and Asouhidou, 2003). As mentioned previously, the adsorption equilibrium of the
Fig. 6. Results obtained during the application of the pseudo-second-order kinetic model for the adsorption of paraquat onto PUF. [SDS]/[paraquat] ¼ 49.
206
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
performed by applying diffusion models, which were used to evaluate whether the solute transference from the liquid to the solid phase was regulated by film-diffusion and/or intraparticlediffusion processes (Qiu et al., 2009). Diffusion models are generally based on the observation that three processes occur before the solute molecules are truly adsorbed: (i) diffusion of solute molecules through the interfacial film that encompasses the adsorbent particles, which is usually referred to as film diffusion; (ii) diffusion of solute molecules within the adsorbent pores (intraparticle diffusion); and (iii) actual adsorption/absorption of the solute molecules on/in active sites of the adsorbent. The occurrence of such processes can be verified by testing the MorriseWeber and Reichenberg models. The MorriseWeber model is useful for verifying whether a diffusion process (intraparticle diffusion, film diffusion, or both) is controlling the adsorption rate. This model considers that a linear relationship exists between the adsorption capacity and the square root of time. If a straight line is obtained when t1/2 is plotted against qt, a diffusion process is active in controlling the adsorption kinetics. The MorriseWeber model was firstly applied in this scenario, which can be described by Equation (6):
qt ¼ kd $t 1=2 þ C
Fig. 7. Application of the intraparticle-diffusion model (MorriseWeber) for the adsorption of paraquat onto PUF. [SDS]/[paraquat] ¼ 49.
(Reichenberg, 1953). The Reichenberg model has been used to evaluate diffusion processes, providing information about their nature (film or intraparticle diffusion). The Reichenberg model can be expressed by Equation (7):
(6)
where kd is the intraparticle diffusion constant. It is important to remember that C is a parameter of the function that indicates the resistance of the interfacial film (Crini et al., 2007) The application of the MorriseWeber model to paraquat adsorption by PUF resulted in the appearance of a multilinear relationship between t1/2 and qt (Fig. 7). This behavior confirmed that a diffusion process controls the adsorption rate of paraquat by PUF. Within the first 4 min, a straight line passing through the origin was observed. According to Crini et al. (2007), the resistance of the interfacial film is practically null (C z 0) under these conditions, and the intraparticle diffusion process must regulate the transference of the solute from the liquid to the solid phase. Between 4 and 10 min, the slope decreases, leading to the occurrence of a value of C (if the line is extrapolated) higher than 0. In this situation, it can be concluded that the resistance of the interfacial film increases, probably due to its occupation with the solute molecules. The smaller inclination of the straight line indicates a decrease in the influence of the intraparticle diffusion process on the control of solute adsorption, as the film-diffusion process becomes active. Finally, after 10 min extraction, the function becomes parallel to the t1/2 axis, indicating that the equilibrium condition has been reached and that the film resistance (C) has achieved its maximum value. Under these conditions, diffusion processes no longer influence adsorption (slope ¼ 0) and the exchange of solute molecules between the liquid and solid phases should be neglected, owing to a very low concentration of paraquat remaining in the solution. In order to confirm the occurrence of intraparticle diffusion as the rate-limiting step of adsorption, especially in the beginning of the extraction process, the Reichenberg model was tested
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Bt ¼ 6:28318 3:2899F 6:28318 ð1 1:047FÞ
(7)
where Bt is the Reichenberg parameter and F is the qt/qe ratio. The results obtained during the application of the Reichenberg model showed, as expected, that intraparticle diffusion was very active in the beginning of the adsorption process. Straight lines (r2 higher than 0.90), passing through the origin, were obtained by the application of the Reichenberg model to the data obtained in the first 10 min of extraction, proving that intraparticle diffusion controlled the adsorption rate within this period of time. 4. Conclusions The results obtained in this work showed that PUF can be used to remove paraquat from aqueous media in the presence of SDS. The adsorption process was dependent upon the concentration of SDS in solution, which controlled the formation of an ion-associate between paraquat and the dodecylsulfate anion. The highest removal percentage was observed when the concentration of SDS was approximately 50 times higher than the paraquat concentration. Under optimized conditions, and using 300 mg of the adsorbent, PUF was able to retain more than 95% of the paraquat present in 45 mL of a 2.05 105 mol L1 solution, which represents approximately 2.2 mg of paraquat. Computational studies indicated that the ion-associate that formed in aqueous solution was composed of one unit of paraquat surrounded by four dodecylsulfate anion, which were bonded to the pesticide through interactions between the oxygen atoms of the
Table 1 Evaluation of the pseudo-second-order kinetic model for the adsorption of paraquat by PUF at different initial concentrations. Parameter
r2 k2 (g min1 mol1) qpredicted (mol g1) e qobserved (mol g1) e Difference (%)
Initial concentration of paraquat ( 106 mol L1) 3.90
7.80
15.5
21.5
29.5
39.0
1.000 4.96 106 5.00 107 4.95 107 1.1
1.000 3.18 106 1.03 106 1.03 106 0.6
0.9999 1.76 106 2.13 106 2.12 106 0.8
0.9999 1.54 106 3.03 106 2.94 106 3.0
0.9999 2.80 105 4,05 106 4.03 106 0.6
1.000 2.75 105 5.31 106 5.27 106 0.7
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
sulfate moiety and the nitrogen and hydrogen atoms of the paraquat. The system could be modeled by using the Freundlich isotherm and followed pseudo-second-order kinetics. The application of diffusion models (MorriseWeber and Reichenberg) demonstrated that solute transference from the liquid to the solid phase was controlled by both film- and intraparticle-diffusion processes. Acknowledgments ~o de Aperfeiçoamento de The authors are grateful to Coordenaça ~o Carlos Chagas Filho de Pessoal de Nível Superior (CAPES), Fundaça Pesquisa do Estado do Rio de Janeiro (FAPERJ), Fundaça ~o Amparo a Pesquisa do Estado de Minas Gerais (FAPEMIG) and to de Amparo a gico Conselho Nacional de Desenvolvimento Científico e Tecnolo (CNPq) for providing grants, scholarships and financial support. References Al-Duri, B., 1995. Adsorption modeling and mass transfer. In: McKay, G. (Ed.), Use of Adsorbents for the Removal of Pollutants from Wastewaters. CRC Press Inc., Boca Raton. Almeida, G.N., Souza, L.M., Pereira Netto, A.D., Cassella, R.J., 2007. Characterization of solid-phase extraction of Fe(III) by unloaded polyurethane foam as thiocyanate complex. J. Colloid Interface Sci. 315, 63e69. Baldez, E.E., Robaina, N.F., Cassella, R.J., 2008. Employment of polyurethane foam for the adsorption of methylene blue in aqueous medium. J. Hazard. Mater. 159, 580e586. Baldez, E.E., Robaina, N.F., Cassella, R.J., 2009. Study of rhodamine B retention by polyurethane foam from aqueous medium in presence of sodium dodecylsulfate. Sep. Sci. Technol. 44, 3128e3149. Bowen, H.J.M., 1970. Adsorption by polyurethane foams: new method of separation. J. Chem. Soc. A 7, 1082e1085. Braun, T.D., Navratil, J.D., Farag, A.B., 1985. Polyurethane Foam Sorbents in Separation Science. CRC Press Inc., Boca Raton. Brigante, M., Zanini, G., Avena, M., 2010. Effect of humic acids on the adsorption of paraquat by goethite. J. Hazard. Mater. 184, 241e247. Brigante, M., Schulz, P.C., 2011. Adsorption of paraquat on mesoporous sílica modified with titania: effects of pH, ionic strength and temperature. J. Colloid Interface Sci. 363, 355e361. Bromilow, R.H., 2004. Paraquat and sustainable agriculture. Pest Manage. Sci. 60, 340e349. Cassella, R.J., Garrigues, S., Santelli, R.E., de la Guardia, M., 2000. Retention of carbaryl by polyether type polyurethane foam: a critical study. Analyst 125, 257e261. Cheah, U.-B., Kirkwood, R.C., Lum, K.-Y., 1997. Adsorption, desorption and mobility of four commonly used pesticides in Malaysian agricultural soils. Pestic. Sci. 50, 53e63. Clark, D.G., McElligott, T.F., Hurst, E.W., 1966. The toxicity of paraquat. Brit. J. Ind. Med. 23, 126e132. Crini, G., Peindy, H.N., Gimbert, F., Robert, C., 2007. Removal of C.I. Basic Green 4 (Malachite Green) from aqueous solutions by adsorption using cyclodextrinbased adsorbent: kinetic and equilibrium studies. Sep. Purif. Technol. 53, 97e110. Dewar, M.J.S., Zoebisch, E.G., Healy, E.F., 1985. AM1: a new general purpose quantum mechanical molecular model. J. Am. Chem. Soc. 107, 3902e3909. ~o, F., Carmo, H., Duarte, J.A., Navarro, A.S., Bastos, M.L., Dinis-Oliveira, R.J., Remia Carvalho, F., 2006. Paraquat exposure as an etiological factor of Parkinson's disease. Neurotoxicol 27, 1110e1122. Ditchfield, R., Hehre, W.J., Pople, J.A., 1971. Self-consistent molecular orbital methods. 9. Extended Gaussian-type basis for molecular-orbital studies of organic molecules. J. Chem. Phys. 54, 724e728. El-Shahawi, M.S., 1993. Preconcentration and separation of some organic water pollutants with polyurethane foam and activated carbon. Chromatographia 36, 318e322. El-Shahawi, M.S., 1994. Retention and separation of some organic water pollutants with unloaded and tri-n-octylamine loaded polyester-based polyurethane foams. Talanta 41, 1481e1488. El-Shahawi, M.S., Aldhaheri, S.M., 1996. Preconcentration and separation of acaricides by polyether based polyurethane foam. Anal. Chim. Acta 320, 277e287. El-Shahawi, M.S., 1997. Retention profiles of some commercial pesticides, pyrethroid and acaricide residues and their application to tomato and parsley plants. J. Chromatogr. A 760, 179e192. Farag, A.B., El-Wakil, A.M., El-Shahawi, M.S., 1986. Collection and separation of some organic insecticides on polyurethane foam columns. Fresenius J. Anal. Chem. 324, 59e60. Frisch, M.J., Trucks, G.W., Schlegel, H.B., Scuseria, G.E., Robb, M.A., Cheeseman, J.R., Scalmani, G., Barone, V., Mennucci, B., Petersson, G.A., Nakatsuji, H., Caricato, M., Li, X., Hratchian, H.P., Izmaylov, A.F., Bloino, J., Zheng, G., Sonnenberg, J.L., Hada,
207
M., Ehara, M., Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M., Nakajima, T., Honda, Y., Kitao, O., Nakai, H., Vreven, T., Montgomery, J.A.Jr., Peralta, J.E., Ogliaro, F., Bearpark, M., Heyd, J. J., Brothers, E., Kudin, K.N., Staroverov, V.N., Kobayashi, R., Normand, J., Raghavachari, K., Rendell, A., Burant, J.C., Iyengar, S.S., Tomasi, J., Cossi, M., Rega, N., Millam, N.J., Klene, M., Knox, J.E., Cross, J.B., Bakken, V., Adamo, C., Jaramillo, J., Gomperts, R., Stratmann, R.E., Yazyev, O., Austin, A.J., Cammi, R., Pomelli, C., Ochterski, J.W., Martin, R.L., Morokuma, K., Zakrzewski, V.G., Voth, G.A., Salvador, P., Dannenberg, J.J., Dapprich, S., Daniels, € Foresman, J.B., Ortiz, J.V., Cioslowski, J., Fox, D.J., 2009. Gaussian A.D., Farkas, O., Inc., Wallingford, CT, USA. Gough, K.M., Gesser, H.D., 1975. The extraction and recovery of phthalate esters from water using porous polyurethane foam. J. Chromatogr. 115, 383e390. Hamadi, N.K., Swaminathan, S., Chen, X.D., 2004. Adsorption of paraquat dichloride from aqueous solution by activated carbon derived from used tires. J. Hazard. Mater. 112, 133e141. Hehre, W.J., Ditchfield, R., Pople, J.A., 1972. Self-consistent molecular orbital methods. 12. Further extensions of Gaussian-type basis sets for use in molecular-orbital studies of organic-molecules. J. Chem. Phys. 56, 2257e2261. Ho, Y.S., McKay, G., 1999. Pseudo-second order model for sorption processes. Process Biochem. 34, 451e465. Hsu, S.-T., Chen, L.-C., Lee, C.-C., Pan, T.-C., You, B.-X., Yan, Q.-F., 2009. Preparation of methacrylic acid-modified rice husk improved by Na experimental design and application for paraquat adsorption. J. Hazard. Mater. 171, 465e470. pez, R., Gondar, D., Antelo, J., Fiol, S., Arce, F., 2010. Adsorption of Iglesias, A., Lo paraquat on goethite and humic acid-coated goethite. J. Hazard. Mater. 183, 664e668. Jarvie, D.R., Stewart, M.J., 1979. The rapid extraction of paraquat from plasma using an ion-pairing technique. Clin. Chim. Acta 94, 241e251. Kohn, W., Sham, L.J., 1965. Self-consistent equations including exchange and correlation effects. Phys. Rev. 140, 1133e1138. Lazaridis, N.K., Asouhidou, D.D., 2003. Kinetics of sorptive removal of chromium(VI) from aqueous solutions by calcined Mg-Al-CO3 hydrotalcite. Water Res. 37, 2875e2882. Lee, C., Yang, W., Parr, R.G., 1988. Development of the ColleeSalvetti correlationenergy formula into a functional of the electron density. Phys. Rev. B 37, 785e789. Leite, B.T., Robaina, N.F., dos Reis, L.G.T., Netto, A.D.P., Cassella, R.J., 2012. Removal of malachite Green from aqueous medium employing polyurethane foam as adsorbent and sodium dodecylsulfate as carrier. Water, Air, Soil Pollut. 223, 1303e1313. Leite, M.P., Reis, L.G.T., Robaina, N.F., Pacheco, W.F., Cassella, R.J., 2013. Adsorption of paraquat from aqueous medium by Amberlite XAD-2 and XAD-4 resins using dodecylsulfate as counter ion. Chem. Eng. J. 215e216, 691e698. Magdy, Y.H., Daifullah, A.A.M., 1998. Adsorption of a basic dye from aqueous solutions onto sugar-industry-mud in two modes of operations. Waste Manage 18, 219e226. Mandel, J.S., Adami, H.-O., Cole, P., 2012. Paraquat and Parkinson's disease: an overview of the epidemiology and a review of two recent studies. Regul. Toxicol. Pharmacol. 62, 385e392. dasy, M., Andriska, V., 1988. Studies in Environmental Science 32: Matolcsy, G., Na Pesticide Chemistry. Elsevier, New York, NY, USA. , G., 1996. ParaMelchiorri, D., Reiter, R.J., Sewerynek, E., Hara, M., Chen, L., Nistico quat toxicity and oxidative damage: reduction by melatonin. Biochem.. Pharmacol. 51, 1095e1099. Miehlich, B., Savin, A., Stoll, H., Preuss, H., 1989. Results obtained with the correlation-energy density functionals of Becke and Lee, Yang and Parr. Chem. Phys. Lett. 157, 200e206. Mori, M., Cassella, R.J., 2009. Sorption of crystal violet by polyurethane foam from aqueous medium containing sodium dodecylsulfate. Quím. Nova 32, 2039e2045. Nanseu-Njiki, C.P., Dedzo, G.K., Ngameni, E., 2010. Study of the removal of paraquat from aqueous solution by biosorption onto Ayous (Triplochiton schleroxylon) sawdust. J. Hazard. Mater. 179, 63e71. Parr, R.G., Yang, W., 1989. Density-functional Theory of Atoms and Molecules. Oxford University Press, New York, NY, USA. Proctor, A., Toro-Vazquez, J.F., 1996. The Freundlich isotherm in studying adsorption in oil processing. J. Am. Oil Chem. Soc. 73, 1627e1633. Qiu, H., Lv, L., Pan, B.-C., Zhang, Q.-J., Zhang, W.-M., Zhang, Q.-X., 2009. Critical review in adsorption kinetic models. J. Zhejiang Uni. Sci. A 10, 716e724. Reichenberg, D., 1953. Properties of ion-exchange resins in relation to their structure. III. Kinetics of exchange. J. Am. Chem. Soc. 75, 589e597. Robaina, N.F., Soriano, S., Cassella, R.J., 2009. Polyurethane foam loaded with SDS for the adsorption of cationic dyes from aqueous medium: multivariate optimization of the loading process. J. Hazard. Mater. 167, 653e659. Sant’Ana, O.D., Jesuino, L.S., Cassella, R.J., Carvalho, M.S., Santelli, R.E., 2003. Solidphase extraction of Cu (II) as diethyldithiocarbamate complex by polyurethane foam. J. Braz. Chem. Soc. 14, 728e733. Sant’Ana, O.D., Jesuino, L.S., Cassella, R.J., Carvalho, M.S., Santelli, R.E., 2004a. Determination of lead by electrothermal atomic absorption spectrometry employing a novel sampling strategy of polyurethane foam impregnated with thiazolylazo-p-cresol (TAC). J. Braz. Chem. Soc. 15, 96e102. Sant’Ana, O.D., Jesuino, L.S., Cassella, R.J., Carvalho, M.S., Santelli, R.E., 2004b. Solidphase extraction of Cd (II) as diethyldithiocarbamate complex by polyurethane foam: kinetic and thermodynamic characterization. Sep. Sci. Technol. 39, 2695e2709.
208
J.O. Vinhal et al. / Journal of Environmental Management 156 (2015) 200e208
Seki, Y., Yurdakoç, K., 2005. Paraquat adsorption onto clays and organoclays from aqueous solution. J. Colloid Interface Sci. 287, 1e5. Shao, Y., Molnar, L.F., Jung, Y., Kussmann, J., Ochsenfeld, C., Brown, S.T., Gilbert, A.T.B., Slipchenko, L.V., Levchenko, S.V., O'Neill, D.P., DiStasio Jr., R.A., Lochan, R.C., Wang, R., Beran, G.J.O., Besley, N.A., Herbert, J.M., Lin, C.Y., Van Voorhis, T., Chien, S.H., Sodt, A., Steele, R.P., Rassolov, V.A., Maslen, P.E., Korambath, P.P., Adamson, R.D., Austin, B., Baker, J., Byrd, E.F.C., Dachsel, H., Doerksen, R.J., Dreuw, A., Dunietz, B.D., Dutoi, A.D., Furlani, T.R., Gwaltney, S.R., Heyden, A., Hirata, S., Hsu, C.-P., Kedziora, G., Khalliulin, R.Z., Klunzinger, P., Lee, A.M., Lee, M.S., Liang, W.Z., Lotan, I., Nair, N., Peters, B., Proynov, E.I., Pieniazek, P.A., Rhee, Y.M., Ritchie, J., Rosta, E., Sherrill, C.D., Simmonett, A.C., Subotnik, J.E., Woodcock III, H.L., Zhang, W., Bell, A.T., Chakraborty, A.K., Chipman, D.M., Keil, F.J., Warshel, A., Hehre, W.J., Schaefer, H.F., Kong, J., Krylov, A.I., Gill, P.M.W., Head-Gordon, M., 2006. Advances in methods and algorithms in a modern quantum chemistry program package. Phys. Chem. Chem. Phys. 8, 3172e3191. Silva, M.S., Cocenza, D.S., Grillo, R., Melo, N.F.S., Tonello, P.S., Oliveira, L.C.,
Cassimiro, D.L., Rosa, A.H., Fraceto, L.F., 2011. Paraquat-loaded alginate/chitosan nanoparticles: preparation, characterization and soil sorption studies. J. Hazard. Mater. 190, 366e374. Silveira Neta, J.J., Moreira, G.C., Silva, C.J., Reis, C., Reis, E.L., 2011. Use of polyurethane foams for the removal of the Direct Red 80 and Reactive Blue 21 dyes in aqueous medium. Desalination 281, 55e60. Tomasi, J., Mennucci, B., Cammi, R., 2005. Quantum mechanical continuum solvation models. Chem. Rev. 105, 2999e3093. Tsai, W.T., Hsien, K.J., Chang, Y.M., Lo, C.C., 2005. Removal of herbicide paraquat from an aqueous solution by adsorption onto spent and treated diatomaceous earth. Bioresour. Technol. 96, 657e663. Tsai, W.-T., Lai, C.-W., 2006. Adsorption of herbicide paraquat by clay mineral regenerated from spent bleaching earth. J. Hazard. Mater. 134, 144e148. Tseng, R.-L., Wu, F.-C., Juang, R.-S., 2010. Characteristics and applications of the Lagergren's first-order equation adsorption kinetics. J. Taiwan Inst. Chem. Eng 41, 661e669.
